Field / plotLeafClassificationStress / disease detectionDisease symptoms / severity
Plant diseases threaten agricultural productivity, and automated image analysis can support early identification of visible disease symptoms. This study introduces HybOptic-CNN, a convolutional neural network (CNN) whose learning rate and batch size are selected using a hybrid Whale Optimization Algorithm--Grey Wolf Optimizer (WOA-GWO). Nine disease classes were selected from the 22-class CCMT field-image dataset, and 152 local farm leaf images were collected in Enugu State, Nigeria. Of the local images, 122 (80.3%) were added to the model-development data for training and validation, whereas 30 (19.7%) formed an independent Nigerian hold-out set excluded from augmentation, class balancing, early stopping, validation, and hyperparameter selection. Across 10 model-development runs, the optimized model achieved 96.8 ± 0.4% mean validation accuracy, 95.2 ± 0.5% macro-precision, 94.9 ± 0.6% macro-recall, and 95.0 ± 0.5% macro-F1, compared with 90.3 ± 0.9% validation accuracy and 86.3 ± 1.2% macro-F1 for the baseline. The optimized model improved mean validation accuracy by 6.5 percentage points and converged 14.6 epochs earlier. On the independent 30-image Nigerian hold-out, HybOptic-CNN achieved 93.3% accuracy and 93.1% macro-F1 across four represented disease classes. A web application integrating the trained classifier was also demonstrated. These results support improved model-development performance through hybrid hyperparameter selection and motivate broader multi-location field evaluation.
Why it matches plant phenotyping methods植物の可視病徴を画像から分類するCNN手法を開発・検証しており、病害状態のフェノタイピング手法が中心である。
abstractautomated image analysis can support early identification of visible disease symptoms
Reproduction assets foundThe paper's Data availability statement points to two public sources: a Mendeley dataset (the locally collected Nigerian field images) and the Kaggle CCMT plant disease dataset used as the principal image source. Only the Kaggle URL matches an allowed URL; the Mendeley URL is not in the allowed list, so only the CCMT/KDataset · publicnt and independent field-test data and should pub-
lish the class-wise split manifest, random seeds, WOA-GWO
numerical settings, and evaluation code so that the reported pro-
cedure can be reproduced and extended.
Data availability
The data used in this study are available at https://
data.mendeley.com/datasets/bwh3zbpkpv/1 and https://www.kaggle.com/datasets/rahimanshu/ccmt-plant-disease-dataset.Declaration of competing interest
The authors declare that they have no known competing fi-
nancial interests or personal relationships that could have ap-
peared to influence the work reported in this manuscript.
Funding
The authors received no specific funding from any public,
commercial, or not-fOpen asset ↗Kaggle · rahimanshu/ccmt-plant-disease-datasetpdf-raw-page:12 lines:1-78Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Abstract Tomato leaf diseases (TLDs) such as Early Blight (EB), Late Blight (LB), and Leaf Mold (LM) have a negative impact on crop yield and quality, result-ing in economic losses in agriculture. Traditional diagnosis methods depend on the visual recognition of a disease, which are time-consuming, subjective and may be difficult to reach in rural settings. Keeping these drawbacks in mind, the present work introduces a Transfer Learning model for tomato leaf disease classification based on EfficientNetB3 network. A pretrained Ef-ficientNetB3 network, trained on ImageNet, is used to obtain discriminative features like lesion boundaries, discoloration, fungal textures, and infection spots. Images are cropped to 224 × 224 pixels and split into training, validation and test set. The proposed architecture employs Global Average Pooling, a dense layer with ReLU activation, drop out regularization and Softmax classifier for classification of tomato leaf images into four classes: Early Blight, Late Blight, Leaf Mold and Healthy. Experimental results show the high classification accuracy and low training, validation and test-ing losses. Inter-class misclassifications of the confusion matrix show very few, which supports good generalization. Moreover, Grad-CAM visualiza-tions can generate interpretable heat maps that point to the regions of the image affected by the disease, and multi-class ROC analysis gives high AUC values, which means that it has excellent class separability. The proposed framework provides a precise, reliable, and interpretable approach for auto-mated tomato leaf disease diagnosis, contributing to precision agriculture by assisting in early detection and prompt management of tomato diseases.
Why it matches plant phenotyping methodsトマト葉画像から病害状態を推定する深層学習手法が研究の中心であり、Grad-CAMによる病変領域の解釈と性能評価も行っているため、植物表現型計測手法として採択。
abstractthe present work introduces a Transfer Learning model for tomato leaf disease classification based on EfficientNetB3 network
Reproduction assets foundThe paper's tomato leaf disease classification uses a publicly available Kaggle dataset (PlantVillage-derived tomato leaf images, 4000 images across 4 classes), explicitly declared in the Data Availability statement with a public URL. No author code, trained models, or other paper-specific assets are disclosed.Dataset · publicitted in accordance with the Journal policies.
Permission to use third-party material
Images or figures are never published previously and did not take from any
internet resources.
Data Availability
The dataset used in this study is publicly available from the PlantVillage
tomato leaf disease dataset on Kaggle’s following link:
https://www.kaggle.com/datasets/kaustubhb999/tomatoleaf
Acknowledgements
The authors would like to thank SR University, Warangal, Telangana,
INDIA and Sharda University, Greater Noida, Uttar Pradesh, INDIA for
providing research facilities and computational resources to carry out this
work.Open asset ↗Kaggle · kaustubhb999/tomatoleafpdf-raw-page:20 lines:1-18Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Abstract Maize plant plays crucial role not only in the field of agriculture but also in global economy, since it is the third most cultivated crop across the globe. However, these plants are usually affected by various types of diseases such as blight, common rust, gray leaf spot, etc., protecting the plants from these disease is very important. This research proposes a new deep learning model for disease detection to perform well than other models. The hybrid model uses MobileNetV3 as the backbone architecture integrated with attention, fusion and head. The framework includes data preprocessing, feature extraction, classification and interpretation. We will compare the performance of this model with other models such as VGG16, ResNet50, DenseNet121, ALEXNET, etc. criteria for the final evaluation includes Accuracy, Precision, Recall, F1score, Specificity, Logloss, AUC-ROC curve. Through this proposed model we have achieved an accuracy of 98% which is high than the other models compared. The lightweight nature of MobileNetV3 enables us to implement the model in the mobile and IoT devices also. The present study contributes to the development of deep learning model in the field of agriculture, offering a efficient solution for early maize leaf disease detection.
Why it matches plant phenotyping methodsトウモロコシ葉の病徴を画像から検出・分類する深層学習手法の開発と比較が中心であり、植物病害状態の画像ベース表現型計測に該当する。
abstractThis research proposes a new deep learning model for disease detection to perform well than other models.
Reproduction assets foundThe paper's maize leaf disease detection model (MAFH) was trained and evaluated entirely on a public Kaggle image dataset, which the authors explicitly declare in the Data Availability statement. No author code, trained model checkpoints, or other paper-specific assets are stated as publicly available.Dataset · publicig and real-
time datasets and in all the environmental situations.
Funding: This research received no external funding.
Disclosure statement: The authors declare no conflict of interest.
Data Availability
The datasets generated and/or analyzed during the current study are available in the CORN OR
MAIZE LEAF DATASET repository,
https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-datasetOpen asset ↗Kaggle · corn-or-maize-leaf-disease-datasetpdf-raw-page:27 lines:1-34Code / dataset availability confirmedEurope PMC · Crossref · checked 15 Sept 2026
Published4 Sept 2026Springer Science and Business Media LLC
Abstract Plant diseases substantially reduce global crop yields, and cotton production is particularly vulnerable to field-acquired variability in symptom appearance, background clutter, and illumination changes that limit the reliability and scalability of expert visual inspection. This study aimed to develop an accurate, computationally efficient, and explainable framework for real-time cotton leaf disease recognition that is suitable for deployment on resource-constrained edge devices. Using the SAR-CLD-2024 dataset (322 RGB images captured under natural agricultural conditions across seven categories, including healthy and diseased leaves), images were preprocessed via resizing and normalization and augmented online in the training set (random rotations, flips, brightness/contrast adjustments, and random cropping). An EfficientNet-B3 backbone initialized with ImageNet-pretrained weights was fine-tuned using categorical cross-entropy loss and Adam optimization, with early stopping, checkpointing, regularization, and a fixed-seed 70/15/15 train–validation–test partition to enhance reproducibility and reduce leakage. Performance was evaluated on an independent test set using accuracy, precision, recall, F1-score, MCC, balanced accuracy, Cohen’s kappa, confusion matrix, multi-class ROC/AUC, and precision–recall analysis, alongside computational benchmarking (parameters, FLOPs, memory, and inference latency) and comparative experiments against contemporary CNN, lightweight, and transformer-based models. The model showed stable convergence over 30 epochs with a small training–validation gap, predominantly correct predictions with limited confusion among visually similar classes, consistently high precision–recall behavior under moderate class imbalance, and stable performance across repeated runs with low variability and a tight confidence interval. Grad-CAM heatmaps localized necrotic lesions, discoloration, and infected tissues while largely ignoring background, and failure cases were associated with early-stage symptoms, occlusion, shadows, and inter-class similarity. Overall, the framework provides a reproducible, interpretable, and efficient solution for cotton leaf disease classification with practical implications for trustworthy, low-latency, on-device decision support in precision agriculture.
Why it matches plant phenotyping methods綿葉の病徴を画像から分類する深層学習手法の開発が中心で、独立テスト、比較評価、計算性能評価、Grad-CAMによる病徴局在化を実施しているため、植物病害フェノタイピング手法に該当する。
abstractThis study aimed to develop an accurate, computationally efficient, and explainable framework for real-time cotton leaf disease recognition
Reproduction assets foundThe paper's Data Availability statement explicitly names the SAR-CLD-2024 cotton leaf dataset used for all experiments as publicly available on Kaggle with a direct URL. No author analysis code or trained model checkpoint is deposited.Dataset · publicntribute to the development of fully automated, scalable, and real-time smart agriculture
systems.
Declaration
Funding
Datta Meghe Institute of Higher Education and Research Wardha, Maharashtra, India
Data Availability: The SAR-CLD-2024 cotton leaf dataset used in this study is publicly
available through the Kaggle platform at: https://www.kaggle.com/datasets/pantho12/sar-cld-2024-dataset-for-cotton
This dataset includes annotated images of various cotton leaf diseases collected under diverse
environmental conditions. All data utilized in this work are freely accessible, and the data
processing methodology has been described in detail to facilitate reproducibility.
Conflict of interest The aOpen asset ↗Kaggle · SAR-CLD-2024pdf-raw-page:32 lines:1-38Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Abstract Maize is the staple crop for millions of people in Sub-Saharan Africa, particularly for Zambia. Unfortunately, maize crops are exposed to several serious threats due to their susceptibility to foliar diseases like Maize Rust, Leaf Blight, Leaf Spot, Maize Streak Virus, and Maize Lethal Necrosis that may lead to great yield losses. Conventional methods of crop disease identification consist of field surveys that are not only subjective but also difficult to conduct for smallholder farmers. This paper presents the design and evaluation of a highly optimized version of the EfficientNet-B0 Convolutional Neural Network for the automatic detection of maize leaf diseases using maize leaf images obtained from real-world scenarios. The proposed model utilized the concept of transfer learning with ImageNet pre-trained weights and was trained on the Mendeley Maize Crop Disease (Leaf) Dataset which consists of 30,120 images in nine maize disease classes. The developed fine-tuned EfficientNet-B0 yielded 97.57% classification accuracy, macro precision of 97.61%, macro recall of 97.64%, and macro F1-score of 97.61%. From these results, it is evident that transfer learning and fine-tuning greatly boost maize disease classification accuracy while ensuring high computational efficiency. This study makes a significant contribution to precision agriculture as it offers an accurate and computationally efficient AI-based maize disease classification model, which could help smallholder farmers in early maize disease classification.
Why it matches plant phenotyping methodsトウモロコシ葉画像から病害状態を推定する深層学習モデルを開発・評価しており、植物病害表現型の取得・分類手法が中心的な研究です。
abstractThis paper presents the design and evaluation of a highly optimized version of the EfficientNet-B0 Convolutional Neural Network for the automatic detection of maize leaf diseases using maize leaf images obtained from real-world scenarios.
Reproduction assets foundThe paper's sole qualifying asset is the public Mendeley Maize Crop Disease (Leaf) Dataset of maize leaf images used for all phenotyping/classification measurements, explicitly declared publicly available with a URL. No author analysis code, trained model checkpoints, or other paper-specific assets are disclosed.Dataset · publiconflicts of interest to publish the paper.
Consent to Publish
All authors have read and approved the final version of the manuscript and agree to its
submission to Discover Networks.
Consent to Participate
Not applicable
Data Availability
The Mendeley Maize Crop Disease (Leaf) Dataset used in this study is publicly available at
https://data.mendeley.com/datasets/6w6gsvghfw
Clinical Trial Number
Not applicable.
Ethics Declaration:
Not applicable.
Competing interests
All authors declare no competing interests.Open asset ↗Mendeley · 6w6gsvghfwpdf-raw-page:55 lines:1-22Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Fruit quality is a critical determinant of economic returns in pear production, and maintaining an appropriate fruit load (FL) is essential for achieving high yield and quality. As a direct indicator of canopy photosynthetic capacity and assimilate supply, leaf number constitutes the key biological basis of reasonable FL determination under the leaf-to-fruit ratio concept. However, accurate and efficient estimation of leaf number in mature pear trees remains technically challenging, limiting its practical use in precision FL regulation. Here, we propose a data-driven framework for leaf number and reasonable FL estimation by integrating 3D point cloud-derived canopy structure with machine learning. A pipeline for extracting 3D architectural traits was developed and implemented in the software tool FTPCT, enabling rapid and standardized trait acquisition. Through correlation analysis, multicollinearity diagnosis, and variance inflation factor screening, five key traits strongly associated with leaf number were identified and incorporated into five machine learning models optimized using Bayesian optimization. Among them, the optimized random forest regression model achieved the highest and most stable performance, with R 2 of 0.85, RMSE of 239.74, and MAE of 149.26 for test dataset. SHAP analysis identified tree crown volume as the dominant contributor to leaf number estimation. Field validation demonstrated that FL regulation guided by the proposed framework significantly improved fruit weight and size without reducing yield compared with conventional practices. Notably, the proposed approach avoids explicit leaf-level reconstruction and relies on less canopy-scale traits, substantially reducing data requirements and computational cost, and thereby offering strong potential for rapid, field-deployable FL regulation in large-scale orchards.
Why it matches plant phenotyping methods3D点群から樹冠構造形質を抽出し、葉数と適正着果量を推定する手法およびソフトウェアを開発・検証しており、植物表現型取得が中心である。
abstractA pipeline for extracting 3D architectural traits was developed and implemented in the software tool FTPCT, enabling rapid and standardized trait acquisition.
Reproduction assets foundThe paper's phenotyping analysis assets are the authors' publicly released LeafNumPred source code and trained models, and the FTPCT software for 3D trait extraction from pear tree point clouds. Phenotype/point-cloud datasets are only available on request.Code · public. Supplementary data
The following is the Supplementary data to this article:
Multimedia component 1
mmc1.docx (1.6MB, docx)
Data availability
Data will be made available on request. Anyone who wants to obtain other public data can contact us at taost@njau.edu.cn. The source codes and models have been made publicly available at https://github.com/Zhang-Fanhang/LeafNumPred, and the FTPCT software has been released at https://github.com/Zhang-Fanhang/FTPCT/tree/Installation-package.
References
1.Tao S., Khanizadeh S., Zhang H., Zhang S. Anatomy, ultrastructure and lignin distribution of stone cells in two Pyrus species. Plant Sci. 2009;176:413–419. [Google Scholar]
2.Zhang F., Wang Q., Yuan K.Open asset ↗Zhang-Fanhang/LeafNumPredhtml-lines:284-315Code · publical variations [34,35]. The method for calculating these traits are shown in the Supplementary information 1.
2.5. Software implementation for 3D trait extraction (FTPCT)
To facilitate efficient and standardized extraction of canopy structural traits from point cloud data, we used a standalone software tool, FTPCT (available at: https://github.com/Zhang-Fanhang/FTPCT/tree/Installation-package), which integrates the trait extraction procedures applied in this study. The software provides a graphical user interface, enabling users to process tree-level point cloud data and extract key 3D structural traits without requiring advanced programming skills.
FTPCT implements a series of predefined proOpen asset ↗Zhang-Fanhang/FTPCThtml-lines:138-149Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Automated plant disease classification from leaf images demands models that jointly achieve high accuracy and efficient training convergence. Standard deep learning approaches process images through a single feature pathway, limiting their ability to capture diverse visual manifestations such as color changes, texture patterns, and spatial distributions. This paper introduces DMCNNA-FBVL, a framework integrating two complementary innovations: (1) a Deep Multi-Component Neural Network Architecture (DMCNNA) employing three specialized pathways processing color-space statistics, texture descriptors, and raw image features before fusing them through SoftMax-weighted aggregation; and (2) Feedback-Based Validation Learning (FBVL), a training strategy that periodically blends validation-set gradients into weight updates to accelerate convergence. Experiments on the New Plant Diseases Dataset (87,000 images, 38 classes) show that DMCNNA-FBVL achieves 98.7% accuracy, 98.8% precision, 98.6% recall, and 98.7% F1-score, outperforming ResNet-50 by 3.2 percentage points ( 𝑝 < 0 . 0 0 1 ). The primary reported metrics are computed exclusively on an independent 10% hold-out test set, whereas the separate 10% validation partition is used during training for FBVL gradient blending and does not contribute to final test evaluation. Five-fold cross-validation confirms stability (98.7% ± 0.15%). Ablation experiments confirm additive gains, while FBVL reduces wall-clock training time by 15% through faster convergence.
Why it matches plant phenotyping methods葉画像から植物病害を分類するニューラルネットワークと学習戦略の開発・検証が研究の中心であり、植物の病害状態を直接推定している。
abstractAutomated plant disease classification from leaf images demands models that jointly achieve high accuracy and efficient training convergence.
Reproduction assets foundThe paper's plant disease classification experiments use the New Plant Diseases Dataset, which the authors state is publicly available on Kaggle. The authors' code and trained models are only promised 'upon acceptance' (no public URL), so they do not qualify as public assets.Dataset · publicral
monitoring systems.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
Data availability
The New Plant Diseases Dataset used in this study is publicly available on Kaggle (https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset). Code and trained models will be made available upon
acceptance.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit
sectors.
Acknowledgments
[Removed for double-blind review.]
CRediT authorship contribution statement
[Removed fOpen asset ↗Kaggle · new-plant-diseases-datasetpdf-raw-page:19 lines:1-59Code / dataset availability confirmedOpenAlex · Crossref · checked 14 Sept 2026
The automation and standardization of hyperspectral imaging, particularly at high spatial resolution, are essential for advancing plant phenomics and supporting diverse plant science research. We developed HyperBird, a hyperspectral microscopic imaging robot, to automate the acquisition of up to 351 leaf-disc samples in a single tray within approximately 2.4 h and thereby support scalable plant experiments. System calibration established a spatial resolution of 24.3 μ m (full width at half maximum; FWHM), a spectral resolution of 1.78 nm (FWHM), and a depth of field of 4.53 mm, producing hyperspectral data cubes of 2195 × 2000 × 950 pixels across the 400–1000 nm spectral range. System-level and biological-sample consistency assessments confirmed stable spectral measurements during extended scanning sessions and across independently prepared trays. Thermal evaluation confirmed that the 150 W illumination design introduced minimal sample heating during scanning. HyperBird was first evaluated using grape downy and powdery mildews, where no consistent measurable effect of repeated imaging on pathosystem development was detected under the tested experimental conditions. Subsequently, HyperBird was used to characterize spatiotemporal spectral progression in grapevine leaves inoculated with Plasmopara viticola . An automated regional spectral tracing pipeline was developed to isolate spectra from retrospectively traced regions and compare them with whole-leaf averaged spectra across 0–9 days post-inoculation (DPI). Categorical mixed-effects modeling showed that these spatially resolved regional spectra exhibited significant disease-associated spectral change by 5 DPI, with a sharp transition at 6 DPI, whereas whole-leaf spectra showed delayed and weaker disease-aligned changes beginning at 7 DPI. These results demonstrate that HyperBird enables high-throughput, spatially resolved hyperspectral imaging for quantifying localized plant disease progression and provides a scalable platform for studying spectral biology across plant phenotyping applications.
Why it matches plant phenotyping methodsHyperBirdの高スループット・ハイパースペクトル画像取得ロボットを開発し、校正、再現性、加熱影響、病害進展の定量性能を検証しており、植物フェノタイピング手法が中心である。
abstractWe developed HyperBird, a hyperspectral microscopic imaging robot, to automate the acquisition of up to 351 leaf-disc samples in a single tray within approximately 2.4 h and thereby support scalable plant experiments.
Reproduction assets foundThe paper's Data Availability statement points to a public GitHub repository containing the authors' code and processed data supporting the phenotyping analyses; raw hyperspectral image data are request-only.Code · publicData Availability
857
The code and processed data supporting the findings of this study are available in the GitHub
858
repository at https://github.com/jy773Cornell/HyperBird-Robot. Raw hyperspectral image
859
data are available from the corresponding author upon reasonable request due to file size and storage
860
constraints.
861
Supplementary Materials
862
Supplementary materials accompany this article as a separate document (supplementary.pdf).
863
Supplementary Figure S1. Representative GSAM-based segOpen asset ↗HyperBird-Robotpdf-raw-page:39 lines:1-75Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Southern corn leaf blight (SCLB) is caused by the fungal pathogen Bipolaris maydis (syn. Cochliobolus heterostrophus Drechsler) and is a common disease of fall crops of sweet corn. Phenotyping for SCLB resistance is performed through visual scoring, which is subjective and may limit genetic gain for this quantitative trait. As an alternative, we integrated computer vision (CV)-based phenotyping, genome-wide association studies (GWASs), and predictive breeding approaches to dissect the genetic basis of SCLB resistance. We utilized a sweet corn diversity panel with 693 genotypes, for which whole-genome resequencing produced a high-density single-nucleotide polymorphism (SNP) dataset. Broad-sense heritability for visual scoring ranged from 0.44 to 0.73, while CV-based phenotyping produced estimates ranging from 0.56 to 0.73 in multi-environment resistance trials conducted across 5 years and three locations. We performed GWAS using 16,755,210 SNPs and identified 41 associated SNPs. Genomic selection (GS) models on visual scoring phenotypes achieved moderate prediction accuracies under cross-validation of untested genotypes across characterized environments (0.22-0.47) and high prediction accuracies when predicting tested genotypes in uncharacterized environments (0.49-0.68). Using CV-based phenotypes for GS, we observed prediction accuracies of 0.45-0.47 under the untested genotypes in the characterized environments cross-validation scheme and 0.59-0.62 under the tested genotypes in the uncharacterized environments scheme. GS demonstrated reliability for ranking the individuals across a gradient of environments. These findings identify candidate loci and predictive breeding strategies to accelerate the development of resistant sweet corn cultivars.
Why it matches plant phenotyping methodsCVベースの病害抵抗性表現型測定を視覚評定と比較し、多環境・多年次試験で妥当性を評価しており、フェノタイピング手法が研究の中心である。
abstractPhenotyping for SCLB resistance is performed through visual scoring, which is subjective and may limit genetic gain for this quantitative trait.
Reproduction assets foundThe authors state that all datasets (phenotype data) and analysis code (CV phenotyping script, customized GAPIT script) are publicly available in their GitHub repository, which is listed in allowed_urls.Code · publiche images taken for each plot were saved in JPG format and analyzed using a CV method. Here, we refer to the CV method as a custom Python script written using the OpenCV library version 4.5.0, a set of tools for CV (Bradski, 2000 ). The Python script used for leaf CV image analysis is available in our public GitHub repository ( https://github.com/Resende‐Lab/SCLB‐Disease ).
FIGURE 1
Leaf imaging set up with QR‐coded plot IDs (bottom right) and color checker for computer vision phenotyping of southern corn leaf blight disease severity in sweet corn.
In the CT19 environment, a black cloth attached to a wooden board was used as the background. A wooden frame was used to clamp the leaves down toOpen asset ↗Resende‐Lab/SCLB‐Diseaselines:199-209Dataset · publicBLUP and BayesB model implemented in BGLR.
ACKNOWLEDGMENTS
This work was supported by the National Institute of Food and Agriculture USDA‐NIFA2018‐51181‐28419, USDA‐NIFA2019–05410, and USDA‐NIFA 2022–51181‐38333.
DATA AVAILABILITY STATEMENT
All the datasets and codes used in this study are available in the following repository: https://github.com/Resende‐Lab/SCLB‐Disease
REFERENCES
Amadeu , R. R.
,
Cellon , C.
,
Olmstead , J. W.
,
Garcia , A. A. F.
,
Resende , M. F. R.
, &
Muñoz , P. R.
( 2016 ).Open asset ↗Resende‐Lab/SCLB‐Diseaselines:566-596Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Surface phenotyping underpins plant science, preclinical animal research and entomology, yet across all three the measurement is almost always a photograph, which records a projection and not the surface itself. Here we present the Gentschinator3000 , an open structured-light platform that brings high-end metric surface measurement within reach of laboratories with no optics expertise, combining documented open hardware, open reconstruction software and analysis workflows for under 4000 Euro in components. It resolves a planar reference to 45 µm local flatness, registers full rotations to a loop closure of 156 µm, and performs stably across acquisition ranges that we define. Applying one workflow to a leaf before and after desiccation, to murine anatomy and to a spread lepidopteran, we find that projection underestimates surface area by 11 to 41 %. That error grows with the condition under study, with the evaluation scale and with the direction of view, so it can confound phenotype comparisons dramatically. In murine limbs a 15-degree change of viewing direction shifts a projected inter-segment angle by up to 23.2 degrees, while the three-dimensional angle does not move. Projection geometry can therefore contribute as much to a measured phenotype as the biology it is meant to quantify.
Why it matches plant phenotyping methods植物表面の三次元形状を測定するオープンな構造化光プラットフォームと再構成・解析ワークフローを開発し、葉で投影バイアスを評価しており、表現型取得手法が中心である。
abstractHere we present the Gentschinator3000 , an open structured-light platform that brings high-end metric surface measurement within reach of laboratories with no optics expertise, combining documented open hardware, open reconstruction software and analysis workflows for under 4000 Euro in components.
Reproduction assets foundThe paper explicitly deposits three public Zenodo records: reconstructed 3D surfaces of all specimens (including the leaf and hop cone phenotyping measurements), the authors' analysis notebooks with derived and per-panel source data, and the reconstruction software with build documentation and working examples. All areDataset · publicData availability
1149
The reconstructed surfaces supporting this study are available at Zenodo under
1150
https://doi.org/10.5281/zenodo.22167250.54
1151
Source data for all graph panels are provided with this paper; for panels showing
1152
rendered surfaces, the underlying reconstructions are in the same record.
1153
1154
Code availability
1155
The analysis notebooks, environment specifications, derived data and per-panel
1156
source data are available at Zenodo under
1Open asset ↗Zenodo · 10.5281/zenodo.22167250pdf-raw-page:35 lines:1-52Code · public1151
Source data for all graph panels are provided with this paper; for panels showing
1152
rendered surfaces, the underlying reconstructions are in the same record.
1153
1154
Code availability
1155
The analysis notebooks, environment specifications, derived data and per-panel
1156
source data are available at Zenodo under
1157
https://doi.org/10.5281/zenodo.22167598.55
1158
The reconstruction software, build documentation and minimal working examples
1159
are available at https://doi.org/10.5281/zenodo.22167471.56
1160
The software and analysis notebooks are released under the MIT licence and the
1161
hardware design files under CERN-OHL-P v2. The visible-light platform described
1162
hereOpen asset ↗Zenodo · 10.5281/zenodo.22167598pdf-raw-page:35 lines:1-52Code / dataset availability confirmedCrossref · checked 11 Sept 2026
Rice leaf diseases pose a significant threat to global food security by reducing crop productivity and causing substantial economic losses. The traditional diagnosis method is manual method, which is low in efficiency, subjective and not suitable for large-scale agricultural monitoring. Despite the advances in automated disease detection using deep learning methods like CNNs, GANs, and transfer learning models, these techniques remain highly computational, not very flexible, and struggle to perform well in different imaging scenarios. Considering these drawbacks, this paper introduces an Explainable Deep Q-Learning based CNN framework which employs CNN-based feature extraction and Deep Q-learning-based adaptive policy optimization for rice leaf disease classification. The proposed model continuously refines the classification actions through reward-based learning, which makes the model more robust in various agricultural imaging environments, in contrast to traditional supervised CNN models that have static classification decisions. The proposed model achieved 98.5% accuracy, 98.52% precision, 98.50% recall, and a 98.51% F1-score, outperforming existing CNN, GAN, reinforcement learning, and transformer-based methods. It also offers a high computational efficiency of 14.2 GFLOPs, 248 MB memory consumption, ~ 48 min of training time, and 6.8 ms inference time per image suitable for resource constrained applications in agriculture. The results demonstrate the effectiveness, scalability, and practical applicability of the proposed framework. The proposed framework performs well on benchmark datasets but more research in the deployment of the edge-devices under different real-world agricultural settings will be investigated in future work.
Why it matches plant phenotyping methodsイネ葉の病害状態を画像から分類する深層学習手法の開発・評価が研究の中心であり、植物の病害表現型を直接推定しているため。
abstractthis paper introduces an Explainable Deep Q-Learning based CNN framework which employs CNN-based feature extraction and Deep Q-learning-based adaptive policy optimization for rice leaf disease classification.
Reproduction assets foundThe paper trains its Deep Q-CNN rice leaf disease classifier on public Kaggle rice leaf image datasets, which are cited with explicit public URLs and qualify as paper-specific phenotyping image inputs. The authors' own derived data/analysis artifacts are only available upon request, so no authors' code or trained modelDataset · publicSoni Gautam. Rice Leaf Bacterial and Fungal Disease Dataset. Kaggle. Available:Open asset ↗Kagglepdf-page:24 lines:1-94Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Plant diseases are still posing a challenge to the productivity, quality of crops, and food security, especially in locations where field diagnosis is based on manual visual inspec-tion. This paper assesses deep learning network-based automated classification of plant leaf diseases on public RGB leaf-image datasets, such as the Kaggle New Plant Diseases Dataset (Augmented) and PlantVillage images. They investigated four archi-tectures: EfficientNetV2B0, ResNet152V2, DenseNet201, and one hybrid Vision Trans-former (ViT)-based model. The steps of the experiment involved loading the dataset, exploratory analysis, preprocessing, resizing, normalizing, augmentation, transfer learning, independent model training, and evaluation metrics such as accuracy, preci-sion, recall, F1-score, training curves, testing results, and confusion matrices. The hy-brid ViT-based model was reported to have the best accuracy of 99.5%. On the smaller seven class subset of PlantVillage, EfficientNetV2B0 scored 98.11%. On the 38-class dataset, DenseNet201 improved test accuracy (97.34) and validation classification ac-curacy (around 98). ResNet152V2 scored 97.01 on the 38-class test set. The results demonstrate that CNN and transformer-based models can help to recognize plant diseases accurately whereas hybrid attention-based structures provide a promising path to enhance fine-grained classification. Since the model notebooks had varying class settings and splits, the comparison is seen as a model-structured assessment as opposed to a precisely identical benchmark across all architectures.
Why it matches plant phenotyping methods植物葉画像から病害状態を分類する深層学習手法を複数モデルで比較・評価しており、病害表現型の取得・抽出と技術検証が研究の中心である。
abstractThis paper assesses deep learning network-based automated classification of plant leaf diseases on public RGB leaf-image datasets
Reproduction assets foundThe paper's phenotyping inputs are two public plant leaf-image datasets explicitly named in the Data Availability statement: the Kaggle New Plant Diseases Dataset (Augmented) and the PlantVillage dataset, both with public URLs. No author code, models, or supplementary materials are deposited (supplementary materials: 'Dataset · publicy available. Plant leaf images were
obtained from the Kaggle New Plant Diseases Dataset (Augmented) and PlantVillage
datasets. The datasets contain publicly accessible RGB images of healthy and diseased
plant leaves used for supervised image classification research.
DATASET SOURCES
Kaggle New Plant Diseases Dataset (Augmented):
https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset
PlantVillage Dataset: https://plantvillage.psu.edu/All processed data, experimental configurations, and model implementation details are
described within the manuscript. Additional materials may be made available from the
corresponding author upon reasonable request.
ACKNOWLEDGMENTS
The author acknowOpen asset ↗Kaggle · new-plant-diseases-datasetpdf-raw-page:24 lines:1-23Dataset · publicgmented) and PlantVillage
datasets. The datasets contain publicly accessible RGB images of healthy and diseased
plant leaves used for supervised image classification research.
DATASET SOURCES
Kaggle New Plant Diseases Dataset (Augmented):
https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset
PlantVillage Dataset: https://plantvillage.psu.edu/All processed data, experimental configurations, and model implementation details are
described within the manuscript. Additional materials may be made available from the
corresponding author upon reasonable request.
ACKNOWLEDGMENTS
The author acknowledges Istanbul Aydin University for academic support and research
guidance duringOpen asset ↗PlantVillagepdf-raw-page:24 lines:1-23Code / dataset availability confirmedOpenAlex · Europe PMC · bioRxiv · checked 14 Sept 2026
Abstract Background and aims Plant volatile organic compounds (VOCs) change dynamically with plant development and in response to environmental conditions. However, their potential as non-invasive indicators of phenological progression remains poorly explored. In this study, we developed a framework integrating automated VOC sampling, time-resolved VOC profiling, and machine-learning analysis for the non-invasive assessment of plant phenology. Using soybean ( Glycine max (L.) Merr.), we investigated whether development-associated temporal variation in VOC emissions could delineate and predict developmental phases. Methods We collected VOCs daily under controlled environmental conditions from 16 to 43 days after sowing, spanning the transition from vegetative to reproductive stages, using an automated sampling system coupled with thermal desorption-gas chromatograph-mass spectrometer (TD- GC-MS). To characterise temporal changes in VOC profiles associated with phenological progression, we analysed the daily VOC data using a multi-step pipeline combining statistical filtering and similarity-based network analysis. We defined VOC-derived developmental phases from similarity patterns in the VOC profiles, then developed and evaluated machine-learning models to predict these phases. Key results Seven VOCs exhibited distinct phase-dependent dynamics, including green leaf volatiles and monoterpenes showing characteristic temporal changes during phenological progression. Network-based clustering of VOC profiles resolved five developmental phases closely aligned with conventional developmental stages. A machine-learning model predicted these phases from the VOC profiles with high predictive accuracy on independent test data, demonstrating that phenological progression could be quantitatively inferred from VOC emission patterns. Conclusions Our findings support VOC profiling as a reliable and non-invasive approach for assessing phenological progression in soybean. By extracting temporally structured VOC signals, this framework captures developmental information that may be difficult to obtain through visual observation alone, particularly after canopy closure. VOC profiling offers a practical tool for monitoring crop developmental dynamics and has broader potential for plant phenotyping and precision crop management.
Why it matches plant phenotyping methods自動VOCサンプリング、時系列VOCプロファイリング、機械学習を統合し、VOCから植物の発育段階を非破壊推定する方法を開発・評価しており、フェノタイピング手法が中心である。
abstractwe developed a framework integrating automated VOC sampling, time-resolved VOC profiling, and machine-learning analysis for the non-invasive assessment of plant phenology.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Supplement · publicThe peak area matrix obtained from the MS- DIAL analysis (Supplementary Dataset S1) was filtered to remove unreliable features.Open asset ↗lines:66-69Code / dataset availability confirmedOpenAlex · arXiv · checked 5 Sept 2026
While most phenotyping platforms rely primarily on image-based measurements, advanced plant characterization requires the integration of active physiological sensing modali- ties such as chlorophyll fluorescence. We present an autonomous robotic platform designed to perform targeted fluorescence measurements on plant leaves. The system combines 3D plant reconstruction, geometric analysis, and motion planning to localize suitable measurement points and generate collision-free trajectories for a robotic manipulator. A dense 3D model of the plant is reconstructed from multi-view data and used to extract candidate leaf surfaces based on orientation, accessibility, and sensing constraints. These targets are then integrated into a task-level planning framework that guides the end-effector to precise contact or near-contact configurations required for point-based fluorescence acquisition. The platform enables automated, repeatable, and spatially resolved physiological measurements that go beyond passive imaging. By tightly coupling perception, geometric reasoning, and manipulation, the proposed system provides a robotics-driven approach to high-resolution plant phenotyping and opens new directions for autonomous agricultural inspection and plant-aware manipulation.
Why it matches plant phenotyping methods植物葉の蛍光を自律ロボットで空間的・反復的に取得するプラットフォームを開発しており、植物表現型の取得手法が研究の中心です。
abstractWe present an autonomous robotic platform designed to perform targeted fluorescence measurements on plant leaves.
Reproduction assets foundThe paper states its code is publicly available in the authors' SonyCSLParis GitHub repository (Plant3DImager), which implements the phenotyping perception and motion-planning pipeline. The exact full URL is split across a line break in the supplied text, so the verifiable allowed URL prefix is used.Code · public2 The code is available at https://github.com/SonyCSLParis/ 3 See for example at https://www.youtube.com/watch?v=Open asset ↗SonyCSLParis/pdf-page:4 lines:1-61Code / dataset availability confirmedEurope PMC · bioRxiv · checked 5 Sept 2026
ABSTRACT Leaf shape is a fundamental trait of plant ecological strategies, influencing biotic interactions and ecosystem functioning. However, established quantitative metrics fail to capture subtle variations and irregularities, require user-based reference points or are challenging to compare among taxa with broadly different leaf shapes. In addition, established metrics typically conflate (aggregate) leaf edge complexity and macro-shape complexity, despite their independent functional significance and genetic foundations. Here, we introduce an entropy-based framework to quantify two new complexity metrics: edge complexity and macro-shape complexity. Based on three case studies, we show that these metrics outperform aggregate metrics in predicting Quercus robur chemical traits, provide more intuitive interspecific classifications, and strongly align with human perception. In addition, edge and macro-shape complexity show high complementarity, while aggregate metrics are highly redundant and typically strongly related to leaf area. Emerging as the strongest predictor of leaf chemistry and key visual cue for complexity as perceived by humans, the effects of edge complexity highlight the under-appreciated functional significance of leaf margins. Our framework and the proposed entropy-based complexity metrics thus promise to help unlock the potential of growing digital image archives of leaves, including images from herbaria and fossils, and are technically readily applicable to shapes of algae, bacteria, pollen, and beyond. The accompanying package ShapeComplexity enables the broad application of entropy-based metrics, providing a powerful tool to explore how the shape of organisms and biological structures influences ecological strategies, biotic interactions, and ecosystem functioning while tracking spatial and temporal variation.
Why it matches plant phenotyping methods葉の画像からエッジ複雑性とマクロ形状複雑性を定量化する新規指標とソフトウェアを開発しており、植物形質抽出法が研究の中心である。
abstractHere, we introduce an entropy-based framework to quantify two new complexity metrics: edge complexity and macro-shape complexity.
Reproduction assets foundThe paper's authors publicly release their ShapeComplexity analysis code (Rust) on GitHub, used to compute the paper's leaf edge- and macro-shape complexity metrics. Supplementary data/analysis code are on Dryad, but that URL is not in the allowed list. RMBG is a generic third-party background-removal model, not a phenCode · publicThe complete, open-source Rust-code (The Rust Team, 2025 ) is publicly available on GitHub ( https://github.com/Thornbach/ShapeComplexity ), ensuring transparency and reproducibilityOpen asset ↗Thornbach/ShapeComplexitylines:86-94Code / dataset availability confirmedEurope PMC · Crossref · checked 14 Sept 2026
Abstract Deep convolutional networks now classify leaf images on curated benchmarks such as PlantVillage with accuracies close to the measurement ceiling of those datasets, which has shifted the open questions away from raw accuracy toward two under-reported issues: which components of a composite pipeline actually cause the result, and whether the components that matter on laboratory images are the same ones that matter on field photographs. We address both with a classification framework evaluated under a protocol that fixes every development decision before the independent test data are read. A Mask R-CNN stage localizes the dominant leaf, the accepted box is expanded by a validation-selected 5% margin and resized to a shared 224 by 224 input, and the same localized image is passed to ResNet50, InceptionV3, and MobileNetV2 together with an extractor that produces eighteen colour and shape descriptors. The forty-five class probabilities and eighteen descriptors form a sixty-three-dimensional input to a neural meta-classifier developed by three repetitions of stratified five-fold cross-validation. On a locked 3,101-image PlantVillage test partition of fifteen classes the framework reached 99.77% accuracy and 99.75% macro F1 with seven misclassifications, and a one-component-at-a-time ablation confirmed that every stage contributed. The same design was then trained and evaluated entirely within a separate thirteen-class PlantDoc field-image dataset, where it reached 90.03% accuracy and 89.40% macro F1. This is a within-PlantDoc experiment and not a controlled-to-field transfer test, so the figure measures how the pipeline behaves on field imagery rather than how a PlantVillage-trained model survives a domain shift. The central finding comes from running the identical ablation on both datasets: on clean images the ensemble breadth and descriptors provide the incremental gains, but under field conditions the ordering changes, and leaf localization and learned fusion become the decisive components. Removing localization cost 2.83 accuracy points and replacing the neural fusion with soft voting cost a further 2.08 points, the two largest effects on PlantDoc. The contribution is a controlled and transparent account of where each component of a localization-aware plant disease classifier earns its place, and of how that ordering shifts between laboratory and field acquisition.
Why it matches plant phenotyping methods葉画像から植物病害状態を推定する分類パイプラインの開発・アブレーション検証が中心であり、単なる病害実験や routine measurement ではない。
titleA Localization-Aware Heterogeneous CNN Ensemble with Neural Meta- Fusion for Plant Disease Classification, with a Component Analysis on Laboratory and Field Images
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe classification subset used here (20,638 images spanning fifteen pepper-bell, potato, and tomato classes) was obtained from PlantVillage, which is openly accessible at https://www.kaggle.com/datasets/emmarex/plantdisease.Open asset ↗Kaggle · emmarex/plantdiseaselines:314-336Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Accurate quantification of plant disease severity is essential for evaluating host-pathogen interactions and assessing the effectiveness of disease management strategies. Traditional visual scoring methods and manual estimation of infected tissue are widely used but are often subjective and prone to observer bias. Digital image analysis offers an objective alternative by enabling automated identification and quantification of symptomatic plant tissues based on color and spatial characteristics. Here, we present a MATLAB-based image processing protocol for differentiating diseased and healthy plant tissue from digital leaf images. The workflow involves acquisition of standardized leaf images, conversion of RGB images into hue-saturation-value (HSV) color space, segmentation of diseased tissue using defined HSV thresholds, refinement of the segmented mask through morphological operations, and extraction of the whole leaf area. The protocol then calculates the diseased area and total leaf area in pixels and computes the percentage of infected tissue. The method uses MATLAB together with the Image Processing Toolbox and can be implemented using simple scripts. This protocol enables rapid and reproducible quantification of disease severity in plant leaves exhibiting visually distinct symptoms such as necrotic lesions or blight patches. By minimizing observer bias and providing quantitative measurements of infected area, the protocol offers a practical and reproducible approach for plant disease phenotyping and evaluation of disease management strategies across diverse plant-pathogen systems where diseased tissues can be clearly distinguished from healthy tissues under reasonably controlled imaging conditions. Key features • A reproducible MATLAB-based workflow for separating diseased and healthy plant tissue using color-space segmentation. • Applicable to plant diseases where symptomatic tissue contrasts clearly with healthy tissue (necrosis, blight lesions, rot patches). • Requires digital leaf images, MATLAB, and the MATLAB Image Processing Toolbox for image processing and disease quantification. • Enables rapid calculation of diseased leaf area and disease severity using automated pixel-based quantification.
Why it matches plant phenotyping methods植物病斑を画像から分割・定量し、感染面積と病害重症度を算出するMATLAB画像解析プロトコルが研究の中心であり、植物病害表現型の取得・抽出手法に該当する。
abstractHere, we present a MATLAB-based image processing protocol for differentiating diseased and healthy plant tissue from digital leaf images.
Reproduction assets foundThe protocol explicitly deposits its authors' MATLAB image-processing workflow (HSV segmentation, mask refinement, pixel-based disease quantification) in a public GitHub repository with README instructions and example images.Code · publicGitHub repository containing the MATLAB source code, README file with installation and execution instructions, and representative example image(s): https://github.com/pankajborahmajuli-source/Leaf-Disease-Detection-MATLAB-Code/blob/main/README.mdOpen asset ↗Leaf-Disease-Detection-MATLAB-Codehtml-lines:112-148Code / dataset availability confirmedEurope PMC · Crossref · checked 15 Sept 2026
Background and aims Phytolith analysis is widely applied in palaeoecological and archaeological research, but its interpretive strength depends on the availability of robust modern reference collections. This study expands the modern Australian phytolith reference collection by analysing 42 native plant species representing 24 families and 37 genera with emphasis on silicification patterns across major growth forms, including forbs, shrubs, trees, and C3 grasses. Methods Phytoliths were extracted from available plant parts, including leaves, stems, flowers, seeds, seed pods, cones, and roots, depending on sample availability. Morphotypes were identified following ICPN 2.0, with grass silica short cell phytoliths (GSSCPs) further classified by shape and size to examine subfamily-level patterns. Phytolith morphotype percentage data were analysed using Hellinger transformation, PerMANOVA, PCA, LDA, and hierarchical clustering to assess compositional differences among plant growth forms and grass subfamilies. Key results Phytolith production varied strongly among growth forms and plant parts. Grasses were abundant producers, whereas most forbs, shrubs, and trees were trace producers or non-producers. Leaves were the most consistent source of phytoliths, while seeds and seed pods were predominantly non-producers. Grass silica short-cell phytolith (GSSCP) morphotypes showed clear subfamily-level differentiation. Pooideae produced Rondel morphotypes. Danthonoideae produced Rondel as well as wide Bilobate types. Panicoideae and Oryzoideae exhibited a pronounced Bilobate signature, commonly associated with Polylobate and Cross forms. Non-grass taxa (woody, shrubs, and forbs) were dominated by Spheroids, Tracheary elements, Epidermal, and Polygonal sheets and other non-diagnostic forms. Phytolith assemblages differed significantly among plant families, with Poaceae uniquely producing GSSCPs, while non-grass families showed greater overlap in assemblage composition. Conclusion By expanding taxonomic and anatomical coverage, this study strengthens the capabilities of phytoliths in the reconstruction of grasslands and in general paleo vegetation in Australia, especially where other proxies such as pollen are limited.
Why it matches plant phenotyping methods植物部位の珪酸体を抽出・形態分類し、成長形態やイネ科亜科を識別する現代参照コレクションを構築しており、植物形質の取得・判別手法が研究の中心である。
abstractThis study expands the modern Australian phytolith reference collection by analysing 42 native plant species representing 24 families and 37 genera with emphasis on silicification patterns across major growth forms
Reproduction assets foundThe authors explicitly state that the R scripts used for data analysis and figure generation are publicly available on their GitHub repository, which directly reproduces this paper's phytolith statistical analyses (PCA, LDA, PerMANOVA, clustering, plots). Supplementary data files contain the paper's measurements but noCode · publicntification of all plant specimens collected for this
study.
A
14 FUNDING
M
15 Funding for this study was provided by ARC Discovery grant DP210100508 and a Ph.D.
D
16 fellowship (UQGSS) to MH.
TE
17 DATA AVAILABILITY
18 The R scripts used for data analysis and figure generation are publicly available on GitHub
EP
19 repository: https://github.com/Manoshi-sporo/Australian-Phytolith-Reference-Collection.
20 CONFLICTS OF INTEREST
CC
21 The authors declare no competing financial or commercial interests.
A
22 AUTHOR CONTRIBUTIONS
23 MH: writing original draft, conceptualization, software, investigation. AC: Supervision, Writing -
24 Review and editing, FM: Supervision, Writing-Review and editing.Open asset ↗Manoshi-sporo/Australian-Phytolith-Reference-Collectionpdf-layout-page:34 lines:1-87Code / dataset availability confirmedOpenAlex · Europe PMC · bioRxiv · Crossref · checked 5 Sept 2026
Reliable and objective phenotyping is essential for plant breeding programs to characterize genetic variation and accelerate crop improvement. Conventional disease assessment relies on expert visual scoring, which is labor-intensive, subjective, and prone to inter- and intra-rater variability. Although image-based phenotyping methods have been proposed, many require manual intervention, specialized imaging setups, or single time-point measurements, limiting their ability to capture disease progression over time. Here, we present a pipeline for longitudinal plant disease phenotyping that quantifies wheat stripe rust and leaf rust progression from time-series images. The pipeline performs semi-automated leaf and automated pustule segmentation from images acquired in situ , enabling objective disease severity estimation with minimal user intervention and without requiring solid backgrounds or manual leaf manipulation or detachment. By extracting temporal traits, including disease severity trajectories and standardized area under the disease progress curve, the method provides a comprehensive characterization of disease development throughout infection. Association between automated and expert assessments was moderate for stripe rust ( R 2 = 0.58) and strong for leaf rust ( R 2 = 0.85), while expert inter-rater reliability was moderate for both diseases (ICC = 0.675 and 0.800, respectively). The proposed approach establishes a scalable and reproducible framework for longitudinal disease phenotyping in controlled environments, with broad applications in disease resistance screening and crop breeding.
Why it matches plant phenotyping methods画像時系列から植物病害の進展と重症度を抽出する半自動・自動解析パイプラインを開発し、専門家評価との比較で検証しており、表現型取得手法が研究の中心です。
abstractHere, we present a pipeline for longitudinal plant disease phenotyping that quantifies wheat stripe rust and leaf rust progression from time-series images.
Reproduction assets foundThe paper's Code and Data Availability section explicitly states that software and datasets (the phenotyping pipeline and imaging datasets) are publicly available at the authors' GitHub repository and project website, both of which are in the allowed URL list.Code · publicSoftware and datasets are available at:
https://github.com/USask-BINFO/greenskeye_analysis and https://greenskeye.usask.ca/speedbreeding/ .Open asset ↗USask-BINFO/greenskeye_analysislines:195-225Dataset · publicSoftware and datasets are available at:
https://github.com/USask-BINFO/greenskeye_analysis and https://greenskeye.usask.ca/speedbreeding/ .Open asset ↗lines:195-225Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published17 Aug 2026International Journal of Drug Delivery TechnologyCited by 0 · OpenAlex ↗
Why it matches plant phenotyping methodsトマト葉の病害を画像と深層学習で検出する手法が題名上の中心であり、植物の病害状態を観察的に推定するフェノタイピング研究に該当する。
titleDetection of Tomato Leaf Disease in Leaves with Deep Learning MobileNetV2 with Gaussian and Gabor Preprocessing
Reproduction assets foundThe paper's phenotyping analysis is based on the publicly available PlantVillage plant leaf disease image dataset hosted on Kaggle (54,303 labeled leaf images across 38 classes), which the authors explicitly state was sourced from a publicly available Kaggle dataset. No author-specific code, models, or derived datasetsDataset · publicThe research incorporated PlantVillage dataset(24)
accessible on Kaggle that contains 54,303 plant leaf
images showing both healthy and diseased
conditions spanning across 38 specific categories.Open asset ↗Kagglepdf-raw-page:2 lines:1-105Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract The classification of banana leaf disease has a large impact on agricultural output and relies heavily on timely early detection, with reliability as a fundamental component of effective crop management. The framework proposed in this study comprises a Hamiltonian Full Node Coverage Graph Attention Network (HFNC-GAT) and Fuzzy C-Means super pixel graph learning to explain the classification of banana leaf diseases. The HFNC-GAT allows for the representation of segmented leaf areas as the nodes in a graph. This framework also makes optimal use of an attention learning model to represent the spatial dependence of diseased leaf regions, allowing it to leverage both local and global spatial dependencies. The HFNC-GAT was demonstrated through observed experiments to achieve high performance with 96.11 and 94.19 accuracy, 0.9111 Cohen's Kappa, 0.9111 MCC, 0.9344 F2-score, and 0.9939 ROC-AUC compared to the performance of conventional CNN, GCN, and baseline GAT models.
Why it matches plant phenotyping methodsバナナ葉の病斑領域を画像から抽出・分類するグラフ学習手法が研究の中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として適格です。
abstractThe framework proposed in this study comprises a Hamiltonian Full Node Coverage Graph Attention Network (HFNC-GAT) and Fuzzy C-Means super pixel graph learning to explain the classification of banana leaf diseases.
Reproduction assets foundThe paper's Dataset Availability statement explicitly declares two public Kaggle banana leaf image datasets used for the phenotyping/classification experiments: Banana Leaf Disease Dataset V4 and BananaLSD. No author analysis code or trained model is reported as publicly available.Dataset · publicThe Banana Leaf Disease
Dataset V4 is available at https://www.kaggle.com/datasets/rayhanarlistya/banana-leaf-disease-dataset-v4.Open asset ↗Kaggle · banana-leaf-disease-dataset-v4pdf-page:19 lines:1-55Dataset · publicThe Banana Leaf Spot Diseases (BananaLSD) dataset is available at
https://www.kaggle.com/datasets/shifatearman/bananalsdOpen asset ↗Kaggle · bananalsdpdf-page:19 lines:1-55Code / dataset availability confirmedCrossref · checked 11 Sept 2026
Abstract Purpose Long-term monitoring of crop biophysical and biochemical traits remains challenging in high-latitude regions due to short growing seasons, frequent cloud cover, and highly variable weather. In this context, unmanned aerial vehicles (UAVs) offer flexible, high-resolution observations, but their added value relative to low-cost proximal sensors and their effectiveness for radiative transfer model (RTM) inversion across diverse crop canopies remain insufficiently quantified. This study evaluated the potential of a two-band proximal spectral reflectance sensor (SRS) and a five-band multispectral UAV sensor for retrieving leaf area index (LAI), leaf chlorophyll content (LCC), and canopy chlorophyll content (CCC) using PROSAIL inversion across major crops in Northern Europe over two growing seasons (2023–2024). Methods and Results Two inversion approaches – look-up table (LUT) and artificial neural network (ANN) were applied to PROSAIL simulations. UAV–PROSAIL–ANN outperformed LUT-based inversion and SRS observations, achieving the highest accuracy for LAI (R 2 = 0.81–0.95; RMSE = 0.27–0.77 m 2 /m 2 ), followed by CCC (R 2 = 0.58–0.94; RMSE 2 ), while LCC remained less accurately estimated (R 2 = 0.26–0.78; RMSE 2 ). Across sensors and methods, retrieval accuracy decreased in the order of LAI, CCC, and LCC, reflecting the stronger spectral control of canopy structure compared to biochemical traits. Conclusions The UAV–PROSAIL–ANN framework effectively captured spatial and temporal variability in crop traits, producing canopy-scale maps consistent with field observations. These results demonstrate the robustness and scalability of hybrid PROSAIL–ANN inversion for high-latitude crop monitoring, while highlighting current limitations in biochemical trait retrieval using multispectral data.
Why it matches plant phenotyping methodsUAV・近接分光センサーとPROSAIL反転、ANNを用いてLAIや葉・群落クロロフィルを推定し、精度比較と圃場観測との整合性評価を行うことが研究の中心である。
abstractThis study evaluated the potential of a two-band proximal spectral reflectance sensor (SRS) and a five-band multispectral UAV sensor for retrieving leaf area index (LAI), leaf chlorophyll content (LCC), and canopy chlorophyll content (CCC) using PROSAIL inversion across major crops in Northern Europe over two growing seasons (2023–2024).
Reproduction assets foundThe paper's Data availability statement explicitly deposits the authors' UAV image processing code (irradiance normalization, vignetting, exposure compensation, radiometric calibration) in a public GitHub repository under GPL v3.0; other data are available only upon request.Code · publicData availability Code to perform irradiance normalization, vignetting, exposure compensation, and radio-
metric calibration is available at https://github.com/fieldSITES/scripts/tree/main/UAV under GNU General
Public License v3.0. Other data will be made available upon request.Open asset ↗UAVpdf-page:34 lines:1-40Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Abstract Sustainable agriculture has substantial share on improvement of food security and optimization of resources utilization particularly for high value crops like rice leaf. Rice varieties should be properly classified in order to benefit the harvest management, reduced loss after harvest and improved agriculture methods. The traditional classification method usually brings the low precision and the traditional classification method is also subjected to human error, which is difficult to bring about reliable output. This study introduces an optimized multi-class rice leaf disease classification system utilizing “Rice Feature Selection” (RiceFS) and ensemble machine learning approaches. RiceFS is realized based on a feature selection mechanism based on Recursive Feature Elimination. Selected classifiers such as KNN, Random Forest, Gradient Boosting, Ensemble Learning and Optimized SVM are analyzed based on the extracted subset of features and the proposed system is used to classify the seven classes of rice leaf disease. The experimental results show that the Optimized SVM has the best classification results among the different classifiers with accuracy of 92.10%, Precision of 92.20%, balanced Recall and F1 Score, which shows that Optimized SVM is very effective in multi-class rice leaf disease classification. The performance can be improved by feature reduction, generalization capability and computational complexity reduction, which are realized with the help of RiceFS. The proposed framework is designed to provide an intelligent decision support system for timely intervention, loss minimization and sustainable agriculture. Results indicate that these algorithms are applicable for rice leaf disease classification since they are accurate, reliable and scalable.
Why it matches plant phenotyping methodsイネ葉の病害状態を観察データから分類する計算手法が研究の中心であり、RiceFSと複数の機械学習器を用いた分類フレームワークを開発・評価しているため。
abstractThis study introduces an optimized multi-class rice leaf disease classification system utilizing “Rice Feature Selection” (RiceFS) and ensemble machine learning approaches.
Reproduction assets foundThe paper's RiceFS phenotyping/classification experiments are built on two public Kaggle rice leaf disease image datasets, explicitly cited with URLs and a data availability statement. No author code or models are deposited.Dataset · publicThe RiceFS framework proposed initially
performs a feature selection, followed by training
several classifiers: K-Nearest Neighbors (KNN),
Random Forest (RF), Gradient Boosting (GB),
Ensemble Learning, and Optimized Support
Vector Machine (Optimized SVM). The data is
published on the Kaggle website. The data is
open-source at:
https://www.kaggle.com/datasets/vbookshelf/rice-leaf-diseases [44].
This data consists of 120 jpgs of disease infected
rice leaves. The photos are divided into 3
categories according to the kind of disease. There
are 40 images in each class.
Classes
• Leaf smut
• Brown spot
• Bacterial leaf blight
The datasets are preprocessed by eliminating
redundant information, normOpen asset ↗Kaggle · vbookshelf/rice-leaf-diseasespdf-raw-page:11 lines:1-103Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published11 Aug 2026Engineering, Technology & Applied Science ResearchCited by 0 · OpenAlex ↗
Corn is a staple crop of global significance; however, foliar diseases may lead to 30–60% yield loss if not detected at an early stage. Conventional visual inspection is time-consuming, subjective, and difficult to scale for smallholder farmers worldwide. To overcome these issues, we present Corn Transfer Learning Network (CTL-Net), an end-to-end hybrid deep learning model for corn leaf disease identification. CTL-Net, which combines Inception-ResNet-v2 as the backbone and MobileNetV3 as a feature extractor in parallel convolutional streams, simultaneously learns diverse scales of texture information from low-level textures, mid-level structural patterns, and high-level disease semantics of RGB leaf images. Adaptive feature fusion is formulated through learnable weighting coefficients and bi-directional spatial–channel attention mechanisms, which further enhance feature discriminability and robustness. The proposed approach is tested on an extensive dataset of 12,456 images from 10 corn diseases, including Northern Leaf Blight, Common Rust, Gray Leaf Spot, and Cercospora Leaf Spot, acquired under controlled and real-field conditions. CTL-Net attains the highest classification accuracy of 99.42%, outperforming DenseNet121 (97.92%), EfficientNetB3 (97.35%), and general stacking models (97.89%). Robustness experiments demonstrate the effectiveness of the proposed method against illumination variations, additive noise, and partial occlusions. CTL-Net enables real-time inference with a latency of 42 ms on an NVIDIA RTX 3090 GPU. Gradient-weighted Class Activation Mapping++ (Grad-CAM++)-based interpretability analysis results in a mean Intersection over Union (IoU) of 87.6% with expert-annotated disease regions. Five-fold cross-validation, ablation studies, and statistical significance testing (p
Why it matches plant phenotyping methodsトウモロコシ葉画像から病徴・病害状態を推定する深層学習手法を開発し、複数モデルとの比較、頑健性評価、交差検証、アブレーションを行っており、植物フェノタイピング手法が中心である。
abstractwe present Corn Transfer Learning Network (CTL-Net), an end-to-end hybrid deep learning model for corn leaf disease identification.
Reproduction assets foundThe paper states its final curated corn leaf disease dataset (12,456 images, 10 classes) is publicly available via the authors' GitHub repository vishruthkp/maizedataset (reference [30] and Data Availability statement). The Kaggle PlantVillage and Corn or Maize Leaf Disease datasets are cited source inputs, not paper-Dataset · public"Prediction of Crop Yield using Machine Learning," International Available: https://github.com/vishruthkp/maizedataset.Open asset ↗vishruthkp/maizedatasetpdf-page:7 lines:1-53Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Field / plotLeafClassificationStress / disease detectionDisease symptoms / severity
Automated plant disease recognition from field photographs remains challenging: models trained on laboratory datasets collapse on in-the-wild images, and the current state of the art on PlantWild-the largest open in-the-wild benchmark (18,542 images, 89 classes)-relies on text prototypes derived from a language model to reach 76.18% top 1 accuracy. We ask whether text supervision is truly necessary or whether the bottleneck is the diversity of the visual representation. Our central methodological finding is that backbone selection and fusion matter far more than classifier-head engineering : across 118 experiments, a simple fixed-weight linear-prototype combination on top of three complementary frozen backbones yields larger and more reproducible gains than any head-level adaptive routing mechanism we test. Specifically, through a systematic study of six frozen vision backbones (three CLIP, two DINOv3, and one DINOv2), three classifier heads, and four fusion configurations, completed in a single day on one consumer GPU, we establish three findings. (i) A single self-supervised backbone (DINOv2 ViT-L/14) already surpasses text-augmented MVPDR (77.56% vs. 76.18%). (ii) Concatenating three complementary backbones (DINOv2 + DINOv3 + CLIP) reaches 80.23% ± 0.41% (five seeds), exceeding the published MVPDR accuracy by +4.05 points and our own reproduction of MVPDR under an identical evaluation protocol by +7.96 points, without any language supervision. (The difference between the two deltas reflects evaluation-protocol differences-our split, model-selection criterion, and training schedule-rather than any discrepancy in the reported numbers; see Section 4.6.6 for a full reconciliation.) (iii) Every form of learned routing we test-per-class gating, backbone gating, sample-wise gating-is inessential; the gain is entirely attributable to backbone diversity and a simple linear-prototype scoring combination. On the smaller PlantDoc benchmark, the same principle transfers but with substantially higher seed variance: the best configuration reaches 80.09% at a favourable seed but 76.97% ± 1.48% over five seeds-a suggestive rather than robust gain. Beyond the accuracy headline, we provide a pathology-aware per-class analysis showing that DINOv2/v3 dominate on fine-texture lesion classes (rusts, mildews, and leaf spots) whilst CLIP's narrow advantage concentrates on organ/species-level identification (rice leaf and potato late blight). All primary claims are validated over five seeds, and all code, feature caches, and result files are released for full reproducibility.
Why it matches plant phenotyping methods植物病害の画像認識を対象に、複数の視覚基盤モデル融合と分類器を体系比較・検証しており、病害状態の推定手法が研究の中心である。
abstractOur central methodological finding is that backbone selection and fusion matter far more than classifier-head engineering
Reproduction assets foundThe paper states that its code, cached features, and result files will be released, but no authors' public URL or repository is provided in the supplied blocks (future-tense availability language only). The DINOv3 GitHub link and OpenReview link are third-party backbone resources, not paper-specific assets.Code · publicAll primary claims are validated over five seeds, and all code, feature caches, and result files are released for full reproducibility.Open asset ↗lines:332-335Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published11 Aug 2026Engineering, Technology & Applied Science ResearchCited by 0 · OpenAlex ↗
In recent years, areca nut plants have been vulnerable to different diseases that appear as distinct colors on leaves, caused by bacteria or fungi. These symptoms disrupt photosynthesis and reduce yield, affecting productivity and crop health. Therefore, accurate plant disease classification is essential for detecting distinct disease shapes and sizes. Existing Deep Learning (DL) models have several limitations that prevent them from distinguishing between various plant diseases due to similar characteristics. To overcome this limitation, a Dynamic elastic boundary strategy Sailfish Optimization Algorithm and Convolution Neural Network (DSFO-CNN) method is proposed to identify and accurately classify arecanut plant diseases. The Visual Geometry Graph-19 (VGG-19) model extracts features that have significant information about disease in arecanut plants. The proposed arecanut plant disease classification model employed feature selection and drop cyclic learning rate, which adjusts the CNN learning rate to efficiently learn the subtle information about various leaf and nut diseases to enhance classification. The experimental results of the DSFO-CNN demonstrate superior performance compared to existing approaches.
Why it matches plant phenotyping methodsアレカヤシの葉・果実に現れる病徴を画像から分類するCNNベース手法を提案・評価しており、植物病害状態の取得・推定が中心的な方法論的貢献である。
abstractTherefore, accurate plant disease classification is essential for detecting distinct disease shapes and sizes.
Reproduction assets foundThe paper's phenotyping inputs are two public image datasets: the collected Arecanut dataset (Kaggle) and the PlantVillage dataset (Kaggle), both explicitly cited and declared openly available. No author analysis code or trained model is released.Dataset · publicDATA AVAILABILITY
The data used in this study are openly available at [19] and
[20].Open asset ↗pdf-page:7 lines:1-63Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
leaf development was tracked dynamically from high-throughput phenotyping image series using the SAM-2 deep learning model, enabling automated segmentation and tracking of individual leaves across hundreds of plants from different accessions grown under different water and nitrogen conditions. From these time series, leaf-level growth dynamics were reconstructed and used to calculate key developmental traits, including phyllochron and relative expansion rates. These measurements were further integrated into the GreenLab model to infer plant-scale parameters such as radiation use efficiency (RUE) and leaf demand parameters. Statistical analyses revealed significant genotype, treatment, and interaction effects on several traits, with stable genotype rankings but contrasted sensitivities to environmental conditions. While whole-rosette growth captured strong and consistent responses, individual leaf contributions to global growth remained limited, highlighting the integrative nature of rosette-scale dynamics. Model-derived parameters provided additional insight into plant strategies, showing that genotypes differ in both the duration and timing of leaf demand, with generally stable patterns across environments. In parallel, QTL mapping was performed using leaf-level measurements and phyllochron traits. This analysis identified shared and trait-specific genetic loci, including loci associated with leaf emergence dynamics that were not detected using traditional rosette-scale integrative traits. Overall, this study demonstrates that combining deep learning-based image analysis with mechanistic modeling enables large-scale and quantitative characterization of plant development, providing biologically interpretable traits and new insights into their genetic and environmental determinism. It highlights the potential of combining high-throughput image-based phenotyping, deep learning, and mechanistic modeling to generate biologically interpretable traits and to assess genetic and environmental influences on these traits.
Why it matches plant phenotyping methods深層学習による葉の自動セグメンテーション・追跡を開発的に適用し、時系列画像から葉レベルおよび植物体レベルの発達形質を定量化しているため、表現型取得・抽出が研究の中心である。
abstractleaf development was tracked dynamically from high-throughput phenotyping image series using the SAM-2 deep learning model, enabling automated segmentation and tracking of individual leaves across hundreds of plants
Reproduction assets foundThe paper explicitly states that the authors' leaf segmentation software (AraLeaf_segmentation), the GreenLab Arabidopsis model (AraGreenlab), and the statistical analysis code (AraParameters_statistical_analyses) are open-source and publicly hosted on forge.inrae.fr, and that the data and results used in the paper areCode · publicThe leaf segmentation software (AraLeaf_segmentation) and the GreenLab model of Arabidopsis thaliana (AraGreenlab) are open-source software, and distributed under the GNU GPL v3 licence. The two software packages, and the code to run statistical analyses (AraParameters_statistical_analyses) are available at:https://forge.inrae.fr/phenoscope-public/plant_phenomics_suppmat.Open asset ↗phenoscope-public/plant_phenomics_suppmathtml-lines:419-444Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
The tomato plant is considered one of the most important crops in the world, yet it is vulnerable to various diseases that affect crop quality and agricultural productivity. These challenges have driven the need for an efficient and intelligent plant disease detection system. With the development of computer vision and artificial intelligence, this proposed methodology based on deep learning for tomato leaf diseases has been presented. Two public datasets: Taiwan DS with nine classes and Tomato Leaf Diseases Detection Computer Vision Dataset (TLDDCV DS) with seven classes have been used to test this system. This system begins with plant image processing, which includes gamma correction and bilateral filtering, to enhance image quality and clarity while preserving key disease features. Then, a genetic metaheuristic algorithm was used to automatically select the most significant hyperparameters, further optimizing both processing time and accuracy. After that, the tomato leaf disease detection applies the You Only Look Once version 11 Nano (YOLOv11n) model. The YOLOv11n backbone is edited through a Data-efficient Image Transformer (DeiT) to improve the system's capacity for learning global contextual information and long-range dependencies. Experimental results demonstrate that the proposed system outperforms existing methods. It achieved an average mAP@50 of 97.8%, mAP@50-95 of 93.4%, precision of 97.3%, recall of 93.8%, and F1-score of 95.5% on the Taiwan dataset. Additionally, it achieved an average mAP@50 of 87%, mAP@50-95 of 48%, precision of 83.9%, recall of 70.3%, and F1-score of 76.4% on the TLDDCV dataset. These results demonstrate the generalizability and effectiveness of the proposed system in real-world agricultural situations.
Why it matches plant phenotyping methodsトマト葉の病徴を画像から検出・分類する深層学習手法の開発と2データセットでの性能評価が研究の中心であり、植物病害状態のフェノタイピングに該当する。
abstractthe tomato leaf disease detection applies the You Only Look Once version 11 Nano (YOLOv11n) model.
Reproduction assets foundThe paper uses two public Roboflow tomato leaf disease image datasets and states its source code is publicly available on Zenodo, all with explicit availability statements and URLs.Dataset · publicThe first dataset is the Taiwan dataset, which can be found at the following link: (https://universe.roboflow.com/bryan-b56jm/tomato-leaf-disease-ssoha).Open asset ↗tomato-leaf-disease-ssohalines:317-328Dataset · publicThe second dataset is the TLDDCV dataset, which can be found at the following link: (https://universe.roboflow.com/sylhet-agricultural-university/tomato-leaf-diseases-detect)Open asset ↗tomato-leaf-diseases-detectlines:317-328Code · publicThe source code of the proposed framework, including the implementation of the proposed methodology and experimental setup, is publicly available in the Zenodo repository: https://doi.org/10.5281/zenodo.20777853 .Open asset ↗Zenodo · 10.5281/zenodo.20777853lines:317-328Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Identifying plant health conditions is an emerging precision-agriculture and food-security challenge, intensified by deploying deep-learning models on memory- and power-constrained edge devices. We present a TinyMLOps pipeline spanning model design, optimization, quantization, and deployment across diverse edge devices, evaluated under controlled laboratory conditions. Using a dataset derived from the PlantVillage benchmark, 39 fine-grained classes are aggregated into three superclasses: healthy leaf, unhealthy leaf, and no leaves. The resulting system therefore performs plant health-status classification and background filtering rather than diagnosing specific diseases. We train a MobileNet-based convolutional neural network jointly optimized for classification accuracy and computational efficiency, adopting state-of-the-art hyperparameter optimization (HPO) tools. Five models are selected, four from the Pareto Front and one as the biggest evaluated model during HPO, converted to LiteRT and ONNX, and evaluated at float32 and post-training int8 precision on a Raspberry Pi Zero 2 W and an STM32H743ZI microcontroller. At float32, LiteRT is 1.87–2.65× faster than ONNX Runtime on the Raspberry Pi across all five models. Relative to their float32 LiteRT counterparts, the int8 LiteRT models are 2.83–3.67× smaller on disk and 21.3–31.2% faster on the same board, at a cost in F1-score of between 0.0010 and 0.0068. On the microcontroller, only the two smallest models deploy at both precisions; for these, the fully quantized int8 variants are 4.3× faster and 3.85× smaller in MCU flash footprint than the float32 counterparts. The mid-range model fits the 2 MB flash and 1 MB RAM budget only when quantized, while the two largest models exceed it in every configuration tested.
Why it matches plant phenotyping methods葉画像から植物の健康状態を推定する分類手法と、エッジデバイス向けの最適化・量子化・展開パイプラインが研究の中心であり、植物状態の画像ベースフェノタイピングに該当する。
abstractWe present a TinyMLOps pipeline spanning model design, optimization, quantization, and deployment across diverse edge devices, evaluated under controlled laboratory conditions.
Reproduction assets foundThe paper's plant-phenotyping measurements are based on a derived PlantVillage dataset (39 classes aggregated into three superclasses) that the authors explicitly state is openly available in their own GitHub repository, also catalogued in the AgrifoodTEF Data Space. No author analysis code or trained model checkpointsDataset · publicThe data used in this study are derived from the openly available GitHub repository available at https://github.com/FBK-OpenIoT/PlantVillage-AugNoLeaves , accessed on 1 July 2026.Open asset ↗FBK-OpenIoT/PlantVillage-AugNoLeaveslines:1355-1403Dataset · publicPlantVillage-AugNoLeaves—AgrifoodTEF Data Space Catalogue. 2025. [(accessed on 1 July 2026)]. Available online: https://dataspace.agrifoodtef.eu/asset/did:op:3091fdcf83a05784416e585e4e45ea24afaf8c925a891921a3949acd98426f65Open asset ↗did:op:3091fdcf83a05784416e585e4e45ea24afaf8c925a891921a3949acd98426f65lines:1424-1474Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Precise and reliable diagnosis of leaf diseases in tomato is essential for enhancing crop cultivation and minimizing agricultural losses. While deep learning models have performed well on benchmark datasets, the majority of present techniques rely on flat multi-class classification, which predicts all disease categories simultaneously. Such formulations promotes inter-class confusion, particularly when biologically different diseases with similar visual symptoms are learned within a same model. To overcome this constraint, we propose a biologically structured hierarchical deep learning framework in this study. Instead of directly classifying 10 disease classes, the proposed method first classifies leaf images into meaningful biological groups such as bacterial, fungal, pest-associated and healthy using a vision transformer (ViT) model. Then, specialized convolutional neural network (CNN) experts perform fine-grained classification within each category. The proposed hierarchical model shows an overall accuracy of 97.8%, when validated on PlantVillage tomato dataset. A flat ViT model trained with class-weighted loss obtained 96.3% accuracy, whereas a flat CNN model reached 99.3% under clean conditions but decreased sharply to 41% under Gaussian perturbation ( σ = 0.05). On the other hand, the hierarchical model performed steadily under noise with 97.4% accuracy at the same perturbation level. These results indicate that adding biological structure to model design reduces confusion, helps prevent imbalance effects and increases robustness, providing a more trustworthy and interpretable solution for real-world agricultural disease diagnosis.
Why it matches plant phenotyping methodsトマト葉画像から病害状態を分類する深層学習手法を開発し、既存モデルとの比較およびノイズ下での技術検証を行っており、植物表現型（病害状態）の取得・推定が中心である。
abstractwe propose a biologically structured hierarchical deep learning framework in this study.
Reproduction assets foundThe paper's phenotyping/classification measurements are based on the public PlantVillage tomato leaf image dataset, which the authors explicitly state was analyzed and provide a public Kaggle URL. No author analysis code, trained models, or paper-specific supplementary assets are described with availability language.Dataset · publicIsabel Luna-Maldonado , Autonomous University of Nuevo León, Mexico
Reviewed by: Noredine Hajraoui , Moulay Ismail University, Morocco
Tri Handhika , Universitas Gunadarma Pusat Studi Komputasi Matematika, Indonesia
Data availability statement
Publicly available datasets were analyzed in this study. This data can be found here: https://www.kaggle.com/datasets/charuchaudhry/plantvillage-tomato-leaf-dataset .
Author contributions
HG: Conceptualization, Formal analysis, Methodology, Writing – original draft, Writing – review & editing. SR: Supervision, Validation, Writing – review & editing. BL: Supervision, Validation, Writing – review & editing.
Conflict of interest
The author(s) declared thaOpen asset ↗Kaggle · charuchaudhry/plantvillage-tomato-leaf-datasetlines:588-616Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
TomatoLiDAR / point cloudLeafMorphology / geometry measurementSegmentationLeaf traits
Leaf parameters are crucial indicators reflecting the growing status of plants. Monitoring and analysis of leaf parameters significantly contributes to the improvement of crop yield and food quality. This study focused on three tomato plant varieties commonly grown in the Netherlands and proposed a fully automatic pipeline for leaf phenotyping. Three-dimensional (3D) point clouds of target plants were acquired with a specially designed imaging unit naming Maxi-Marvin. A semantic segmentation of plant organs was performed with PointNet++ model. To mitigate point cloud resolution decrement, the down-sampling operation in the baseline model was replaced with a distributed segmentation strategy. Leaf instances were further identified with Density-Based Spatial Clustering of Applications with Noise (DBSCAN), followed by a morphological phenotypic trait quantification based on 3D geometrical analysis. Target phenotypic traits including leaf length, leaf width, and leaf area. The evaluation results indicated that the distributed segmentation strategy achieved the best F 1 scores of 0.98 with block size set to 30,000. The Mean Average Errors (MAE) of leaf length, leaf width, and leaf area estimation were 2.09 cm, 1.78 cm and 8.98 cm 2 respectively. The estimation accuracies for leaf length, leaf width, and leaf area were 91.98%, 92.66%, and 89.67%, respectively.
Why it matches plant phenotyping methodsトマト葉の3D画像取得、器官セグメンテーション、葉インスタンス識別、形態形質推定を統合した自動フェノタイピング手法を開発・評価しており、方法が研究の中心です。
abstractproposed a fully automatic pipeline for leaf phenotyping
Reproduction assets foundThe paper's tomato point cloud dataset (with semantic and leaf instance annotations used for the phenotyping pipeline) is publicly available on Kaggle via a footnote. NPEC website is a facility page, and Open3D is a generic library, so neither qualifies.Dataset · public2. ^ The dataset used in this study is available at: https://www.kaggle.com/datasets/xinbolai/vtc-tomatoOpen asset ↗Kaggle · xinbolai/vtc-tomatolines:545-624Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
The simultaneous diagnosis of diseases and evaluation of age quality grades in tea leaves are critical for precision agriculture and the economic valuation of tea products. Although deep learning has shown promise in agricultural vision tasks, current multi-task models often suffer from performance degradation due to feature conflicts: tea leaf disease recognition relies heavily on macro-structural lesions, whereas tea leaf-age quality grading depends on micro-textural features such as trichome density and color uniformity. To address this discrepancy, we propose a novel dual branch fusion network. Our architecture fundamentally decouples the feature extraction process by utilizing a dual branch mechanism. The first branch employs global average pooling to capture first-order spatial statistics; it can retain the global structural layout necessary for macro-lesion detection. The second branch introduces a dimensionality-reduced self-bilinear pooling module to compute second-order covariance matrices; it can effectively capture the fine-grained textural patterns essential for micro-grade classification. These decoupled features are subsequently fused and optimized through a weighted multi-task loss function. Experimental results on a comprehensive tea leaf dataset demonstrate that the proposed dual fusion framework significantly outperforms baseline models. The proposed network can rescue the disease classification accuracy drop observed in standard bilinear models while maintaining exceptional grading performance. Furthermore, the proposed network maintains a compact parameter footprint and low computational complexity. This balance renders it suitable for deployment on agricultural Internet of Things edge devices where inference speed is critical.
Why it matches plant phenotyping methods茶葉の病害状態と葉齢品質を画像から推定する深層学習手法を開発し、データセット上でベースラインと比較評価しており、表現型取得・推定法が中心である。
abstractwe propose a novel dual branch fusion network
Reproduction assets foundThe paper's Data Availability Statement publicly releases the two tea leaf image datasets used for its phenotyping tasks (disease recognition and leaf-age quality grading) via Mendeley Data. The authors' analysis code and trained models are only promised 'upon acceptance' with no public URL, so they do not qualify.Dataset · publicThe tea leaf disease recognition dataset analyzed in this study is available from https://data.mendeley.com/datasets/744vznw5k2/3 (accessed on 11 February 2026)Open asset ↗744vznw5k2/3lines:514-565Dataset · publicthe tea leaf grading dataset is available from https://data.mendeley.com/datasets/7t964jmmy3/1 (accessed on 11 February 2026)Open asset ↗7t964jmmy3/1lines:514-565Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Reliable traits are needed for identification of tea ( Camellia sinensis ) cultivars, yet the stability of leaf morphology and color across leaf positions remains unclear. This study evaluated inter-cultivar variation and positional stability in leaf morphological, RGB color, and SPAD traits in six predominant cultivars. One-year-old shoots were sampled in a completely randomized design, and five fully expanded leaves below the apical bud were analyzed. SPAD values were measured with a chlorophyll meter, and scanned images were used to extract contour and RGB traits. Data were analyzed using ANOVA, correlation analysis, PCA, and discriminant analysis. Leaf morphology differed among cultivars and leaf positions, with significant cultivar-by-position interactions; however, the width-to-length ratio differed among cultivars but remained stable across positions in these cultivars. SPAD values increased with leaf position and were strongly associated with RGB components, being negatively correlated with R and G and positively correlated with B. Morphological traits explained 52.988% of total variance in PCA and yielded 64.6% overall classification accuracy, with LaoHan showing the highest accuracy (83.3%). Misclassification was concentrated among genetically similar cultivars. These findings suggest that stable leaf shape proportions and SPAD-RGB relationships provide useful descriptors, whereas genetic relatedness limits morphology-based cultivar identification under the present conditions.
Why it matches plant phenotyping methods茶品種識別のため、葉の形態・RGB・SPAD特性の取得と安定性、分類性能を中心に評価しており、画像由来形質抽出を含む実質的な表現型解析である。
abstractscanned images were used to extract contour and RGB traits
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/biology15151283/s1 , Table S1: Original data of leaf morphological traits, RGB values, and SPAD values from six tea cultivars in this study.Open asset ↗lines:368-409Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published1 Aug 2026International Journal of Advances in Data and Information SystemsCited by 0 · OpenAlex ↗
Plant diseases have continued to threaten agricultural productivity, while manual inspection methods have remained inefficient and prone to subjectivity. This study proposed and assessed a hybrid framework integrating a Convolutional Neural Network (CNN) with a Large Language Model (LLM) to perform image-based plant leaf disease classification accompanied by interpretable diagnostic explanations. EfficientNetV2-M was employed as the visual backbone and trained on 11 selected classes of apple, grape, and potato leaf images derived from the PlantVillage dataset. A structured data splitting strategy was applied to ensure reliable model validation and unbiased testing. The classification capability of the CNN component was examined through standard multi-class evaluation indicators, including class-wise predictive consistency and error distribution analysis. Experimental results indicated that the model delivered highly consistent predictions, reaching a peak test accuracy of 99.79%, reflecting its robustness in distinguishing visually similar disease patterns. To overcome the black-box limitation, prediction outputs were transformed into structured prompts and processed by GPT-4o to generate contextual explanations. The generated narratives systematically described observable symptoms, highlighted distinguishing characteristics, and suggested initial management actions. Overall, the proposed hybrid system demonstrated that combining high-performance visual recognition with language-based reasoning enhanced both diagnostic reliability and interpretability in digital agriculture applications.
Why it matches plant phenotyping methods植物葉の画像から病害状態を推定するCNN・LLM統合手法を開発・評価しており、病害分類と説明生成が研究の中心であるため。
abstractThis study proposed and assessed a hybrid framework integrating a Convolutional Neural Network (CNN) with a Large Language Model (LLM) to perform image-based plant leaf disease classification accompanied by interpretable diagnostic explanations.
Reproduction assets foundThe paper uses the public PlantVillage color dataset (11 apple/grape/potato classes, 9,385 images) and explicitly points to it in the DATA AVAILABILITY statement as the replication package data. The authors also provide a public Streamlit demonstration of their hybrid CNN–LLM system. No author analysis code or trained-Dataset · publicaper. The research was conducted for academic purposes, and no financial, commercial, or
personal relationships influenced the study design, data analysis, interpretation of results, or
preparation of the manuscript.
DATA AVAILABILITY
The data associated with this study are publicly available online in the replication package.
[https://github.com/spMohanty/PlantVillage-Dataset/tree/master/raw/color]
AUTHOR CONTRIBUTIONS
Frenky Riski Gilang Pratama: Conceptualization; Programming and coding
implementation; Methodology; Writing-Original Draft. Sugiarto Surono: Conceptualization;
Methodology; Supervision; Writing-Review & Editing. Aris Thobirin: Proofreading Paper;
Writing-Review & Editing; FunOpen asset ↗https://github.com/spMohanty/PlantVillage-Dataset/tree/master/raw/colorpdf-raw-page:10 lines:1-52Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Shoot apical meristem (SAM) homeostasis integrates environmental and genetic cues to regulate growth dynamics that drive biomass accumulation and crop yield; however, no robust, non-destructive, quantitative proxy has been established for modeling or monitoring SAM-homeostasis-associated dynamics. Here, we developed a novel robot-based 3D imaging system and a custom pot-chamber gas exchange system to non-destructively measure plant occupation volume (POV) and whole-plant photosynthetic rate in wild-type Arabidopsis plants and nine mutants with disrupted SAM homeostasis. We demonstrate that POV robustly captures 3D plant architecture, whereas whole-plant photosynthetic rate serves as a superior proxy for optimal growth dynamics and final biomass associated with SAM homeostasis, outperforming conventional traits such as leaf number, leaf size, total leaf area, and rosette diameter. The strong positive correlations among POV, whole plant photosynthesis, and biomass accumulation establish a powerful new framework for quantitative studies of SAM homeostasis and data-driven evaluation of plant architecture.
Why it matches plant phenotyping methodsロボット3D画像とカスタムガス交換による非破壊的な植物形態・光合成表現型測定系を開発し、従来形質との比較検証も行っており、方法が研究の中心である。
abstractwe developed a novel robot-based 3D imaging system and a custom pot-chamber gas exchange system to non-destructively measure plant occupation volume (POV) and whole-plant photosynthetic rate
Reproduction assets foundThe paper's authors explicitly state that the Python source code for whole-plant leaf-area segmentation, 3D point cloud processing, POV calculation, and Mask3D-based segmentation is publicly available on GitHub at https://github.com/songqingfeng/AtPOVcalculator. This is a paper-specific, public, actionable analysis/PhDCode · publicThe Python source code for whole-plant leaf-area segmentation and calculation is publicly available on GitHub ( https://github.com/songqingfeng/AtPOVcalculator ).Open asset ↗songqingfeng/AtPOVcalculatorlines:224-233Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Sugarcane plant diseases pose a significant threat to agricultural productivity, yet early and accurate identification remains challenging for farmers due to the limitations of manual inspection. This study proposes a sugarcane leaf disease classification system using ConvNeXt V2 Tiny, a modern convolutional architecture with a Global Response Normalization (GRN) mechanism, combined with an ensemble Stratified K-Fold Cross Validation strategy (K=6) to improve generalization on real-world field data. A dataset of 2,948 leaf images spanning five classes (Red Rot, Mosaic, Rust, Yellow Leaf, and Healthy) was used, with field-collected images held out as a fixed test set. The ensemble model achieved a mean validation accuracy of 98.49% ± 0.58% across six folds and a test accuracy of 98.39% on 427 unseen field images, with macro-average precision, recall, and F1-score each reaching 98%. ConvNeXt V2 Tiny substantially outperformed ResNet-50 (87.35%) and EfficientNetV2-S (83.37%) under identical experimental settings, demonstrating superior generalization across the domain gap between curated and field data. The primary contribution of this study is the first application of ConvNeXt V2 Tiny with ensemble K-Fold strategy for sugarcane disease classification, offering high accuracy with moderate computational complexity (28.6M parameters) and practical deployability, as demonstrated through the SugarScan web application.
Why it matches plant phenotyping methodsサトウキビ葉画像から病害状態を推定する画像ベースの表現型解析手法が研究の中心であり、モデル性能の検証・比較も実施しているため含める。
abstractThis study proposes a sugarcane leaf disease classification system using ConvNeXt V2 Tiny
Reproduction assets foundThe paper's phenotyping inputs include a public Kaggle dataset (Sugarcane Leaf Disease Dataset, SLD) of sugarcane leaf disease images used for training/validation, plus field-collected images. Only the Kaggle dataset qualifies as a paper-specific public asset with an authors' URL; no author analysis code, trained modelDataset · publicsecondary data from the Sugarcane Leaf Disease Dataset (SLD)
available publicly on Kaggle (https://www.kaggle.com/datasets/pritpal2873/sug arcane-leaf-disease-dataset)Open asset ↗Kaggle · pritpal2873/sugpdf-page:2 lines:54-60Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Camera-based visual sensing provides a non-destructive and scalable approach for monitoring strawberry diseases and pests in greenhouse environments. However, greenhouse images acquired under practical cultivation conditions often contain early-stage tiny lesions, complex leaf backgrounds, uneven target scales, illumination variations, and partial occlusions, making accurate and efficient visual detection challenging. To address these issues, this study proposes YOLOv8n-DSLW (YOLOv8n enhanced by Dense reuse, Shuffle attention, LSKA-LAMP lightweight modeling, and Wise-IoU optimization), an AI-enabled vision-sensing detection model based on YOLOv8n for tiny strawberry disease and pest detection. Specifically, Shrink Residual Dense Block (ShrinkRDB) dense connection blocks and the C2f with Shuffle Attention (C2fSA) module are introduced to preserve weak lesion textures and suppress background interference in greenhouse visual data. A high-resolution P2 detection layer combined with Wise-IoU (WioU) dynamic regression loss is further incorporated to enhance tiny-target perception and localization. In addition, the Spatial Pyramid Pooling-Fast with Large Separable Kernel Attention (SPPF-LSKA) module strengthens contextual modeling under occlusion and clutter, while Layer-Adaptive Magnitude-based Pruning (LAMP) is adopted to mitigate model redundancy and improve the accuracy-efficiency balance. Experiments on a self-collected greenhouse strawberry disease and pest dataset show that YOLOv8n-DSLW achieves a mean Average Precision at 0.5 IoU threshold (mAP@0.5) of 94.3% and a mAP@0.5:0.95 of 77.5%, outperforming the YOLOv8n baseline. The final model has a parameter count of 4.386 M and a computational cost of 27.6 GFLOPs, achieving a frame rate of 45 FPS on the test workstation. It shows application potential for real-time visual monitoring in greenhouses under controlled data acquisition conditions. The results demonstrate that the proposed method improves tiny lesion detection under dense targets, complex backgrounds, and leaf occlusions, providing an AI-enabled vision-sensing framework for automated strawberry health monitoring in greenhouses. Nevertheless, due to limitations associated with imaging equipment, dataset representativeness, and the inherent constraints of the algorithm, further optimization and validation are required to support large-scale field deployment.
Why it matches plant phenotyping methodsイチゴ葉の病斑を画像から検出・局在化する新規YOLOモデルを開発し、データセット上で性能評価しており、植物病害状態の画像ベース表現型取得が中心である。
abstractthis study proposes YOLOv8n-DSLW (YOLOv8n enhanced by Dense reuse, Shuffle attention, LSKA-LAMP lightweight modeling, and Wise-IoU optimization), an AI-enabled vision-sensing detection model based on YOLOv8n for tiny strawberry disease and pest detection.
Reproduction assets foundThe paper's self-collected greenhouse strawberry disease/pest image dataset (with COCO annotations and train/test splits) is explicitly stated as publicly deposited on GitHub at the allowed URL. No author analysis code or trained model checkpoint is mentioned as publicly available.Dataset · publicThe dataset used in this study, including the training and independent test subsets, has been uploaded to a GitHub repository for dataset verification and is available at: https://github.com/dataset-review-2026/strawberry-dataset (accessed on 26 July 2026).Open asset ↗dataset-review-2026/strawberry-datasetlines:111-131Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Introduction Timely and accurate detection of plant diseases is essential for ensuring global food security and supporting sustainable agriculture. Conventional diagnostic approaches, such as manual inspection and laboratory testing, are often time-consuming, labor-intensive, and impractical for large-scale or remote agricultural environments. Although deep learning models, particularly Convolutional Neural Networks (CNNs), have significantly improved automated plant disease classification, they often lack interpretability and struggle to generalize under diverse field conditions. Methods This study proposes AgriX-SENet, an explainable deep learning framework that integrates Squeeze-and-Excitation (SE) blocks with a DenseNet121 backbone to enhance disease classification performance. The SE blocks recalibrate channel-wise feature responses to emphasize disease-relevant information while suppressing background noise. To improve model transparency, Grad-CAM, SHAP, and LIME were incorporated to provide visual and feature-level explanations of the model’s predictions. The framework was trained and evaluated using the Plant Pathology 2020 dataset containing four classes: healthy, rust, scab, and multiple diseases. Results AgriX-SENet achieved a training accuracy of 97.47% and a validation accuracy of 95.07%, outperforming fourteen state-of-the-art deep learning models. The classification report demonstrated high precision and recall across most disease categories, although the scab class exhibited comparatively lower recall, indicating an opportunity for further improvement. The explainability analyses consistently showed that the model focused on pathologically relevant regions of leaf images, validating the reliability of its predictions. Discussion The proposed AgriX-SENet framework effectively combines high classification performance with model interpretability, addressing a key limitation of existing CNN-based plant disease detection systems. Its ability to provide accurate and explainable predictions makes it a promising solution for scalable agricultural diagnostics. Future work will focus on improving classification performance for challenging disease categories and optimizing the framework for deployment on mobile and edge computing devices to enable real-time field applications.
Why it matches plant phenotyping methods葉画像から植物病害状態を推定する説明可能な深層学習フレームワークを開発・評価しており、植物表現型取得・判定手法が中心である。
abstractThis study proposes AgriX-SENet, an explainable deep learning framework that integrates Squeeze-and-Excitation (SE) blocks with a DenseNet121 backbone to enhance disease classification performance.
Reproduction assets foundThe paper trains and evaluates AgriX-SENet on the public Plant Pathology 2020 (FGVC7) Kaggle image dataset, which is the paper-specific plant image input for its disease-classification measurements. No author analysis code, trained model checkpoints, or supplementary code/data deposit is mentioned; the data statement (Dataset · publicPlant Pathology 2020 - Fgvc7 . Available online at: https://www.kaggle.com/competitions/plant-pathology-2020-fgvc7/data .Open asset ↗Kaggle · plant-pathology-2020-fgvc7lines:895-974Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Plants encounter multiple abiotic stresses. Among them, heat and drought stress play a substantial role in reducing the agricultural productivity of commercial plants. Hence, wild and underutilized plants can be a potential alternative as they are naturally tolerant to extreme climatic conditions and are a rich source of nutrition. Manual stress and disease detection is a laborious and expensive process, and hence automation in this field is required to reduce agricultural losses. This study evaluates the prediction and detection of abiotic stress in Acacia senegal bipinnate leaves, exploring various stress-induced changes using machine learning (ML) algorithms and biochemical analysis. A. senegal , an underutilized edible desert legume, was grown under controlled greenhouse conditions. After 2 months, these plants were segregated into groups and subjected to heat and drought treatments. Image acquisition was performed to obtain a dataset of 3,454 images of A. senegal leaves. Physiological parameters, such as fresh and dry leaf weight, shoot length, number of leaves, and biochemical assays like antioxidant assay (DPPH), total phenolic content (TPC), and total flavonoid content (TFC), were determined. LC-MS/MS analysis was conducted to identify over 50 phytochemical compounds. A hybrid model was developed consisting of a fine-tuned EfficientNet-based Convolutional Neural Network (CNN) followed by a Support Vector Machine (SVM) for the binary classification of A. senegal leaves. The model distinguishes between healthy and stress-affected unhealthy leaves and achieved an accuracy score of 86.6%. This report provides a significant lead toward stress phenotyping and prediction of a bipinnate leaf plant using ML algorithms. The overall study is useful to understand how the stress encountered by arid plants alters the nutritional quality.
Why it matches plant phenotyping methods画像データと機械学習モデルを用いて、アカシア葉の健全・ストレス状態を自動分類する手法を開発・評価しており、植物表現型取得が中心です。
abstractThis study evaluates the prediction and detection of abiotic stress in Acacia senegal bipinnate leaves
Reproduction assets foundThe paper's data availability statement explicitly makes the 3,454-image A. senegal leaf imaging dataset public on Zenodo and the ML implementation source code public on GitHub; both are paper-specific, public, and actionable.Dataset · publicThe plant leaf imaging data used in the work is publicly available at https://doi.org/10.5281/zenodo.16531486.Open asset ↗zenodo · 10.5281/zenodo.16531486html-lines:480-497Code · publicThe source code of the implementation is available at https://github.com/softwareinnovationslabBITS/CDRF_ASenegal_MLImagingOpen asset ↗github · softwareinnovationslabBITS/CDRF_ASenegal_MLImaginghtml-lines:480-497Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Early and accurate identification of plant diseases is essential for improving crop productivity and ensuring food security. Many existing deep learning-based plant disease classification methods rely solely on leaf images collected from a controlled environment, which limits their applicability in real-world agricultural conditions where symptoms may be visually unclear and influenced by environmental factors. To address these challenges, this study discusses AgriFusionNet, a context-aware multimodal deep learning framework that integrates leaf images, textual symptom descriptions, and environmental data for robust plant disease classification. The proposed architecture employs EfficientNet-B0 for visual feature extraction, BERT for semantic representation of symptom descriptions, and a lightweight multilayer perceptron for modeling environmental factors such as temperature, humidity, rainfall, and soil moisture. Features from all three modalities are fused into a unified representation to train the CNN model. The model is trained and tested upon the Context-Aware Multimodal Augmented PlantVillage dataset covering 38 plant diseases and healthy classes. Experimental results show that AgriFusionNet gives an overall accuracy of 98.94% on the dataset Context-Aware Multimodal Augmented PlantVillage, with competitive precision and recall and F1-score. The multimodal framework facilitates the co-learning of visual, semantic, and contextual environmental representations and the analyses of the confusion matrix and feature interactions give insights into cross-modal relationships. The proposed approach aims to explore context-aware multimodal representation learning for agricultural AI applications, with emphasis on integrating complementary visual, semantic, and contextual information.
Why it matches plant phenotyping methods葉画像を中心に、症状記述と環境情報を統合して植物病害状態を分類する手法を開発・評価しており、植物フェノタイピング手法が中心である。
abstractthis study discusses AgriFusionNet, a context-aware multimodal deep learning framework that integrates leaf images, textual symptom descriptions, and environmental data for robust plant disease classification.
Reproduction assets foundThe paper's data availability statement points to the Context-Aware Multimodal Augmented PlantVillage dataset (leaf images, symptom text, environmental data used for the phenotyping/classification analysis) deposited publicly on IEEE Dataport with a DOI matching an allowed URL.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: Dataset. IEEE Dataport. https://dx.doi.org/10.21227/9jat-r836 [Accessed on August 2025].Open asset ↗IEEE Dataport · 10.21227/9jat-r836lines:1029-1047Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Field / plotLeafClassificationDisease symptoms / severity
Early and reliable diagnosis of hibiscus leaf diseases is critical to protect horticultural yield. Yet, it remains challenging under real-time field conditions where uncontrolled lighting, clutter, and the non-contiguous nature of pathological symptoms blur diagnostic cues. To address these challenges, we introduce CNN-FusionViT-GNN. This explainable hybrid multi-branch framework synergizes the fine-grained texture extraction of a DenseNet201 backbone, the global contextual modeling of a Vision Transformer (ViT), and the relational reasoning of a Graph Neural Network (GNN). The model is trained and validated on 'Hibiscus,' a curated field dataset of 1165 images from Bangladesh, which is strategically augmented to 8000 samples for robust training following a strict train-validation-test split. The proposed framework achieves a state-of-the-art accuracy of 98.33% with a macro F1-score of 0.98. The framework's generalization is confirmed through high performance on external datasets: 98.78% accuracy on the 52-class Plant City dataset and 83.88% on the 10-class Tomato Leaf Disease dataset, while maintaining a rapid inference time of 10-45 ms. Furthermore, a multi-faceted Explainable AI (XAI) audit using LIME, Grad-CAM++, ViT Attention Maps, and Occlusion Sensitivity validates that the model's decisions are driven by biologically meaningful symptom patterns rather than background artifacts. This study establishes a computationally efficient, transparent, and robust pathway for automated disease diagnosis in precision agriculture.
Why it matches plant phenotyping methodsハイビスカス葉の病徴を画像から分類するCNN-ViT-GNN手法を開発し、複数データセットで性能検証しているため、植物病害表現型の取得・推定が中心である。
abstractwe introduce CNN-FusionViT-GNN. This explainable hybrid multi-branch framework synergizes the fine-grained texture extraction of a DenseNet201 backbone, the global contextual modeling of a Vision Transformer (ViT), and the relational reasoning of a Graph Neural Network (GNN).
Reproduction assets foundThe paper's primary Hibiscus leaf disease image dataset is publicly deposited on Mendeley Data, and the external Tomato Leaf Disease dataset used for validation is also publicly available on Mendeley Data. No author analysis code or trained model checkpoints are reported.Dataset · publicThe primary dataset generated and analyzed during the current study,“Hibiscus Leaf Diseases Classification Dataset,”is publicly available in Mendeley Data 7 .Open asset ↗Mendeley Datalines:307-347Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Premise Accurate species identification is crucial for ecological restoration and can be especially challenging for understudied non-model species. Quercus garryana is the only native oak species in the Pacific Northwest and is an important component of the endangered oak savanna ecosystem. Quercus robur is an imported ornamental species from Europe and has been found to be mistakenly planted as Q. garryana in habitat restoration projects. Methods We measured leaf morphological traits sampled from herbarium collections in their native ranges using the digital morphometric tools MorphoLeaf and Tomato Analyzer. We then used Lasso logistic analysis to generate a predictive model and tested it on leaves from Portland, Oregon. To streamline this species detection process, we developed Garryanalyzer, an ImageJ plug-in that automatically measures leaf traits and outputs species predictions. Results Garryanalyzer demonstrated 95% accuracy in predicting the species identity of herbarium specimens of oaks. Garryanalyzer correctly identified all Q. robur individuals sampled in Portland but showed lower accuracy for Q. garryana . Discussion Many existing morphometric software are not open source, which makes them unable to be customized to specific study systems. Garryanalyzer is built upon the widely used open-source ImageJ platform. This study also demonstrates a viable workflow for developing similar tools for other ecologically important non-model plant species.
Why it matches plant phenotyping methods葉の形態形質を自動測定し、種予測まで行うImageJプラグインとワークフローの開発・評価が中心であり、植物フェノタイピング手法として適格です。
abstractTo streamline this species detection process, we developed Garryanalyzer, an ImageJ plug-in that automatically measures leaf traits and outputs species predictions.
Reproduction assets foundThe paper's authors publicly released the Garryanalyzer ImageJ plug-in source code on GitHub, all original and modified leaf images used in the morphometric analyses on Zenodo, and the full leaf morphometric measurement dataset plus R Lasso analysis code in a second Zenodo repository. All are paper-specific, public,可直接Code · publicThe source code and installation instructions for Garryanalyzer can be accessed on GitHub at https://github.com/zxie8561/Garryanalyzer.Open asset ↗https://github.com/zxie8561/Garryanalyzer · zxie8561/Garryanalyzerhtml-lines:210-274Dataset · publicAll images used in the morphometric analyses, both original and modified, are available on Zenodo (https://doi.org/10.5281/zenodo.17462266).Open asset ↗https://doi.org/10.5281/zenodo.17462266 · 10.5281/zenodo.17462266html-lines:210-274Dataset · publicThe full dataset of leaf morphometric measurements of both GBIF and Portland samples, R code for Lasso analysis, and other miscellaneous files are available on a separate Zenodo repository (https://doi.org/10.5281/zenodo.17546152).Open asset ↗https://doi.org/10.5281/zenodo.17546152 · 10.5281/zenodo.17546152html-lines:210-274Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published23 Jul 2026Journal of Intelligent Decision Making and Information ScienceCited by 0 · OpenAlex ↗
The early and effective diagnosis of tomato leaf diseases is very important to enhance crop yield and reduce economic loss in precision agriculture. The conventional image-based methods are typically based on single architecture model, which cannot capture fine-grained lesion details and global contextual patterns simultaneously in the real-field. To this end, we introduce a deep hybrid Convolutional Neural Network (CNN) –Transformer architecture by combining ConvNeXt Large (ConvNeXt-L) (as local feature extractor) and Swin Transformer (as global context encoder). The concatenated features vector is then fed to a shallow classifier to predict the disease. The model was tested on two datasets, namely a field dataset in agriculture areas from Madhya Pradesh (India) and a benchmark tomato leaf dataset. Experimental results revealed that the proposed scheme achieved accuracy of 92.83% on a primary dataset, and performance was significantly high with an accuracy of up to 95.65% in terms of generalization rate for computing technique models from various environmental conditions.
Why it matches plant phenotyping methodsトマト葉の病害状態を画像から推定するCNN–Transformer手法を提案し、複数データセットで性能検証しており、植物フェノタイピング手法が中心である。
abstractwe introduce a deep hybrid Convolutional Neural Network (CNN) –Transformer architecture by combining ConvNeXt Large (ConvNeXt-L) (as local feature extractor) and Swin Transformer (as global context encoder).
Reproduction assets foundThe paper uses a public Tomato Leaves Dataset from GTS AI as its secondary/external validation dataset for tomato leaf disease classification. The primary field dataset from Madhya Pradesh is not stated as publicly available, and no author analysis code or trained model is reported as deposited.Dataset · publicSecondary Dataset: The Secondary dataset was extracted from the public Tomato Leaves Dataset
available at GTS AI platform. It involves various disease classes, such as bacterial spot, early blight,
late blight, leaf mold, powdery mildew, septoria leaf spot and spider mites (Figure 1) target spots are
present in tomato mosaic virus leaves yellow curl virus of tomato. This data set was employed as an
external validation to evaluate the generalization of proposed model in different conditions and
diseases types.
Source : https://gts.ai/dataset-download/tomato-leaves-dataset/Open asset ↗GTS AIpdf-page:20 lines:1-23Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published23 Jul 2026Journal of Intelligent Decision Making and Information ScienceCited by 0 · OpenAlex ↗
- Enhancing agricultural productivity and attaining sustainable crop management depend on the early and precise identification of leaf disease. Using state-of-the-art technologies in precision agriculture like machine learning (ML) and image processing greatly increases the effectiveness of disease detection and facilitates well-informed decision-making. But conventional manual inspection techniques are still tedious, unpredictable, and prone to errors. In order to overcome these constraints, this research offers YOLOv12-CropNet, an innovative deep learning-based system for multi-crop leaf disease diagnosis in real time. The proposed YOLOv12-CropNet approach makes use of the Convolutional Block Attention Module (CBAM) for adaptive attention, the Content-Aware Reassembly of Features (CARAFE) up-sampling module to preserve fine-grained disease characteristics, the YOLOv12 architecture improved with Ghost Convolution for effective feature extraction, and Involution layers to capture spatially specific patterns. Inspection techniques are still laborious, arbitrary, and prone to mistakes. A substantial set of data of 38 classes of both healthy and sick leaves from a variety of crops, including tomato, potato, apple, grape, corn, mango and sugarcane, was put together for training and evaluation. Experimental results show that YOLOv12-CropNet finds a suitable balance between computational speed and accurate detection. Accuracy, F1-score, recall, and precision are important performance metrics that verify the model's resilience in challenging environmental and visual circumstances. The suggested technique provides a scalable and field-deployable way to assist effective identification of diseases and precision agricultural decision-making. The proposed YOLOv12-CropNet model exhibits better performance than the other evaluated models, attaining a 98.45% peak accuracy, 98.10% precision ,98.20 % sensitivity and a 98.18% F1 score, thereby highlighting its efficacy in multi-crop leaf disease detection.
Why it matches plant phenotyping methods複数作物の葉の病徴を画像から検出・分類する深層学習手法を開発し、データセットと性能評価を伴うため、植物病害状態のフェノタイピング手法が中心である。
abstractthis research offers YOLOv12-CropNet, an innovative deep learning-based system for multi-crop leaf disease diagnosis in real time.
Reproduction assets foundThe paper uses public Kaggle datasets as its phenotyping image inputs: the PlantVillage dataset (38 crop-disease classes) and the Sugarcane Leaf Disease dataset, both cited with explicit public URLs. No author code, models, or checkpoints are reported as publicly available.Dataset · public[37] PlantVillage Dataset. Available online: https://www.kaggle.com/datasets/abdallahalidev/plantvillage-datasetOpen asset ↗pdf-page:22 lines:1-61Dataset · public[39] Sugarcane leaf Disease Dataset available online: https://www.kaggle.com/datasets/nirmalsankalana/sugarcane-Open asset ↗pdf-page:22 lines:1-61Code / dataset availability confirmedEurope PMC · Crossref · checked 14 Sept 2026
Abstract Blanket-rate agrochemical scheduling — a practice wherein the same quantity of fertilizer or pesticide is spread uniformly across an entire field irrespective of spatial or temporal crop need — persists as the dominant farm management paradigm across rural India and large parts of South Asia. This approach generates cascading inefficiencies: excess nitrogen drains into waterways, off-target pesticide deposits devastate pollinators, input costs erode thin profit margins, and wide-scale greenhouse gas release from soil microbial activity accelerates climate change. The study documented here addresses this challenge through a purpose-built, four-layer intelligent field management platform. The platform ingests continuous data from drone-mounted multispectral cameras, in-field IoT soil probes, a wireless weather station, and cloud-sourced Sentinel-2 satellite imagery, then passes these inputs through a cascaded AI inference stack. A fine-tuned YOLOv8-L network performs real-time pest and foliar disease localisation; a ResNet-50 backbone quantifies canopy health across five stress gradients; a two-layer stacked LSTM projects short-horizon yield trajectories; and a Deep Q-Network autonomously plans drone spray routes weighted by field-specific prescription maps. Field validation spanned two consecutive growing seasons (Rabi 2022–23 and Kharif 2023–24) across six georeferenced plots covering 4.8 ha at Baramati, Maharashtra. Outcome metrics recorded during head-to-head comparison with conventional practice included a disease detection score of 95.6% mAP, a 47.3% reduction in total nitrogen applied, a 38.1% decrease in pesticide volume, and a 22.4% uplift in harvested grain weight. Together, these field-verified numbers substantiate the operational readiness of integrated AI precision agriculture for smallholder deployment.
Why it matches plant phenotyping methodsマルチスペクトル画像・深層学習による病害局在化とキャノピー健康状態の定量化を中核機能とする統合プラットフォームであり、植物の病害状態・生育状態を直接推定して現地検証している。
abstractThe platform ingests continuous data from drone-mounted multispectral cameras, in-field IoT soil probes, a wireless weather station, and cloud-sourced Sentinel-2 satellite imagery, then passes these inputs through a cascaded AI inference stack.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicData Availability The annotated image dataset (14,300 images, 23 classes), trained YOLOv8-L and ResNet-50 weights, LSTM model files, DQN policy checkpoint, and all analysis scripts are archived at https://github.com/precision-agri-ai (Zenodo DOI: 10.5281/zenodo.XXXXXXX).Open asset ↗precision-agri-ailines:161-182Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Plant diseases significantly affect agricultural productivity and global food security, while accurate disease identification remains challenging because of uncertain and overlapping visual symptoms in leaf images. Existing deep learning approaches often require large annotated datasets and suffer from limited interpretability in practical agricultural environments. This study presents a multi-parameter improved fuzzy soft set-based framework for plant disease classification using tomato leaf images from the PlantVillage dataset. The objective is to develop an interpretable and reliable classification model capable of handling uncertainty in plant disease patterns through feature-driven fuzzy similarity analysis. The methodology integrates image preprocessing, color and texture feature extraction, variance-based feature weighting, prototype generation using K-means clustering, and fuzzy similarity computation using Mahalanobis distance and Gaussian membership functions. RGB, HSV, and Gray-Level Co-occurrence Matrix (GLCM) features are extracted from standardized leaf images and evaluated within an improved fuzzy soft classification framework. Performance comparison is carried out using machine learning models including Support Vector Machine (SVM), Random Forest (RF), Linear Discriminant Analysis (LDA), and Naive Bayes (NB) implemented in Python using Scikit-learn libraries. Experimental simulation results demonstrate that the proposed framework achieves competitive classification performance while preserving interpretability and robustness under uncertain feature distributions. Performance evaluation is conducted through accuracy analysis, ROC-AUC curves, confusion matrices, ablation studies, and Wilcoxon Signed-Rank statistical testing. The proposed Improved Fuzzy Soft model achieved an accuracy of 88.57% which is less than LDA (94.92%), Random Forest (97.78%) and SVM (97.94%) classifiers. However, in the cross data set validation, the proposed Improved Fuzzy Soft model achieved an accuracy of 67.35% which is greater than LDA (51.02%), Random Forest (51.02%) and SVM (55.10%) classifiers. Statistical validation using the Wilcoxon Signed-Rank Test produced a p-value of [Formula: see text], confirming that the performance difference between the Improved Fuzzy Soft framework and the Random Forest classifier is statistically significant under the current experimental setting.
Why it matches plant phenotyping methodsトマト葉画像から植物病害状態を推定する解釈可能な画像解析・分類フレームワークを開発し、複数モデル、交差データセット検証、アブレーション、統計検定で評価しており、病害表現型の取得・抽出手法が中心である。
abstractThis study presents a multi-parameter improved fuzzy soft set-based framework for plant disease classification using tomato leaf images from the PlantVillage dataset.
Reproduction assets foundThe paper uses public tomato leaf image datasets (PlantVillage and PlantDoc from Kaggle) as phenotyping inputs and states the authors' Improved Fuzzy Soft Framework implementation is publicly available on Zenodo with source code and reproduction instructions.Dataset · publicThe dataset analyzed during the current study are available in the repository:
https://www.kaggle.com/datasets/abdallahalidev/plantvillage-datasetOpen asset ↗kaggle.com/datasets/abdallahalidev/plantvillage-datasetpdf-page:24 lines:1-75Code · publicThe implementation of the proposed Improved Fuzzy Soft Framework is publicly available through the Zenodo repository:
https://doi.org/10.5281/zenodo.20570546
The repository contains the source code, documentation, and instructions required to reproduce the experiments reported in
this study.Open asset ↗zenodo · 10.5281/zenodo.20570546pdf-page:25 lines:1-74Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Plant disease detection under complex climatic conditions and cross-crop scenarios remains a critical challenge. To address this, we propose a novel SimAM-Inception-StyleRandomization (SIS)-YOLOv11 algorithm based on YOLOv11n for early/late blight detection on potato and tomato leaves. Our core innovations are: 1) A C3k2-SSI module integrating Style Randomization, Inception architecture, and SimAM attention to enhance cross-crop generalization; 2) A Fusion-InceptionConv module for fine-grained feature extraction under rainfall/haze noise; 3) SPPF-Inception and C2PSA-IS modules to optimize multi-scale feature fusion; 4) DepGraph pruning to reduce 47.82% parameters while improving performance. Experiments show that the pruned SIS-YOLOv11 outperforms YOLOv11n by 3.7% in precision, 6.6% in recall, 5.4% in mAP50, and 7.9% in mAP50-95, and surpasses mainstream models (Faster R-CNN, SSD, etc.). This study provides a robust, lightweight solution for automated cross-crop disease detection in complex agricultural environments.
Why it matches plant phenotyping methodsジャガイモとトマト葉の病害状態を画像から検出する新規アルゴリズムを開発し、性能比較・軽量化まで行っており、植物フェノタイピング手法が中心である。
abstractwe propose a novel SimAM-Inception-StyleRandomization (SIS)-YOLOv11 algorithm based on YOLOv11n for early/late blight detection on potato and tomato leaves.
Reproduction assets foundThe paper's image dataset (potato/tomato leaf disease images with annotations and climate-noise augmentation) is explicitly declared publicly available on Baidu AI Studio. No author code or trained model deposit is stated.Dataset · publicData Availability: All image datasets used and analyzed in this study are publicly available from the Baidu AI Studio dataset repository at the URL: https://aistudio.baidu.com/datasetdetail/245434 .Open asset ↗Baidu AI Studio · 245434lines:1-133Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Detecting cotton leaf diseases in open-field environments is challenging due to cluttered backgrounds, scale variation, and irregular lesion morphology. Conventional detectors rely on isotropic receptive fields and coupled box-regression losses, which limit their ability to localize elongated lesions with poorly defined boundaries. We present an anisotropic boundary-aware detection framework that propagates high-frequency boundary information across four successive pipeline stages. In the backbone, an Anisotropic Morphological Contrast Aggregation module (AMCA) enhances direction-aware representation and lesion-background contrast via re-parameterizable strip convolutions and high-frequency residual extraction. A Dynamic Semantic Boundary Transfer mechanism (DSBT) then captures boundary priors from shallow layers before they are lost to downsampling and injects them into the neck. A Morphological-Spectral Synergistic Feature Pyramid Network (MFS-FPN) preserves these cues during multi-scale fusion through spatial-domain operations compatible with edge hardware. Finally, an Anisotropic Boundary-Decoupled IoU loss (ABD-IoU) independently penalizes each of the four box boundaries and sustains optimization signals in high-IoU regimes via a logarithmic modulation factor. On the self-constructed Complex Cotton Leaf Disease dataset (CCLD; 6,856 images, 6 classes), the method achieves 78.50% mAP@50 and 65.00% mAP@50:95, improving the YOLOv11n baseline by 4.80% and 2.70% with only 2.73 M parameters at 202 FPS. Cross-domain evaluations on PlantDoc and RWD confirm consistent improvements. The framework runs in real time on NVIDIA Jetson edge platforms with INT8 quantization.
Why it matches plant phenotyping methods綿花葉の病斑・病害状態を画像から検出する手法を開発し、複数データセットとベースラインで性能検証しているため、植物表現型取得が中心である。
abstractWe present an anisotropic boundary-aware detection framework that propagates high-frequency boundary information across four successive pipeline stages.
Reproduction assets foundThe authors explicitly state that their source code, trained models, and implementation details are publicly available, and the data availability statement points to the same repository, which hosts the self-constructed CCLD cotton leaf disease dataset (6,856 images, 6 classes) used for the paper's phenotyping/disease-Code · publicFurthermore, to facilitate future research, our source code, trained models, and implementation details have been made publicly available at https://github.com/DynaVLA/ABAD-CLD .Open asset ↗DynaVLA/ABAD-CLDlines:331-343Dataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://github.com/DynaVLA/ABAD-CLD .Open asset ↗DynaVLA/ABAD-CLDlines:1278-1317Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
To address this challenge, we propose StyleGAN3-T, the translation-equivariant alias-free variant of StyleGAN3, as a generative framework for producing high-fidelity synthetic plant disease images, integrated with a hybrid Swin Transformer-ResNet50 classifier for precise recognition. Accurate detection of plant leaf diseases is essential for sustainable agriculture and early intervention. However, deep learning models often struggle with small, imbalanced datasets that limit generalization and robustness. To address this challenge, we propose StyleGAN3-T, a novel alias-free generative framework for producing high-fidelity synthetic plant disease images, integrated with a hybrid Swin Transformer-ResNet50 classifier for precise recognition. The proposed approach ensures translation-equivariant, artifact-free image synthesis and enhanced feature diversity. A balanced dataset of 18,000 images was developed by combining real and StyleGAN3-T-generated samples. In pooled GAN benchmarking, StyleGAN2-ADA achieved the strongest generative-quality metrics, whereas StyleGAN3-T was selected as the preferred augmentation model because its alias-free synthesis and spatial consistency yielded superior downstream classification performance in the proposed pipeline.
Why it matches plant phenotyping methods植物病害画像を合成・認識する画像解析手法が研究の中心であり、植物の病害状態を画像から推定するフェノタイピング手法に該当する。
abstractwe propose StyleGAN3-T, a novel alias-free generative framework for producing high-fidelity synthetic plant disease images, integrated with a hybrid Swin Transformer-ResNet50 classifier for precise recognition.
Reproduction assets foundThe paper's grape leaf disease image inputs are two publicly available Kaggle datasets explicitly named in the Data Availability statement. No author code, models, or synthetic dataset deposit is provided; other processed data is request-only.Dataset · publictechnical guidance. Y.L. and A.W. supervised the study, provided critical revisions, and contributed to the interpretation of results. All authors reviewed and approved the final manuscript.
Data availability
The datasets analyzed during the current study are publicly available from Kaggle: Grapevine Disease Dataset (Original) (https://www.kaggle.com/datasets/rm1000/grape-disease-dataset-original; accessed 13 March 2026; license: MIT) and Grape Leaf Disease 4 Class (https://www.kaggle.com/datasets/jawadulkarim117/grape-leaf-disease-4-class; accessed 13 March 2026; license: CC0: Public Domain). Additional processed metadata, label-harmonization records, dataset split definitions, and other daOpen asset ↗Kaggle · rm1000/grape-disease-dataset-originallines:549-576Dataset · publicthors reviewed and approved the final manuscript.
Data availability
The datasets analyzed during the current study are publicly available from Kaggle: Grapevine Disease Dataset (Original) (https://www.kaggle.com/datasets/rm1000/grape-disease-dataset-original; accessed 13 March 2026; license: MIT) and Grape Leaf Disease 4 Class (https://www.kaggle.com/datasets/jawadulkarim117/grape-leaf-disease-4-class; accessed 13 March 2026; license: CC0: Public Domain). Additional processed metadata, label-harmonization records, dataset split definitions, and other data used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Declarations
Competing inOpen asset ↗Kaggle · jawadulkarim117/grape-leaf-disease-4-classlines:549-576Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
The genetic identity of coffee cultivars is fundamental to the specialty coffee sector, where premium prices are paid under the assumption that the purchased planting material corresponds to the declared variety. However, many producing countries lack the certification infrastructure necessary to guarantee this identity in their informal seed systems, exposing producers to undetected varietal non-conformity. In this study, we examine a case from a specialty coffee ( Coffea arabica L.) farm in southern Ecuador where seeds labeled as Sidra (USD 100/kg) and Gesha (USD 500/kg) were purchased without genetic or phytosanitary certification. Using a combination of SSR-based DNA fingerprinting and quantitative morphological characterization, including plant architecture, leaf functional traits, and fruit characteristics, we documented varietal identity and assessed the discriminant capacity of morphological traits across the four resulting morphotypes. Using eleven microsatellite markers for SSR fingerprinting, we found that two of the four morphotypes did not match their declared commercial identity. One plant sold as Sidra was identified as compatible with Batian, a composite variety of Kenyan origin that is genetically unrelated to Ethiopian landraces. The plants acquired as Gesha corresponded to a pure Ethiopian landrace that is genetically similar to, but not identical to, the Panamanian Geisha reference accession T.02722. Only two morphotypes were confirmed as Sidra. Furthermore, the placement of Sidra within the Core Ethiopia genetic group is consistent with prior population-level analyses and with its likely status as a selected Ethiopian landrace rather than a variety of hybrid origin. Morphological linear discriminant analysis achieved 82.4% overall classification accuracy under leave-one-out cross-validation (LOOCV), with internode length dominating the first discriminant function (LD1 = 66.6%). These results demonstrate that varietal nonconformity in the specialty coffee seed sector can extend to the inadvertent introduction of genetically unrelated material and underscore the urgent need for accessible seed certification.
Why it matches plant phenotyping methodsコーヒー品種識別のための形態形質測定と判別分析が研究の中心であり、形態形質の識別性能をLOOCVで検証しているため、植物フェノタイピング手法の適用・検証に該当する。
abstractquantitative morphological characterization, including plant architecture, leaf functional traits, and fruit characteristics
Reproduction assets foundThe paper's morphological/functional trait dataset (used for the phenotyping and LDA analysis) is explicitly stated to be publicly available on Figshare (10.6084/m9.figshare.32841344). No author analysis code repository is stated; other URLs in the text are generic libraries or cited prior work.Dataset · publicThe morphological and functional trait dataset generated and analyzed in this study is publicly available in the Figshare repository at 10.6084/m9.figshare.32841344 .Open asset ↗Figshare · 10.6084/m9.figshare.32841344lines:526-568Code / dataset availability confirmedOpenAlex · Europe PMC · bioRxiv · checked 5 Sept 2026
ABSTRACT While whole-genome sequencing captures millions of single nucleotide polymorphisms (SNPs) and hyperspectral imaging (HSI) enables non-destructive plant phenotyping, integrating these modalities to link genotype to phenotype remains challenging due to their high dimensionality and non-linearity. This study presents DeepPheno a deep learning framework that predicts SNP genotypes from HSI data, using model predictability as a proxy for genotype-phenotype association. HSI data were acquired from 194 lettuce genotypes under field conditions. HSI data patches (20×20 pixels × 224 spectral bands) were used to train a hybrid CNN to predict the variant of a specific SNP. The framework was validated on SNPs with known phenotypic effects (anthocyanin, leaf serration, pale pigmentation), achieving high predictive performance (AUC ranging from 0.806 to 0.935), whereas models trained on randomly shuffled labels performed at chance (mean AUC ≈ 0.51). Extending the workflow to 50 randomly selected putatively neutral SNPs, most yielded low predictability, but two showed high performance (AUC > 0.76), suggesting uncharacterized genotype-phenotype links. Explainable AI, including SHAP and Grad-CAM, identified relevant spectral and spatial features driving these predictions, particularly the green and red-edge wavelengths associated with pigment dynamics and leaf structure. These results establish a framework for understanding complex genotype-phenotype interactions in plants and extracting these links from HSI data without predefining the exact trait values. It provides an avenue for high-throughput trait discovery and description and extends the integration of image-based phenomics with plant genetics.
Why it matches plant phenotyping methodsHSIと深層学習を統合し、遺伝子型関連の植物表現型情報を抽出する枠組みを開発・検証しており、フェノタイピング手法が研究の中心です。
abstractThis study presents DeepPheno a deep learning framework that predicts SNP genotypes from HSI data, using model predictability as a proxy for genotype-phenotype association.
Reproduction assets foundThe paper's Data Availability statement deposits authors' code, scripts, and supplementary material in a public GitHub repository, including a downscaled de-identified sample dataset demonstrating the pipeline. The raw HSI/genotype datasets are proprietary under NDA and not public.Code · publicThe code, scripts, and supplementary material supporting the findings of this study have been deposited in the GitHub repository at https://github.com/frankgyan/Utrecht-University--HSI .Open asset ↗frankgyan/Utrecht-University--HSIlines:195-223Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Plant diseases threaten global agriculture, causing 20–40% yield losses and food insecurity. Current diagnostic methods are costly and lack scalability. While deep learning advances plant disease detection, there remains a need for CNNs with simpler architectures, better generalizability, and lower computational cost. This study presents a novel CNN for multi-class classification of 38 diseases. Trained on a public dataset of over 87,000 RGB images, the architecture comprises five convolutional blocks (filters 32–512) with max pooling and dropout (0.25, 0.4), followed by a 1,500-unit dense layer and SoftMax output. Optimized with Adam (lr=0.0001) and categorical cross-entropy, the model achieved 98% training and 96% validation accuracy with approximately 28.7 million parameters significantly fewer than transfer learning architectures. These results demonstrate an effective balance between predictive performance and computational efficiency, positioning the model as a promising tool for real-world agricultural deployment.
Why it matches plant phenotyping methods葉画像から植物病害を分類するCNNの開発と性能評価が研究の中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法に該当します。
titleLeaf-by-Leaf Diagnosis: A Custom CNN with Pyramidal Feature Extraction for Plant Disease Classification
Reproduction assets foundThe paper's sole qualifying asset is the plant disease image dataset used for all its CNN training/validation measurements: the publicly available New Plant Diseases Dataset (Augmented) on Kaggle, explicitly declared in the Data availability statement. No author code, trained model checkpoints, or other paper-specific Dataset · publict to disclose.
Acknowledgment:
The authors are sincerely grateful to their institutions for their continued support and trust, which greatly contributed to
the completion of this research.
Data availability
The dataset used and analyzed during the current study, “New Plant Diseases Dataset”, is publicly available on Kaggle at:
(https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset?select=New+Plant+Diseases+Dataset%28Augmented%29)
References
[1] P. Arputharaj and K. Karunanithy, "A review on machine learning and deep learning techniques for plant leaf disease
detection and classification with IoT in agriculture industry," Journal of Industrial Information Integration, vol. 50, Open asset ↗Kaggle · vipoooool/new-plant-diseases-datasetpdf-raw-page:11 lines:1-49Dataset · publict to disclose.
Acknowledgment:
The authors are sincerely grateful to their institutions for their continued support and trust, which greatly contributed to
the completion of this research.
Data availability
The dataset used and analyzed during the current study, “New Plant Diseases Dataset”, is publicly available on Kaggle at:
(https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset?select=New+Plant+Diseases+Dataset%28Augmented%29)
References
[1] P. Arputharaj and K. Karunanithy, "A review on machine learning and deep learning techniques for plant leaf disease
detection and classification with IoT in agriculture industry," Journal of Industrial Information Integration, vol. 50, Open asset ↗Kaggle · vipoooool/new-plant-diseases-datasetpdf-raw-page:11 lines:1-49Code / dataset availability confirmedEurope PMC · bioRxiv · checked 15 Sept 2026
Accurate segmentation of plant disease symptoms is essential for crop monitoring and phenotyping, yet it typically requires costly pixel-level annotations. Weakly supervised semantic segmentation (WSSS) alleviates this burden using image-level labels, but its performance depends on the quality of spatial priors such as class activation maps (CAMs). We investigate whether text-guided segmentation with the Segment Anything Model 3 (SAM3) can serve as an alternative weak supervision signal. Three pseudo-mask generation strategies are compared: (i) CAMs refined with SAM or SAM3, (ii) zero-shot text-guided SAM3, and (iii) a hybrid approach combining weak spatial cues with text prompts. The resulting pseudo-masks are used to train a DeepLabV3 model. Text guidance alone matches or outperforms conventional WSSS, achieving up to 0.46 IoU without spatial supervision and 0.61 IoU on a public dataset, although performance is sensitive to text prompt formulation. The hybrid strategy improves robustness, reaching 0.50 IoU on the primary dataset and 0.58 IoU on the additional dataset while reducing prompt sensitivity. Overall, text guidance is a promising alternative to conventional weak supervision, while hybrid approaches provide a more robust solution for plant disease segmentation.
Why it matches plant phenotyping methods植物病害症状の画像セグメンテーション手法を開発・比較し、病斑の疑似マスク生成とセグメンテーション性能を評価しているため、植物フェノタイピング手法が中心です。
abstractAccurate segmentation of plant disease symptoms is essential for crop monitoring and phenotyping
Reproduction assets foundThe paper's own oilseed rape leaf disease image dataset (2,540 RGB images, 601 pixel-level annotations) is explicitly made publicly available on Recherche Data Gouv under DOI 10.57745/R3HZOQ, which is an allowed URL. A companion dataset describing the foliar pathogens (DOI 10.57745/UCEKI8) is also public. The pixel-annDataset · publicAn original dataset of 2,540 RGB images of oilseed rape leaves affected by
7 fungal and bacterial diseases (Table 1) was built to develop and evaluate
the proposed pipelines, and is made publicly available [25].Open asset ↗pdf-raw-page:5 lines:1-43Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Agricultural leaf disease detection is crucial for early intervention and yield protection in precision agriculture. Among representative economic crops, such as apples, leaf lesions are typically small and appear in complex backgrounds, making accurate detection performed on resource-constrained embedded devices challenging. To address this, we propose a lightweight small-object detection models, namely the dynamic Differential Compensation Lightweight-YOLO (DCL-YOLO) model and its pruned version (DCL-YOLO-P), based on YOLO11n. A novel Dual-Aspect Feature Complementary Mapping (DAFCM) module type is embedded in their backbone to recover lost semantic and spatial information, while the original YOLO11n's neck is replaced by an Efficient Enhanced Cross-Scale Feature Fusion (EE-CSFF) module, which incorporates Gated Differential Convolutional Fusion (GDCF) modules to strengthen cross-scale information flow and small-object representation. Experimental results obtained on the ALDSOD dataset show that, compared with the YOLO11n baseline, DCL-YOLO improves recall from 81.9% to 84.6%, mAP50 from 86.8% to 88.4%, and mAP50:95 from 47.0% to 47.8%, while also reducing the parameter count from 2.58 M to 1.91 M and Giga Floating-Point Operations (GFLOPs) from 6.3 to 5.5. After applying Layer-Adaptive Magnitude-based Pruning (LAMP), the parameter count and GFLOPs are further reduced to 0.75 M and 2.7, respectively, with mAP50 and mAP50:95 still exceeding the baseline by 1.2 and 0.5 percentage points, respectively. When deployed on an embedded device, the pruned model achieved 15.2 FPS and 139 msec per image, confirming its applicability in real-time scenarios. Furthermore, cross-domain validation, performed on the Global Wheat Head Detection (GWHD) dataset, indicates the stable generalization capabilities of the proposed models across environmental domain shifts. The DCL-YOLO's source code is publicly available at: https://github.com/q123-code/dcl-yolo.
Why it matches plant phenotyping methodsリンゴ葉の病斑を画像から検出する軽量モデルを開発し、データセットで性能比較・クロスドメイン検証・組込み機器での実装評価を行っており、植物の病害状態推定手法が研究の中心です。
abstractTo address this, we propose a lightweight small-object detection models, namely the dynamic Differential Compensation Lightweight-YOLO (DCL-YOLO) model and its pruned version (DCL-YOLO-P), based on YOLO11n.
Reproduction assets foundThe paper's constructed ALDSOD apple-leaf disease detection dataset is publicly available via Zenodo DOI, and the authors' DCL-YOLO source code is publicly available on GitHub. Both are paper-specific, public, and actionable.Dataset · publicData Availability: The constructed ALDSOD dataset used in this study is available for download from the following DOI: https://doi.org/10.5281/zenodo.17198053 .Open asset ↗zenodo · 10.5281/zenodo.17198053lines:148-159Code · publicThe DCL-YOLO’s source code is publicly available at: https://github.com/q123-code/dcl-yolo .Open asset ↗github · q123-code/dcl-yololines:148-159Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
Abstract Plant diseases affecting leaves and fruits cause substantial yield and economic losses worldwide, particularly in horticultural crops cultivated under diverse agro-climatic conditions. Early and accurate disease diagnosis is essential for effective crop management; however, manual inspection is time-consuming, subjective, and often infeasible at large scale. In this work, we present the Tomato–Chilli–Papaya (TCP) Fruit and Leaf Disease Dataset, a comprehensive multi-crop image dataset designed to support deep learning-based plant disease recognition. The dataset comprises labeled RGB images of healthy and diseased leaves and fruits from three economically important crops—tomato, chilli, and papaya—captured under real-field and semi-controlled environments, reflecting significant variability in illumination, background complexity, and disease severity. To demonstrate the applicability of the dataset, several commonly used convolutional neural network (CNN) architectures, including VGG, ResNet, DenseNet, MobileNet, and EfficientNet models, were trained and evaluated on the TCP dataset using transfer learning. Experimental results show that deep CNN models can effectively learn discriminative visual features corresponding to disease-specific patterns such as leaf spots, lesions, discoloration, curling, and fruit surface abnormalities. Lightweight models such as MobileNet achieve competitive performance with reduced computational cost, while deeper architectures provide improved accuracy at the expense of higher complexity. The results highlight the importance of dataset diversity for robust model generalization across multiple crops and plant organs. The TCP dataset provides a challenging benchmark for single-crop and multi-crop disease classification and supports the development of advanced deep learning, attention-based, and explainable AI models for precision agriculture. By enabling reproducible research and realistic performance evaluation, this dataset contributes toward scalable and practical AI-driven plant disease diagnosis systems aimed at reducing yield losses and supporting sustainable agriculture.
Why it matches plant phenotyping methods植物の葉・果実の病徴を画像から評価する大規模データセットとベンチマークを中心に扱っており、植物病害状態の画像ベース表現型解析に該当する。
abstractwe present the Tomato–Chilli–Papaya (TCP) Fruit and Leaf Disease Dataset, a comprehensive multi-crop image dataset designed to support deep learning-based plant disease recognition.
Reproduction assets foundThe paper introduces the TCP (Tomato-Chilli-Papaya) fruit and leaf disease image dataset and reports CNN experiments on it. The dataset is publicly deposited on Mendeley Data, and the authors state that analysis code is available on GitHub. Both are paper-specific, public, and actionable.Dataset · publicData is available on Mendeley:1Open asset ↗pdf-page:27 lines:1-51Code · publicCode availability: Code is available on GitHub 2Open asset ↗GitHubpdf-page:27 lines:1-51Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Accurate lesion segmentation is essential for automated plant disease analysis in precision agriculture. Although the Segment Anything Model (SAM) exhibits strong generalization ability, its direct application to plant disease images in natural field environments remains challenging due to cluttered backgrounds, dense leaf veins, uneven illumination, and frequent occlusions. In particular, SAM mainly relies on global structural cues and is often insufficiently sensitive to subtle lesion textures and weak local details, which can result in missed small or early-stage lesions and inaccurate boundary delineation. To address these limitations, we enhance SAM with a disease-specific detail compensation module for plant disease lesion segmentation. A ResNet-50based branch is employed to extract fine-grained local texture features that are difficult for SAM to capture. These fine-grained features are fused with SAM encoder representations and then injected into the SAM decoder, enabling more accurate lesion prediction while preserving SAM’s strong global modeling capability. More importantly, we propose a reliability-guided variational fusion framework to further improve the interaction between heterogeneous features. Specifically, instead of conventional similarity or addition-based fusion, we introduce an uncertainty-aware variational fusion strategy that explicitly quantifies the confidence of each feature stream. An uncertainty encoder models feature distributions probabilistically, and a variational fusion module dynamically assigns higher weights to more reliable features while suppressing uncertain or interfering responses. In addition, Kullback-Leibler divergence regularization is introduced to stabilize cross-feature alignment and improve fusion robustness. Extensive experiments on PlantSeg, PlantDoc-Seg, and ATLDSD demonstrate that the proposed method outperforms state-of-theart approaches, achieving DSC scores of 81.05%, 91.12%, and 88.27%, respectively. The proposed method addresses SAM’s weakness in fine-grained disease feature extraction, accurately identifies early and small lesions, and delivers reliable segmentation for field plant disease automatic diagnosis.
Why it matches plant phenotyping methods植物病斑を対象とする画像セグメンテーション手法を開発し、複数データセットで性能検証しているため、病害状態のフェノタイピング手法が中心である。
abstractwe enhance SAM with a disease-specific detail compensation module for plant disease lesion segmentation.
Reproduction assets foundThe paper evaluates ReLeaf-SAM on three public plant disease segmentation datasets. One of them, PlantDoc-Seg, is explicitly a community-provided Kaggle dataset with a verbatim URL matching an allowed URL; it is a public plant image/mask dataset directly used for this paper's segmentation measurements. PlantSeg and ATLDataset · publicTherefore, we used a community-provided segmentation subset from Kaggle 1 , which we refer to as PlantDoc-Seg in this study. This subset is derived from PlantDoc and contains 588 diseased leaf images with corresponding binary masks, enabling supervised leaf disease segmentation.Open asset ↗Kagglelines:48-115Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Agriculture is significant in world food production and global economic stability, but leaf disease and pest infection can cause a threat to crop quantity and quality. Thus, it became crucial to have timely and accurate identification of plant leaf disease to prevent loss in agriculture on a large scale and to have sustainable crop management. This paper introduces LeafLiteX, a lightweight mobile-based deep learning application for real-time detection and classification of leaf diseases. This application uses U-Net segmentation to precisely find leaf regions and MobileNetV3-Large to quickly classify diseases with less computation on the computer. The application performs end-to-end processing, from image acquisition to segmentation and disease prediction on mobile devices. An experiment was performed on publicly available crop disease datasets containing various leaf images having different disease types. The model obtained an accuracy of 98.85% showing improved generalization with minimal latency. The design of the model was such that it was suitable for inference on-device while still being robust enough despite changes in lighting conditions, background noise, and camera resolution. LeafLiteX is a low-cost, easy to use, offline-capable, and in-the-moment decision-making supportive diagnostic application that supports farmers and agrarians who require early detection. This paper demonstrates the capabilities that can be achieved using edge-optimized machine learning and computer vision to support the development of smart agriculture technologies. While traditional methods rely solely on classification, this research focuses more on practical implementation by incorporating segmentation, lightweight classification, and explainability to develop a mobile-friendly model.
Why it matches plant phenotyping methods葉画像から病害状態をセグメンテーション・分類する手法とモバイルアプリ自体が研究の中心であり、植物病害の表現型推定に該当する。
abstractThis paper introduces LeafLiteX, a lightweight mobile-based deep learning application for real-time detection and classification of leaf diseases.
Reproduction assets foundThe paper's Data availability statement explicitly links the public PlantVillage (Mendeley) and PlantDoc (GitHub) leaf-image datasets used for its experiments, and provides the authors' LeafLiteX source code on GitHub.Dataset · publicThe dataset used in this study is publicly available from the repository:
https://data.mendeley.com/datasets/tywbtsjrjv/1Open asset ↗data.mendeley.com · tywbtsjrjv/1pdf-page:29 lines:1-74Code · publicThe source code is available on the following link: https://github.com/phdpawan/LeafLiteX.Open asset ↗github.com/phdpawan/LeafLiteXpdf-page:29 lines:1-74Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
Abstract Background Stomata and pavement cells are fundamental components of the leaf epidermis, jointly regulating gas exchange, water loss, and leaf surface expansion. Stomata size, aperture, density, and pavement cell morphology are critical parameters for assessing plant transpiration efficiency, epidermal growth dynamics, and adaptive responses to environmental constraints. Despite their biological importance, quantifying stomatal and pavement-cell traits remains seldom not generalized, simple, and fast enough . Manual or semi-automated approaches limit large-scale phenotyping and restrict the integration of epidermal morphology into crop-improvement pipelines aimed at developing climate-resilient varieties with optimised stomatal patterning. To address such limitations, we developed Stomatalia , a deep learning-based platform designed to automate and standardise the quantification of stomatal and pavement cell traits. The algorithm was trained on epidermal images of cultivated and wild potato and tomato genotypes grown under optimal and abiotic-stress conditions. Stomatalia automatically detects stomata and pavement cells and extracts a broad range of morphological and integrative epidermal parameters, enabling high-throughput phenotyping within a unified workflow. Results Prior to platform development, we optimised a rapid, minimally-destructive leaf-printing protocol that generates negative impressions of the leaf surface within 40–100 s. Transparent positive prints were subsequently produced and imaged under a light microscope at scale settings ranging from 20 to 200 μm. The resulting images are analysed using Stomatalia’s user-friendly web-based interface, which runs an instance-segmentation deep learning algorithm to detect, count, and calculate stomatal and pavement cell parameters. The platform outputs structured files containing raw measurements, derived integrative traits, and associated metadata, facilitating downstream statistical and physiological analyses. Algorithm evaluation on independent datasets demonstrated high performance within the validated dicot imaging domain, with F 1 -scores ranging from 0.86 to 0.94 depending on image scale, species, and resolution, and high segmentation overlap for both stomata and pavement cells. The generality of stomatal detection was also tested on spring onion, chickpea, balsam poplar, and wheat in cross-species feasibility tests, although performance was more variable in monocots, and pavement-cell segmentation remained species- and architecture-dependent. Benchmarking against another publicly available app further showed that, under the tested web interface settings and image types, Stomatalia exhibited closer agreement with manual counts and substantially faster processing times. The practical performance of Stomatalia was further tested in a proof-of-concept trial with potato plants subjected to optimal irrigation and a long, gradual drought. The platform reliably quantified epidermal traits despite variations in leaf morphology and image quality, supporting the integrated interpretation of stomatal and pavement-cell responses under stress. Conclusions We developed Stomatalia as a robust, user-friendly deep learning platform for automated, high-throughput analysis of bright-field leaf epidermal images across varying magnifications and resolutions. Stomatalia facilitates rapid, reproducible, and coordinated phenotyping of stomatal and pavement cells by integrating methodological standardisation, computational automation, and multi-trait extraction in a single analytical workflow. Its strongest current application is the analysis of high-quality dicot leaf-print images, particularly in species and imaging conditions similar to those used for model training and validation. Cross-species and benchmark analyses further define its current scope: stomatal detection can be transferred to some additional epidermal architectures, whereas robust pavement-cell segmentation in monocots or highly divergent species will require further annotation and model retraining. Within these defined boundaries, Stomatalia provides a flexible and extensible framework for studying stomatal and pavement cell morphology and environmental plasticity, while also supporting broader efforts to dissect and optimise plant responses to abiotic stress.
Why it matches plant phenotyping methods気孔・舗装細胞の形態形質を画像から自動抽出する深層学習プラットフォームを開発し、独立データで性能評価・比較検証しているため、植物フェノタイピング手法が中心である。
abstractwe developed Stomatalia , a deep learning-based platform designed to automate and standardise the quantification of stomatal and pavement cell traits.
Reproduction assets foundThe authors publicly deposited the paper's test image datasets (raw/input leaf-print images, detection outputs, manual ground-truth counts, exported datasets) and the model file on Figshare, and provide a public Google Colab demo for running the Stomatalia algorithm. Both are paper-specific, public, and actionable.Dataset · publicThe test datasets and model file used in this work are available through the following link:
https://doi.org/10.6084/m9.figshare.32532672. The test_sets.zip archive contains the test image datasets
(cross-species and benchmark analysis), including raw/input images, detection output images, manual
ground-truth counts and exported datasets.Open asset ↗Figshare · 10.6084/m9.figshare.32532672pdf-page:25 lines:1-75Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Cotton leaf diseases present a major threat to global cotton production, significantly impacting both yield and fiber quality. Traditional diagnostic methods are labor-intensive, time-consuming, and demand highly skilled professionals, making them inefficient for large-scale agricultural applications. Although earlier deep learning -based approaches have shown promising results in identifying cotton leaf diseases such as Bacterial Blight, Fusarium Wilt, and Curl Virus Disease, their performance is often limited by complex preprocessing requirements and insufficient generalization to real-world field conditions. To address these challenges, this study proposes and optimized transfer learning-based model, CLDP-CNN, designed to enhance feature extraction and classification efficiency using pre-trained deep neural networks. This study demonstrates the development of Cotton Leaf Disease Prediction Convolutional Neural Network (CLDP-CNN) automatically, utilizing Transfer Learning (TL) which operates on meticulously prepared datasets. Two distinct datasets were used to train the model: the first consisted of field images from cotton farms, while the second was sourced from Kaggle. The main goal of this research examines how the model performs on real-world field datasets. The CLDP-CNN model has proven highly accurate by attaining 99.78% detection success rates for cotton leaf diseases when processing primary dataset which surpasses its secondary dataset accuracy rate of 99.62%. Both the primary dataset and secondary dataset resulted in high accuracy values for the VGG16 pre-trained model which achieved 99.56% accuracy on the primary dataset and 98.82% on the secondary dataset. A web-based application enhances the capabilities of the CLDP-CNN model by providing real-time updates on the health status of cotton plants. This technology empowers farmers with valuable information, enabling them to take timely protective actions to prevent potential severe yield losses in their cotton crops.
Why it matches plant phenotyping methods綿花葉の画像から病害状態を推定する深層学習モデルを開発・評価しており、植物フェノタイピング手法が研究の中心です。
abstractThe main goal of this research examines how the model performs on real-world field datasets.
Reproduction assets foundThe paper's cotton leaf disease image datasets are publicly available: the authors' primary field-collected dataset on the first author's GitHub repository, and the secondary Kaggle dataset used for comparison. No analysis code or trained model checkpoints are explicitly deposited.Dataset · publicbia.
Funding: This work was supported by Princess Nourah bint Abdulrahman
University Researchers Supporting Project number (PNURSP2026R760),
Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.
Data and code availability
The data that support the findings of this study are openly available in Github and
Kaggle at. https://github.com/mnaeem303/Cotton-Leaf_Disease-Dataset),
and https://www.kaggle.com/datasets/seroshkarim/cotton-leaf-disease-dataset/data
Author Contributions
All the authors (Muhammad Naeem, Muhammad Ibrahim, Nadeem Sarwar, Oumaima
Saidani, Asma Irshad, Muhammad Shadab Alam Hashmi, Muhammad Tayyab
Qammar) contributed equally to this work in their respective meaningOpen asset ↗https://github.com/mnaeem303/Cotton-Leaf_Disease-Datasetpdf-raw-page:29 lines:1-54Dataset · publicbdulrahman
University Researchers Supporting Project number (PNURSP2026R760),
Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.
Data and code availability
The data that support the findings of this study are openly available in Github and
Kaggle at. https://github.com/mnaeem303/Cotton-Leaf_Disease-Dataset),
and https://www.kaggle.com/datasets/seroshkarim/cotton-leaf-disease-dataset/data
Author Contributions
All the authors (Muhammad Naeem, Muhammad Ibrahim, Nadeem Sarwar, Oumaima
Saidani, Asma Irshad, Muhammad Shadab Alam Hashmi, Muhammad Tayyab
Qammar) contributed equally to this work in their respective meaningful ways. All the
authors have read and approved the final manuOpen asset ↗pdf-raw-page:29 lines:1-54Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Food productivity, quantity and quality are at stake when plant diseases such as rice diseases undermine the food security. Rice leaf disease treatment necessitates accurate and timely diagnosing. This study describes a deep learning model for categorizing and forecasting rice plant diseases. Using the remora optimization algorithm (ROA) on a rice leaf dataset demonstrates its potential for plant disease classification. The ROA-DM method detects rice leaf diseases using the ROA algorithm, a deep maxout network (DMN), and a deep autoencoder (DAE). ROA is applied to the learning parameters of deep model in order to achieve better convergence and avoiding local minima, which usually happens with conventional gradient-based optimizers. Experiments show that the suggested framework is accurate and precise across illness categories. The confusion matrices display the training and validation accuracy, losses of this model. The performance of our optimal learning method with respect to other methods indicated its potential for identifying leaf diseases. The accuracy of the ROA-DM method is 98.5%.
Why it matches plant phenotyping methodsイネ葉の観察画像から病害状態を分類・検出する深層学習手法が研究の中心であり、植物病害フェノタイピング手法に該当する。
abstractThis study describes a deep learning model for categorizing and forecasting rice plant diseases.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe dataset collected from https://www.kaggle.com/datasets/emmarex/plantdisease (PlantVillage dataset) for algorithm testing in plant disease diagnosis 35 . The selected rice leaf disease samples from PlantVillage dataset consisting of 3050 colour leaf images across four classes.Open asset ↗Kaggle · emmarex/plantdiseaselines:85-97Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Published1 Jul 2026The Plant journal : for cell and molecular biologyCited by 0 · OpenAlex ↗
Genetically encoded biosensors are one of the essential tools in biological research. They enable visualization of molecules of interest from the subcellular level to entire organism level in vivo and can be used to monitor the presence of small molecules, gene expression, protein activity, and protein degradation. However, multiplexing fluorescent biosensors in plants is notoriously difficult due to signal bleed-through and strong autofluorescence from chlorophyll. In this study, we investigated the potential of multiplexing biosensors based on the selection of reporter fluorescent proteins. We characterized the emission spectra, fluorescence lifetimes, and relative brightness of diverse fluorescent proteins in plant leaves. We show that selected proteins exhibit comparable brightness, supporting their use in co-expression experiments and reliable quantification of individual signals. To separate three overlapping signals, we applied two different linear unmixing approaches and compared them to results obtained without unmixing. We identified the channel separation unmixing approach as the most suitable for biosensors. Additionally, we show how unmixing with the selected approach can be applied to separate autofluorescence and five fluorescent proteins. We further validated this approach in virus-infected cells by following organelle dynamics in vivo. Finally, we demonstrate the feasibility of high-throughput segmentation and quantification with a custom MATLAB workflow for nuclei, chloroplasts, and cytoplasm signal analysis. Overall, our work demonstrates that biosensors can be multiplexed, even when their emission spectra overlap.
Why it matches plant phenotyping methods植物組織における蛍光シグナルの分離、画像セグメンテーション、定量化ワークフローを開発・比較・検証しており、植物の細胞・細胞小器官状態を測定する方法が中心である。
abstractTo separate three overlapping signals, we applied two different linear unmixing approaches and compared them to results obtained without unmixing.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicThe segmentation and histograms were acquired using MATLAB script ( https://github.com/NIB‐SI/Nuclei‐segmentation ). The parameters used in the script to achieve appropriate segmentation are listed on GitHub, Case 1 ( https://github.com/NIB‐SI/Nuclei‐segmentation ).Open asset ↗NIB‐SI/Nuclei‐segmentationlines:255-341Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Rice leaf diseases pose a major challenge to crop health and agricultural productivity, particularly when timely and accurate diagnosis is required under natural field conditions. The development of automated disease recognition systems depends heavily on the availability of large, well-annotated image datasets. However, many existing rice leaf disease datasets are limited in terms of environmental variability, disease representation, and real-field imaging conditions. To address this gap, this paper presents BanglaRiceLeaf, an original rice leaf image dataset collected and curated by the authors from the experimental fields of the Bangladesh Rice Research Institute (BRRI), Gazipur, Bangladesh, between July 2023 and July 2024. The dataset contains 4152 images belonging to five classes: Bacterial Leaf Blight, Bacterial Leaf Streak, Sheath Blight, Leaf Blast, and Healthy Leaf. The images were acquired from two rice varieties, BR11 and BRRI dhan34, under natural field conditions across varying illumination environments in order to reflect practical disease recognition scenarios. All images were manually annotated by trained annotators under expert supervision. The dataset is systematically organized and publicly released to support reproducible research in rice disease classification. In addition to dataset presentation, benchmark experiments using Xception, NASNetMobile, and InceptionV3 are provided to demonstrate its applicability for deep learning-based disease recognition. BanglaRiceLeaf is expected to serve as a useful resource for plant disease analysis, comparative model evaluation, and future research in precision agriculture and agricultural computer vision.
Why it matches plant phenotyping methodsイネ葉の病徴・健全状態を画像で分類する公開ベンチマークデータセットであり、データ収集・注釈・ベンチマーク評価が中心です。
abstractthis paper presents BanglaRiceLeaf, an original rice leaf image dataset collected and curated by the authors
Reproduction assets foundThe paper's core asset is the BanglaRiceLeaf rice leaf disease image dataset (4152 field images, five classes), publicly released on Harvard Dataverse with DOI 10.7910/DVN/XAOBYW. No author analysis code or trained model checkpoints are stated as publicly available.Dataset · publicData Identification Number: https://doi.org/10.7910/DVN/XAOBYW
Direct URL to Data: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/XAOBYW
Access Instructions: This dataset is publicly available on the Harvard Dataverse repository and can be accessed for academic, research, and instructional purposes.Open asset ↗Harvard Dataverse · doi:10.7910/DVN/XAOBYWhtml-lines:100-131Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published1 Jul 2026International Journal of IoT, Embedded Systems and Industrial AutomationCited by 0 · OpenAlex ↗
Modern agriculture is rapidly adopting Artificial Intelligence (AI) and Internet of Things (IoT) technologies to improve crop monitoring and decision-making. Many existing systems focus either on water stress detection or pest detection separately. The proposed system integrates both functions into a single platform. It uses a camera module and environmental sensors connected to a Raspberry Pi (5/4) as the main controller. A Convolutional Neural Network (CNN) model processes leaf images captured by the AI camera, while a soil moisture sensor supports water stress analysis. The system classifies crops into three categories: healthy, water-stressed, and pest-infected. Based on the output, it provides real-time recommendations for irrigation and pesticide application. This reduces manual inspection, prevents unnecessary chemical usage, saves water, and improves crop productivity.
Why it matches plant phenotyping methods植物の葉画像と土壌水分センサーを用いて、健康・水ストレス・害虫感染という植物の状態を自動分類する統合センシング基盤を開発しており、表現型取得・判定が中心的です。
abstractThe proposed system integrates both functions into a single platform.
Reproduction assets foundThe paper's CNN phenotyping/classification analysis is built directly on two public Kaggle image datasets (PlantVillage plant disease and Crop Water Stress), explicitly cited with URLs. No author code or trained model is deposited.Dataset · publicThe PlantVillage Dataset was used for
plant disease detection, and it is available at https://www.kaggle.com/datasets/emmarex/plantdisease.Open asset ↗Kaggle · emmarex/plantdiseasepdf-page:7 lines:1-57Dataset · publicThe Crop Water Stress Dataset was used for crop water stress analysis, and it can be accessed at
https://www.kaggle.com/datasets/harshilsharma/crop-water-stress.Open asset ↗Kaggle · harshilsharma/crop-water-stresspdf-page:7 lines:1-57Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Paddy leaf disease (PLD) detection has grown more difficult, yet early detection might prevent significant losses due to decreased crop yield. However, existing models struggle to accurately classify diseases under difficult circumstances like intricate backgrounds, fluctuating lighting, and overlapping leaves. Additionally, existing models do not incorporate efficient optimization strategies, leading to suboptimal accuracy and poor generalization on unseen data. To address these challenges, a novel deep learning-based YOLO-LEAFNET method for PLD detection utilizing IGT-YOLO, integrating the YOLOv8 disease detection with the Improved Gorilla Troops (IGT) optimization. The input paddy leaf images are pre-processed using Bilateral Contrast Limited Adaptive Histogram Equalization (B-CLAHE) to enhance image quality and improve local contrast while preserving disease boundaries. YOLOv8 model is utilized to detect and classify paddy leaf diseases by accurately localizing affected regions with bounding boxes. Then, the IGT algorithm boosts the disease detection accuracy by optimizing YOLOv8 through effective hyperparameter tuning. The proposed YOLO-LEAFNET method effectiveness was evaluated using recall, F1 score, specificity, accuracy, and precision. B-CLAHE enhanced noise-free images improve contrast and detection accuracy, while the IGT-YOLO model ensures scalable, efficient early diagnosis of paddy leaf diseases with 99.07% accuracy. The YOLO-LEAFNET enhanced the total accuracy by 3.21%, 5.25%, and 1.98% related to CNN, DeepRice, and FasterR-CNN, respectively.
Why it matches plant phenotyping methodsイネ葉の病害状態を画像から検出・分類するYOLOベース手法を提案し、前処理・最適化・性能評価を中心に扱っているため、植物フェノタイピング手法として該当する。
abstractTo address these challenges, a novel deep learning-based YOLO-LEAFNET method for PLD detection utilizing IGT-YOLO, integrating the YOLOv8 disease detection with the Improved Gorilla Troops (IGT) optimization.
Reproduction assets foundThe paper's phenotyping input is the public UCI Rice Leaf Diseases dataset (paddy leaf images of bacterial leaf blight, leaf smut, brown spot), also mirrored on Kaggle. No author analysis code, trained model, or supplementary assets are disclosed.Dataset · publicThe dataset is publicly available
at: https://archive.ics.uci.edu/dataset/486/rice+leaf+diseases. The dataset is distributed
under the Creative Commons Attribution 4.0 (CC BY 4.0) license.Open asset ↗rice+leaf+diseasespdf-page:7 lines:1-33Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Tomatoes are the most significant and widely consumed crops globally. Leaf diseases cause an important threat to crop production and quality. Further, various fungi, bacteria and viruses can influence the plant's various parts and gradually destroy the quality and production of tomatoes in an agricultural field, thus it impacts the surrounding cultivated plants to cause more economical loss to farmers. Therefore, various techniques are proposed recently to optimally recognize and categorize the epidemic pathogens. To enhance sustainable plant protection practices, accurate identification and classification of diseases is essential to enhance the production rates. Effective pathogen detection and monitoring of plant health are critical areas of research in agriculture. Understanding disease severity is a crucial role for management practices and preventing the spread of infections. Rapid assessment is crucial because early detection of disease can significantly enhance the crop yield and influence the management strategies implemented by farmers. In this proposed model, a deep learning approach is proposed to classify the severity of diseases in tomato plants. At first, the needed images are collected from publicly available resource. Further, the collected images are subjected to the Adaptive and Attention-based Mask Region Convolutional Neural Network (AA-MRCNN) for optimally segmenting the abnormal regions from the gathered image. Further, the hyperparameters, like epoch, steps per epoch, and hidden neuron count in the Adaptive and Attention-based Mask Region Convolutional Neural Network are tuned by the Fitness-based African Vultures Optimization (FAVO) algorithm. Also, the segmented images are passed into Multiscale Recurrent MobileNet (MRMNet) module for categorizing disease severity in the tomato plant epidemic. The assessment of the recommended severity detection approach of tomato plant disease is determined by conducting a simulation experiment. The proposed model attains better outcomes of 93% accuracy, 93% specificity, 93% precision, 7% False Negative Rate (FNR), 86% Matthews Correlation Coefficient (MCC), 93% Fowlkes mallow Index (FM), 86% Bookmaker Informedness (BM), and 86% Threat Score (TS) measures in the ReLu activation function, which is progressed than the conventional frameworks. The result defines that the suggested technique outperformed than other baseline models to ensure the dependability of the tomato plant epidemic pathogens detection performance.
Why it matches plant phenotyping methodsトマト葉画像から病変領域を分割し、植物病害の重症度を分類する画像ベースの表現型推定手法を開発・評価しており、方法が研究の中心である。
abstracta deep learning approach is proposed to classify the severity of diseases in tomato plants.
Reproduction assets foundThe paper uses a public Kaggle tomato leaf disease image dataset as its phenotyping input and states that the authors' source code is available in a public GitHub repository. Both are paper-specific, publicly accessible, and actionable.Code · publiche tomato disease classification performances were
carried out among the performance metrics like Prevalence Threshold
(PT), BM, Precision, FNR, Accuracy, FM, Specificity, MK (Markedness)
and TS to maximize the reliability of the designed approach. The source
code of the public repository on GitHub link is available on “GitHub-
https://github.com/pdeepika6078/Severity-Classification-of-Tomato-Plant-Epidemic-Pathogens-/tree/main”
In order to demonstrating the effectiveness of the designed approach,
several conventional segmentation, optimization, and classification
approaches are adopted to compare the overall process. The reason
behind selecting the traditional approaches to improve the clasOpen asset ↗pdeepika6078/Severity-Classification-of-Tomato-Plant-pdf-raw-page:55 lines:1-29Code / dataset availability confirmedCrossref · checked 14 Sept 2026
This study addresses the need for efficient and accessible plant disease identification systems in the era of Agriculture 4.0, where advances in artificial intelligence (AI) and machine learning (ML) support data-driven agricultural practices. The increasing popularity of home gardening highlights challenges faced by users in identifying plant diseases due to limited knowledge and diagnostic tools. Therefore, this research aims to develop a web plant disease detection system using the Django framework and convolutional neural networks (CNNs). The model was trained on a controlled dataset consisting of 57,320 leaf images collected from the PlantVillage and Turmeric Plant Disease datasets. Image preprocessing was applied, including resizing, normalization, and data augmentation such as image rotation, zooming, image inversion and brightness adjustmen. Class imbalance during training was handled using class weighting. The dataset is divided into a training set and a validation set for model development and evaluation. The CNN model achieved an accuracy of 92% on the labeled validation dataset, with a mean F1 score of 0.79 and a weighted mean F1 score of 0.92. For generalization testing, an uncontrolled (wild) dataset consisting of 223 images collected from online sources was used, resulting in an accuracy of 11%, indicating limited real-world generalization due to domain differences. Despite this limitation, the proposed system demonstrates the feasibility of CNN-based plant disease classification in a web application.
Why it matches plant phenotyping methodsCNNによる葉画像からの植物病害識別システムを開発し、検証データと野外データで性能評価しているため、植物の病害状態を推定する画像ベースのフェノタイピング手法が中心である。
abstractthis research aims to develop a web plant disease detection system using the Django framework and convolutional neural networks (CNNs).
Reproduction assets foundThe paper's CNN plant-disease model was trained primarily on the public Kaggle 'New Plant Diseases Dataset' (PlantVillage-derived, 54,528 images), which is a paper-specific, publicly available image dataset directly used for the study's phenotyping measurements. The Turmeric Plant Disease Dataset (Mendeley DOI 10.17632Dataset · public[23] Samir Bhattarai, “New Plant Diseases Dataset,” San Francisco, CA, USA, 2018. Accessed: Jun. 16, 2025.
[Online]. Available: https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-datasetOpen asset ↗Kaggle · vipoooool/new-plant-diseases-datasetpdf-page:15 lines:1-45Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Global food security is largely based on the accurate and timely diagnosis of crop diseases, where paddy rice is an extremely essential staple of more than half of the world population. The conventional disease identification techniques tend to be laborious, time consuming and demand a great deal of domain knowledge, which becomes a bottleneck in the efficient management of the farms. Although deep learning [and especially Convolutional Neural Networks (CNNs)] have demonstrated a spectacular performance in automated classification of diseases based on leaf images, they tend to overlook important contextual features that are implicitly processed by agronomic experts. The visual defects of a disease might be unclear and this can greatly differ depending on factors like the genetic variety of the plant and the stage of development. We overcome this shortcoming by proposing a new multi-modal deep learning framework, Multi-Modal Factorized Bilinear Pooling (MFBP) model which is capable of a more holistic and precise paddy health measurement. The proposed method is the only one that combines high-level visual information obtained using leaf images and related tabular information, namely the paddy type and number of days. The MFBP model uses Factorized Bilinear Pooling (FBP) rather than the simple feature concatenation which commonly loses the complex relationship between different data types. This systematic method efficiently encodes all the complex interactions between all components of the visual and tabular features vectors in such a way that helps the model to pick up subtle, context-specific patterns. As an example, it will only be possible to educate the model that a specific visual blemish is predictive of a given disease through a specific species at a specific age. We test our model on the Paddy Doctor: Paddy Disease Classification dataset, which is a detailed public dataset comprising of more than 10,000 labeled images and containing relevant metadata, and thus it forms a perfect testing bed to conduct multi-modal research. Through our detailed experiments, we have shown that the proposed MFBP model is much better than a baseline model based on concatenation fusion, which proves that deep, multiplicative interactions can be best modeled in this task. The findings highlight the massive possibilities of multi-modes AI in the development of more robust, more accurate, and more context-aware diagnostic instruments and precision agriculture to enable more sustainable and productive agricultural activities.
Why it matches plant phenotyping methods葉画像とメタデータを統合してイネの健康状態・病害を推定する新規深層学習手法を提案し、ベースライン比較で検証しているため、植物フェノタイピング手法が中心である。
abstractWe overcome this shortcoming by proposing a new multi-modal deep learning framework, Multi-Modal Factorized Bilinear Pooling (MFBP) model which is capable of a more holistic and precise paddy health measurement.
Reproduction assets foundThe paper's phenotyping inputs are the public Kaggle 'Paddy Doctor: Paddy Disease Classification' dataset (10,407 leaf images with tabular metadata for variety and age), explicitly named in the Data Availability statement with a persistent public URL. No author analysis code, trained models, or checkpoints are reportedDataset · publicThe datasets used and/or analysed during the current study are publicly available in the
“Paddy-doctor: paddy disease classification” repository at the following persistent
URL: https://www.kaggle.com/datasets/vbookshelf/paddy-disease-classification.Open asset ↗Kaggle · vbookshelf/paddy-disease-classificationpdf-page:20 lines:1-74Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Image processing plays a vital role in precision agriculture by enabling automated disease detection and crop health monitoring. This research presents a novel Hybrid Vision Transformer with Dense Attention Capsule Network (HVT-DACapNet) model for accurate cotton plant disease detection. The proposed framework integrates Adaptive Wavelet Transform Filtering (AWTF) for noise removal while preserving disease-related features. A Hybrid Vision Transformer (HVT) is employed to extract both local spatial patterns and global contextual dependencies, and the Dense Attention Capsule Network (DACapNet) captures hierarchical spatial relationships with an attention mechanism that emphasizes infected regions. In addition, a hybrid optimization strategy combining Mayfly and Aquila Optimization (HMAO) is used to fine-tune model hyperparameters for improved convergence. The model is evaluated on a publicly available Kaggle cotton leaf disease dataset containing healthy leaves and multiple disease categories including Target Spot, Powdery Mildew, Bacterial Blight, Army Worm, and Aphids, using a 70:15:15 train-validation-test split under the simulation setup and hyperparameter configuration described in the manuscript. The proposed HVT-DACapNet achieves an F1-score of 99.68%, sensitivity of 99.68%, specificity of 98.89%, and an overall accuracy of 99.79%, outperforming existing models such as ConvLSTM-ZOA, GOA, SFO, Inception-V3, and VGG-16.
Why it matches plant phenotyping methods綿花葉の画像から病害状態を推定する深層学習手法を新規開発し、公開データセットで性能評価しているため、植物フェノタイピング手法が中心である。
abstractThis research presents a novel Hybrid Vision Transformer with Dense Attention Capsule Network (HVT-DACapNet) model for accurate cotton plant disease detection.
Reproduction assets foundThe paper's only qualifying asset is the public Kaggle Cotton Plant Disease Dataset used as the input image dataset for all experiments. The authors' code is explicitly not publicly available (institutional restrictions), with only a supplementary algorithm document and on-request implementation details.Dataset · publicthodological workflow of the proposed model. Additional
implementation details may be made available from the corresponding
author upon reasonable request for academic and non-commercial
research purposes.
Data Availability-The datasets generated and/or analysed during the
current study are available in the Kaggle repository,
https://www.kaggle.com/datasets/dhamur/cotton-plant-disease
Author’s contribution – G.Neelavathi– Research proposal –
construction of the workflow and model – Final Drafting– Survey of
Existing works – Improvisation of the proposed model;
Dr.K.Venkatasalam – Initial Drafting of the paper – Collection of
datasets and choice of their suitability – Formulation of pseudOpen asset ↗Kaggle · dhamur/cotton-plant-diseasepdf-raw-page:39 lines:1-42Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Crop leaf diseases cause 10–40% annual yield losses, yet timely field diagnosis remains difficult. Vision-language models (VLMs) lift recognition accuracy with rich textual descriptions, but multimodal pipelines are too slow for real-time field use because they require text processing at inference. We present MTL-AWL, a framework built on a training–inference asymmetry: VLM text serves as privileged training-time supervision, and two coupled mechanisms—one retaining VLM semantics in the image encoder and one exploiting them—enable image-only deployment at multimodal accuracy. A modal-dropout strategy (p=0.6) intermittently masks the VLM text sequence during training, forcing the image encoder to retain cross-modal representations independently. An adaptive multi-task loss jointly optimizes InfoNCE contrastive alignment, attention diversity, and modality consistency under learnable softmax weights, consistently converging to a dominant contrastive weight (55% on soybean, 68% on PlantDoc)—identifying cross-modal alignment as the primary mechanism of VLM knowledge transfer. At inference, the model reaches 818 FPS (3.7× faster than multimodal methods) at only 0.41% accuracy cost, attaining 99.30%/98.89% (multimodal/image-only) on soybean and 72.65%/68.80% on PlantDoc—compact enough for real-time, offline field screening.
Why it matches plant phenotyping methods葉画像から植物病害状態を推定する画像ベース手法を開発し、複数データセットで精度・速度を評価しており、フェノタイピング手法が中心である。
abstractWe present MTL-AWL, a framework built on a training–inference asymmetry: VLM text serves as privileged training-time supervision, and two coupled mechanisms—one retaining VLM semantics in the image encoder and one exploiting them—enable image-only deployment at multimodal accuracy.
Reproduction assets foundThe paper's Data Availability Statement links a public Dryad DOI for the soybean leaf disease image dataset used in the study's phenotyping/recognition experiments. No author code or model release is stated.Dataset · publicThe datasets utilized in this study are openly accessible. The soybean dataset is available at https://doi.org/10.5061/dryad.41ns1rnj3 .Open asset ↗Dryad · 10.5061/dryad.41ns1rnj3lines:441-459Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
Plants are geometrically and topologically complex objects, and methods and devices that produce plant point clouds often miss parts due to self occlusions, making further analysis, such as phenotypic trait extraction or 3D reconstruction, difficult. We introduce A-Occ-Plant , a novel method for point cloud completion. The first novelty of our algorithm is converting point clouds into a set of images, which are then completed using 2D amodal segmentation. The images are then converted into a complete point cloud by using view-consistent Gaussian splats. The second novelty is the use of a coarse-to-fine hierarchical Transformer with cross-scale attention. The completed soft masks are fused into a continuous 3D density field using Gaussian splatting, removing the need for external pose estimation or fixed-size inputs. We introduce a synthetic dataset using a procedural model and a real-world plant reconstruction benchmark with artificially generated occlusions. We further benchmark A-Occ-Plant against representative 3D point-cloud completion methods, demonstrate that it recovers downstream phenotypic traits (leaf count, leaf angle, plant height), and show that it generalizes to another crops (soybean). A-Occ-Plant achieves a 264.8% improvement in LPIPS and an 8.3% gain in SSIM compared to the current state of the art, while using only 2.3% of the parameters and running 39.4× faster. We release our code at https://github.com/JaeLee18/PlantPhenomics_Occlusion.
Why it matches plant phenotyping methods植物の遮蔽点群を補完し、葉数・葉角度・草丈という表現型形質を復元する手法を開発しており、データセット作成とベンチマーク検証も中心的に行っている。
abstractWe introduce A-Occ-Plant , a novel method for point cloud completion.
Reproduction assets foundThe paper explicitly releases authors' code and sample data (inference code, sample data for reproducing results) via a Google Drive project download and a GitHub repository, both with explicit availability statements and public URLs.Code · publicThe full code and data at https://github.com/JaeLee18/PlantPhenomics_Occlusion .Open asset ↗JaeLee18/PlantPhenomics_Occlusionlines:386-410Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Maize is a globally important food crop, and its yield and quality are vulnerable to various leaf diseases. To address issues such as blurred edges of disease spots and difficulty in small target detection, this study proposes the MAC-YOLO11 model improved on the basis of YOLOv11m. The model introduces the MSPP module to enhance the extraction capability of edge and directional features, and incorporates the spatial position attention module CDSA to strengthen the modeling capability for differences in lesion morphology, texture, and spatial distribution. We designed the APC module to expand the effective receptive field at a low parameter cost through asymmetric convolution branches. The dataset covers 11 categories: northern leaf blight, brown spot, common rust, smut, downy mildew, fall armyworm larval damage, gray leaf spot, maize streak virus disease, adult corn borer damage, corn borer larval damage, and healthy maize leaves. Results show that the MAC deep learning model achieves mAP50 and mAP50:95 of 94.1% and 83.5%, respectively, with overall performance superior to YOLOv11m and other mainstream models. This study provides a technical solution for intelligent identification of maize diseases and holds significant value for disease monitoring and precise control in smart agriculture.
Why it matches plant phenotyping methodsトウモロコシ葉の病斑形態・テクスチャ・空間分布を画像から識別する深層学習モデルを開発・評価しており、植物病害状態の画像ベース表現型計測が中心である。
abstractthis study proposes the MAC-YOLO11 model improved on the basis of YOLOv11m.
Reproduction assets foundThe paper's authors explicitly state that the source code and implementation details of the proposed MCA-YOLO11 maize leaf disease detection model are publicly available on GitHub, matching an allowed URL. No public dataset deposit is stated for the 17,729-image maize leaf disease dataset.Code · publicThe source code and implementation details for the proposed model are publicly
available on GitHub at: https://github.com/xuzhiheng0402/MCA-modelOpen asset ↗xuzhiheng0402/MCA-modelpdf-page:18 lines:1-53Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Pearl millet is an important crop in arid regions, but its yield is reduced by foliar diseases like Downy Mildew and Rust. Traditional and deep learning methods struggle with accurate lesion detection, severity estimation, and robustness under complex field conditions, and often lack interpretability for practical agricultural deployment. To address these challenges, this study proposes the Adaptive Severity-Aware Swin Attention Network (ASA-SAN), an integrated framework designed for disease segmentation, classification, and severity estimation in pearl millet leaves. The proposed architecture combines a Swin Transformer encoder for hierarchical feature extraction with a ResUNet++ decoder for accurate lesion segmentation. This is further enhanced using Adaptive Channel Attention to improve feature discrimination and a dual-stream classification network to jointly capture local lesion characteristics and global contextual information. Additionally, an Adaptive Disease Severity Index (ADSI) is introduced to quantitatively assess disease progression based on lesion area ratio, color degradation, edge irregularity, and texture variations. Experimental evaluations conducted on a pearl millet leaf dataset demonstrate that the proposed method achieves a Dice score of 97.8%, IoU of 95.6%, classification accuracy of 98.3%, and F1-score of 98.2%, outperforming several state-of-the-art methods. Furthermore, Grad-CAM visualizations enhance model interpretability by highlighting disease-relevant regions. Overall, the ASA-SAN framework provides a robust, interpretable, and severity-aware solution for automated pearl millet disease analysis, enabling early detection and supporting precision agriculture practices for improved crop protection and yield optimization.
Why it matches plant phenotyping methods真珠粟葉の病斑を画像から分割・分類し、病害重症度を定量推定する手法を中心に開発・評価しているため、植物表現型計測手法として含める。
abstractAdditionally, an Adaptive Disease Severity Index (ADSI) is introduced to quantitatively assess disease progression based on lesion area ratio, color degradation, edge irregularity, and texture variations.
Reproduction assets foundThe paper's phenotyping inputs are drawn from a public, open-access image dataset: the Pearl Millet Leaf Disease dataset (Version 2) hosted on Roboflow Universe, containing annotated images of Downy Mildew, Rust, and healthy pearl millet leaves. This is a paper-specific, publicly available asset directly used for the作者Dataset · publicThe dataset used in this research was taken from the publicly available open-access Pearl
Millet Leaf Disease dataset hosted on Roboflow Universe, which has images of Downy
Mildew, Rust and healthy pearl millet leaves annotated publicly available [26]. To ensure
experimental consistency and reproducibility, all experiments were conducted with Version
2 of the open access dataset.Open asset ↗Roboflow Universepdf-raw-page:10 lines:1-28Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
LeafScans-Orchard is a curated, multi-year RGB image dataset of orchard plant leaves designed to support research in computer vision, machine learning, and plant phenotyping. The dataset comprises 9708 high-quality leaf scans acquired during collection campaigns conducted between 2015 and 2025, covering seven orchard crop species: apple, pear, sweet cherry, sour cherry, plum, peach, and apricot. In total, the dataset includes 67 cultivar labels. All samples were acquired using flatbed scanning under controlled conditions on a uniform background, ensuring high visual consistency and minimal background variability. The original scans were captured at 1200 dpi and subsequently converted into a public release format at 300 dpi, stored as lossless TIFF images to preserve morphological and textural details. Each image corresponds to a single leaf and is organized in a hierarchical directory structure by species, cultivar, and acquisition year, accompanied by image-level metadata and aggregated species–cultivar–year counts. LeafScans-Orchard is suitable for plant species classification, cultivar recognition, leaf morphology analysis, texture analysis, and general visual feature extraction. In addition to the main release, a representative subset of 300 original 1200 dpi scans is provided to support high-resolution analyses. The dataset is particularly suited for fine-grained classification, morphology-driven analysis, and methodological studies under controlled imaging conditions.
Why it matches plant phenotyping methods果樹葉のRGBスキャン画像を収録した公開データセットで、植物フェノタイピングおよび葉形態解析を目的とする。標準化された画像取得と再利用可能なデータ構成が中心であり、フェノタイピング用データセットとして適格。
abstractLeafScans-Orchard is a curated, multi-year RGB image dataset of orchard plant leaves designed to support research in computer vision, machine learning, and plant phenotyping.
Reproduction assets foundThe paper's core asset is the LeafScans-Orchard dataset itself (9708 RGB leaf scans, 300 dpi TIFF release plus 1200 dpi subset, image-level metadata and summary counts), openly deposited on Zenodo with an explicit DOI and CC BY 4.0 license. This is a paper-specific, public, actionable phenotyping image dataset. No codeDataset · publicthe published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The dataset described in this article is openly available in Zenodo
as LeafScans-Orchard Dataset (v1.0.0) at https://doi.org/10.5281/zenodo.20187966 (accessed on
10 May 2026). The repository includes the 300 dpi image release, the 1200 dpi high-resolution subset,
image-level metadata, aggregated species–cultivar–year counts, and supporting documentation. The
complete archive of original 1200 dpi scans is retained locally by the authors but is not included in
the current pubOpen asset ↗Zenodo · 10.5281/zenodo.20187966pdf-raw-page:12 lines:1-46Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published23 Jun 2026International Journal of Innovative Research and Scientific StudiesCited by 0 · OpenAlex ↗
Plant health plays a critical role in agriculture, climate balance, and economic stability. However, plant diseases caused by bacteria, fungi, and viruses can significantly reduce crop productivity if not detected early. Traditional manual inspection methods are time-consuming, labor-intensive, and prone to human error, especially in large-scale farming. To address these challenges, this study proposes an automated and accurate plant disease detection system using deep learning-based object detection models for early disease diagnosis in agriculture. A publicly available dataset containing 38 different plant leaf diseases annotated in You Only Look Once (YOLO) format is used, along with a standardized preprocessing pipeline to ensure data quality and consistency. Three modern architectures: YOLOv8, YOLOv11, and YOLOv26 were trained and evaluated under identical conditions using the Ultralytics framework on Google Colab. Experimental results show that YOLOv11 achieves the highest accuracy in terms of precision, recall, and mean Average Precision (mAP), while YOLOv8 provides the fastest inference speed with lower computational complexity. Based on the results, the study concludes that YOLO-based models show great potential for plant disease detection, with YOLOv11 offering superior detection accuracy among the evaluated models. The practical implications of these findings lie in the potential for precision agriculture to monitor diseases in real-time, minimize crop losses, and aid in timely decision-making for farmers and agricultural stakeholders.
Why it matches plant phenotyping methods植物葉の病害状態を画像から検出するYOLOベース手法を提案し、複数モデルを同一条件で評価しているため、病害表現型の取得・抽出が中心である。
abstractthis study proposes an automated and accurate plant disease detection system using deep learning-based object detection models
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicFor this study, we obtained a publicly accessible dataset (plant disease detection dataset) from Kaggle [30]. The
dataset comprises 2569 images of 13 different plant species, as shown in Figure 2.Open asset ↗Kagglepdf-page:6 lines:1-41Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Wheat (Triticum aestivum L.) is a staple crop of paramount importance to global food security; however, its productivity is significantly compromised by foliar diseases. Conventional diagnostic approaches, relying on manual observation or laboratory analyses, are often labor-intensive and susceptible to inaccuracies. While recent advancements in deep learning present promising avenues for automated disease detection, persistent challenges such as limited annotated datasets, environmental heterogeneity, and model generalization continue to hinder optimal performance. This study proposes a novel hybrid deep learning model called HybridViT, which combines ConvNeXt and Vision Transformer (ViT) architectures with the Convolutional Block Attention Module (CBAM) to improve the classification of wheat leaf diseases. While ConvNeXt ensures local feature extraction and ViT provides global contextual understanding, CBAM dynamically highlights the most discriminative features. Additionally, the Contrast Limited Adaptive Histogram Equalization (CLAHE) method is employed to enhance the visibility of disease symptoms in low-contrast leaf images. Unlike conventional hybrid CNN-Transformer approaches that rely on static feature concatenation, the proposed model employs an adaptive gated fusion mechanism to dynamically balance local and global feature representations. The fused features are further refined using a lightweight CBAM module to enhance discriminative capability. Additionally, Contrast Limited Adaptive Histogram Equalization (CLAHE) is applied to improve feature visibility under varying illumination conditions. Evaluated on three different datasets obtained under both controlled and field conditions, HybridViT achieved 100% accuracy on balanced datasets and 99.10% accuracy on complex images captured in real-world conditions, surpassing existing methods. Furthermore, a 5-fold cross-validation strategy yielded an average accuracy of 99.04% ± 0.22, demonstrating the model's robustness and stability across different data splits. The results demonstrate the model's robustness against environmental noise, lighting variations, and class imbalance. This approach, which enables early and accurate disease diagnosis, supports sustainable agricultural practices, reduces pesticide use, and contributes to global food security.
Why it matches plant phenotyping methods小麦葉の病徴を画像から分類する深層学習手法を開発・検証しており、植物病害状態の取得・推定が研究の中心です。
abstractThis study proposes a novel hybrid deep learning model called HybridViT, which combines ConvNeXt and Vision Transformer (ViT) architectures with the Convolutional Block Attention Module (CBAM) to improve the classification of wheat leaf diseases.
Reproduction assets foundThe paper evaluates HybridViT on three public wheat leaf disease image datasets from Kaggle, cited in the reference list with explicit URLs. These are the paper-specific image inputs used for its disease-classification measurements. No author analysis code, trained model checkpoints, or supplementary code deposit is披露dDataset · publicAvailable: https://www.kaggle.com/datasets/olyadgetch/wheat-leaf-datasetOpen asset ↗Kaggle · olyadgetch/wheat-leaf-datasetpdf-page:51 lines:1-64Dataset · public[78] J. Jayaprakash, “Wheat Leaf Disease,” Kaggle. Accessed: May 1, 2026.
[Online]. Available:
https://www.kaggle.com/datasets/jayaprakashpondy/wheat-leaf-diseaseOpen asset ↗Kaggle · jayaprakashpondy/wheat-leaf-diseasepdf-page:51 lines:1-64Dataset · public[79] S. Kumar, “Multiple Plant Diseases Dataset,” Kaggle. Accessed: May 1, 2026.
[Online]. Available:
https://www.kaggle.com/datasets/samareshkumar/multipleplantdiseasesOpen asset ↗Kaggle · samareshkumar/multipleplantdiseasespdf-page:51 lines:1-64Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Rice is a fundamental food source for more than half of the global population, making stable yields and quality improvements vital for food security and sustainable agricultural development. Early infections of rice leaf diseases often exhibit subtle symptoms, while conventional control methods based on empirical judgment and routine pesticide application result in both yield losses and environmental pollution. A Multi-scale closed-loop tuning via spatial frequency collaborative sensitivity (MCCA-YOLO) model has been proposed in this paper with a multiscale closed-loop tuning and spatial frequency collaborative attention mechanism for the early detection and classification of rice crop diseases. MCCA-YOLO incorporates a closed-loop tuning compound network architecture that combines a dual-backbone feature extractor with a spatial frequency enhancement module to achieve system self-verification feedback, reducing transmission errors and enhancing the texture features of leaves. The framework implements a cross-scale weighted fusion and a deformable spatial hybrid attention enhanced bidirectional feature pyramid fusion network for dynamic feature adaptation, effectively accommodating the complex morphology of rice leaf lesions. By conducting comprehensive ablation studies and comparative experiments with existing techniques on the rice plant diseases v8 dataset, the proposed approach achieves a mean average precision (mAP) of 92.2%, outperforming well-established methods, while delivering superior precision (0.915) and recall (0.900). Extensive empirical validation of additional v9 and Rice Leaf Spot Disease (RLSD) datasets for rice plant diseases further demonstrates the model's outstanding performance.
Why it matches plant phenotyping methodsイネ葉の病徴を画像から検出・分類するYOLOベース手法を開発し、アブレーション、比較実験、複数データセットで性能検証しており、植物病害表現型の取得手法が中心である。
abstractA Multi-scale closed-loop tuning via spatial frequency collaborative sensitivity (MCCA-YOLO) model has been proposed in this paper with a multiscale closed-loop tuning and spatial frequency collaborative attention mechanism for the early detection and classification of rice crop diseases.
Reproduction assets foundThe paper's rice leaf disease image datasets (Roboflow v8/v9, Kaggle RLSD) are explicitly declared publicly available, and the authors' MCCA-YOLO analysis code is stated to be open source on GitHub with a public URL.Code · publicOur code is publicly accessible as open source at: https://github.com/sstan12/MCCA-YOLOOpen asset ↗GitHub · sstan12/MCCA-YOLOlines:147-153Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Accurate and timely identification of plant diseases along with their severity is critical for effective crop management and minimizing agricultural losses. While recent advances in deep learning have demonstrated high performance in plant disease classification, limited attention has been given to quantifying disease severity, which is essential for informed agronomic decision-making. To address this gap, this study proposes TomatoMTL, a unified multi-task learning framework for simultaneous disease classification and severity estimation of tomato leaf diseases from a single image. The proposed architecture employs a shared ResNet50-based convolutional backbone augmented with CBAM-based feature refinement, followed by task-specific branches for disease classification and severity prediction. Furthermore, a cross-task attention mechanism is introduced to enable interaction between disease-specific and severity-related features, thereby enhancing the robustness of severity estimation. To effectively leverage partially labeled data, a masking strategy is incorporated during training. Experimental evaluation on a publicly available tomato leaf disease severity dataset demonstrates that the proposed model achieves 97.85% disease classification accuracy and 77.66% severity prediction accuracy, outperforming state-of-the-art single-task classifiers including EfficientNetV2-S, ViT-B/16, and ConvNeXt-Tiny as well as existing multi-task learning baselines including Cross-Stitch Networks and MTAN. Comprehensive ablation studies confirm the individual contributions of CBAM, MixUp and CutMix augmentation, and the cross-task attention mechanism. Statistical significance analysis across five independent runs yields p-values less than 0.001 and Cohen's d greater than 14, establishing the reliability of the reported improvements. Quantitative localization analysis reveals that the model achieves 89.4% Pointing Game accuracy, confirming that attention maps focus on biologically meaningful disease regions. The proposed framework represents a complete and effective approach for integrated plant disease analysis with strong potential for real-world precision agriculture applications.
Why it matches plant phenotyping methodsトマト葉画像から病害の重症度という植物状態を推定するマルチタスク画像解析手法を開発・評価しており、表現型取得・推定が研究の中心である。
abstractthis study proposes TomatoMTL, a unified multi-task learning framework for simultaneous disease classification and severity estimation of tomato leaf diseases from a single image.
Reproduction assets foundThe paper's Data availability and Code availability statements explicitly link a public Kaggle tomato leaf disease severity dataset (the phenotyping image data used) and the authors' public GitHub repository containing the TomatoMTL implementation scripts and documentation.Dataset · publicThe datasets analysed during the current study are publicly available in the kaggle Data repository at: https://www.kaggle.com/datasets/janiruwalisingha/tomato-leaf-disease-severity-dataset .Open asset ↗kaggle Data repository · tomato-leaf-disease-severity-datasetlines:354-386Code · publicThe implementation, along with relevant scripts and documentation, can be accessed through the following GitHub repository: https://github.com/Parnika798/tomato_leaf_disease .Open asset ↗GitHub repository · Parnika798/tomato_leaf_diseaselines:354-386Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Accurate and efficient identification of cucumber leaf diseases is a critical step in preventing losses and facilitating timely intervention in agricultural activi-ties. However, most state-of-the-art plant disease recognition models, including those employing deep learning, often fail to identify spatial dependencies among symptomatic leaf feature regions, require high computational resources, and lack robustness in their predictions. To overcome these challenges, this paper pro-poses MobileGraph, a graph-aided deep learning model that jointly reasons local texture patterns and spatial dependencies among CNN-derived cucumber leaf feature regions using MobileNetV3 as a lightweight feature extractor. Experi-ments on a publicly available cucumber leaf disease dataset containing 5 classes and 4,000 images show that the proposed model achieves 99.75% accuracy, 99.75% macro F1-score, and 99.69% MCC, outperforming several state-of-the-art models including ResNet-152, EfficientNet-B7, DenseNet-201, ConvNeXt, and VGG16, while having a significantly lower computational cost of 0.465 GFLOPs. Explainability results from Grad-CAM and LIME indicate that the model is focused on biologically important regions of plant lesions. Furthermore, a proto-type mobile application illustrates the feasibility of real-time cucumber disease diagnosis for practical agricultural monitoring. These results indicate that Mobi-leGraph provides an efficient and interpretable solution for intelligent crop health surveillance.
Why it matches plant phenotyping methodsキュウリ葉の病斑という植物状態を画像から診断する深層学習手法を開発し、複数モデルとの性能比較・検証まで行っており、病害表現型の取得・推定が研究の中心である。
abstractExplainability results from Grad-CAM and LIME indicate that the model is focused on biologically important regions of plant lesions.
Reproduction assets foundThe paper's experiments use the publicly available Cucumber Disease Recognition Dataset (4,000 images, 5 classes) hosted on Mendeley Data, which is a paper-specific public phenotype/image asset. The MobileGraph source code is only available upon request, so it does not qualify as a public asset.Dataset · publicThe dataset analysed of this study, titled ”Cucumber Disease Recognition
Dataset” is publicly available in the Mendeley Data repository at
(https://data.mendeley.com/datasets/y6d3z6f8z9/1).Open asset ↗Mendeley Data · y6d3z6f8z9pdf-page:36 lines:1-71Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Abstract Rice and potatoes are major crops in Bangladesh, frequently affected by major disease outbreaks that challenge food security. Inaccurate disease identification often contributes to yield losses. Recently, machine learning garnered much attention in identifying crop diseases. The present study was conducted to develop a deep learning model-based image‑analysis system that automatically identifies key diseases of Bangladeshi rice and potato, and integrates it into a web app to provide farmers with rapid, accurate diagnoses. The system employs a convolutional neural network (CNN) implemented with TensorFlow’s Sequential API, featuring ReLU-activated hidden layers and a Softmax output layer. A dataset of 4,809 images, comprising both healthy and diseased, was collected and processed through pre-processing, feature extraction, and classification. A web-based application was deployed utilizing the Python Streamlit framework. This application integrates the proposed model to predict 2 rice diseases viz. blast ( Magnaporthe oryzae ), bacterial leaf blight ( Xanthomonas campestris ), and 2 potato diseases viz. Early blight (Alternaria solani) and Late blight ( Phytophthora infestans ) from uploaded images, providing a confidence score for the predictions with approximately 92.84% for all detected diseases. The proposed model achieved a training accuracy of 0.9357, a validation accuracy of 0.8983, and a test accuracy of 0.9333. The developed web application indicates strong diagnostic performance for four major diseases, offering Bangladeshi farmers an accessible tool to make timely management decisions.
Why it matches plant phenotyping methodsイネ・ジャガイモ葉の病徴を画像から分類するCNNと実用Webアプリを開発・評価しており、植物の病害状態推定が中心的な方法論的貢献である。
abstractdevelop a deep learning model-based image‑analysis system that automatically identifies key diseases of Bangladeshi rice and potato
Reproduction assets foundThe paper's rice/potato leaf disease image dataset partially comes from Kaggle, and the data availability statement points to PlantVillage for additional image data; both are public image assets used for the paper's CNN phenotyping/disease-classification analysis. No author analysis code, trained model checkpoints, or专Dataset · publicch, M.Y.H. analyzed the data, A.A.J.,
452
M.Y.H. and M.S. wrote this manuscript, M.R.I., F.M.A. and S.O.N. reviewed and edited the
453
manuscript. All authors have read and agreed to the published version of the manuscript.
454
Data availability statement
455
Some of the datasets used in this study was obtained from Kaggle
456
(https://www.kaggle.com/datasets). Additional datasets used and/or analyzed during the current
457
study are available from the corresponding author upon reasonable request. More image data can
458
be found at https://www.plantvillage.org/en/plant_images
459Open asset ↗Kagglepdf-raw-page:24 lines:1-57Dataset · publiche manuscript.
454
Data availability statement
455
Some of the datasets used in this study was obtained from Kaggle
456
(https://www.kaggle.com/datasets). Additional datasets used and/or analyzed during the current
457
study are available from the corresponding author upon reasonable request. More image data can
458
be found at https://www.plantvillage.org/en/plant_images
459Open asset ↗PlantVillagepdf-raw-page:24 lines:1-57Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Early and accurate disease detection is important for increasing the agricultural output, decreasing the financial costs, and ensuring food security. Traditional diagnostic procedures take much time and effort, involve the necessity of having deep expertise, and are not always suitable for large scale farming disease detection. For this purpose, the current research suggests developing an explainable lightweight CNN-based model for crop disease identification based on RGB leaf images. The model utilizes several innovative architectural solutions such as depth-wise separable convolution, SE blocks, skip connections, and guided attention-based feature learning that allow enhancing the effectiveness of features extraction and decreasing computation load. Moreover, Grad-CAM is used to visualize affected areas on a map and thus increase the interpretability of the model. The suggested solution was implemented and tested on the PlantVillage dataset containing 54,305 images for 38 crop diseases out of 14 crops. The results show that the training, validation, and testing accuracies equal 97.6%, 88.3%, and 97.63%, correspondingly, along with the Macro-F1 measure of 0.867 and Micro-ROC-AUC equal to 0.99. A comparative study reveals that the presented model performs comparably well in terms of classification with lightweight structure and built-in interpretability capabilities to be applied in the mobile and edge-enabled agriculture environment. The results show that the presented approach is capable of being used as an effective and interpretable tool for diagnosing plant diseases in real-time.
Why it matches plant phenotyping methods葉画像から植物病害を推定するCNN手法の開発・評価が研究の中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として含める。
abstractthe current research suggests developing an explainable lightweight CNN-based model for crop disease identification based on RGB leaf images.
Reproduction assets foundThe paper's plant-phenotyping input is the public PlantVillage leaf-image dataset (54,305 RGB images, 38 crop-disease classes), explicitly declared in the Data availability statement with a Kaggle URL. No author code, trained model, or checkpoint is deposited.Dataset · publicThe data set analyzed during current study are available in https://www.kaggle.com/datasets/emmarex/plantdisease.Open asset ↗Kaggle · emmarex/plantdiseaselines:366-390Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Crop diseases play a significant role in food production globally; therefore, there is an urgent need to develop quick and accurate diagnostic techniques that are more effective than manual inspection methods. The proposed hybrid multimodal learning framework in this research provides a solution that integrates adaptive therapy suggestion, market price prediction, and image-based disease detection. This study also proposes a framework for pesticide recommendation and the treatment of plants. This study experiment on tomato and cotton crop leaf data for disease detection. Experimental results on a tomato crop disease detection dataset show that the proposed model shows high performance. EfficientNetB0 provides more stability and generalization capabilities in different scenarios compared to other models, such as YOLOv8, ResNet50, and a custom CNN model. The use of a knowledge-based decision support system provides sustainable pesticide recommendations based on environmental and symptom-specific parameters. Forecasting of pesticide prices through LSTM methods yields forecasts within 3.2% and 4.1% MAE, enabling improved decision-making by providing instant points of reference for potential price movements. Research uses SHAP and LIME to provide explainability to users, thus improving user buy-in through transparency. Overall, this modular system provides a data-driven decision-making model to improve the efficiency of managing crops.
Why it matches plant phenotyping methods植物葉画像から病害状態を推定する画像ベース手法を、複数モデルで比較評価しており、植物病害フェノタイピングがシステムの主要構成要素です。価格予測や農薬推薦も含みますが、病害検出の技術評価が明示されています。
abstractThe proposed hybrid multimodal learning framework in this research provides a solution that integrates adaptive therapy suggestion, market price prediction, and image-based disease detection.
Reproduction assets foundThe paper's disease-detection experiments use publicly available cotton and tomato leaf image datasets (Kaggle, IEEE DataPort, Roboflow), all cited with explicit public URLs in the references. No author analysis code or trained model checkpoints are stated as publicly available; the supplementary material is referencedDataset · publiccholar
View reference in article
19
Muppala C. Guruviah V. ( 2020 ). Machine vision detection of pests, diseases, and weeds: a review . J. Phytol. 12 , 9 – 19 . doi: 10.25081/jp.2020.v12.6145
CrossRef
Google Scholar
View reference in article
20
National College of Ireland ( 2025 ). “Cotton Disease Dataset.” Available online at: https://www.kaggle.com/datasets/janmejaybhoi/cotton-disease-dataset (Accessed May 19, 2025).
Google Scholar
View reference in article
21
Naveed Gul and Kaggle ( 2026 ). Tomato Leaf Disease . Kaggle. Available online at: https://www.kaggle.com/datasets/naveedgull/tomato-leaf-disease (Accessed March 29, 2026).
Google Scholar
View reference in article
22
Ngugi H. N. EzugOpen asset ↗Kagglelines:554-633Dataset · publicreference in article
20
National College of Ireland ( 2025 ). “Cotton Disease Dataset.” Available online at: https://www.kaggle.com/datasets/janmejaybhoi/cotton-disease-dataset (Accessed May 19, 2025).
Google Scholar
View reference in article
21
Naveed Gul and Kaggle ( 2026 ). Tomato Leaf Disease . Kaggle. Available online at: https://www.kaggle.com/datasets/naveedgull/tomato-leaf-disease (Accessed March 29, 2026).
Google Scholar
View reference in article
22
Ngugi H. N. Ezugwu A. E. Akinyelu A. A. Abualigah L. ( 2024 ). Revolutionizing crop disease detection with computational deep learning: a comprehensive review . Environ. Monit. Assess. 196 : 302 . doi: 10.1007/s10661-024-12454-z
Pubmed AOpen asset ↗Kagglelines:554-633Dataset · publicComputer Vision and Pattern Recognition (CVPR) ( Las Vegas, NV : IEEE ), 779 – 788 . doi: 10.1109/CVPR.2016.91
CrossRef
Google Scholar
View reference in article
29
Roboflow ( 2026a ). A Comprehensive Dataset of Cotton Plant Diseases for National Disease Identification and Treatment Guidance | IEEE DataPort. Available online at: https://ieee-dataport.org/documents/comprehensive-dataset-cotton-plant-diseases-national-disease-identification-and-treatment (Accessed March 29, 2026).
Google Scholar
View reference in article
30
Roboflow ( 2026b ). Cotton Plant Disease Prediction Object Detection Model by National College of Ireland . Available online at: https://universe.roboflow.com/national-colleOpen asset ↗IEEE DataPortlines:554-633Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Accurate assessment of leaf chlorophyll is essential for understanding plant physiological responses to environmental variation. While solvent extraction provides precise chlorophyll measurements, it is destructive and temporally limited, whereas portable optical meters such as the CCM-300 enable rapid, non-destructive measurement of the chlorophyll fluorescence ratio (CFR) but require species- and season-specific calibration. This study evaluates the performance of CCM-300 measurements and reconstructs seasonal chlorophyll dynamics in field maple (Acer campestre) across two contrasting summers in the United Kingdom. Paired CFR and acetone-extracted chlorophyll data collected in 2023 were used to develop calibration models. RF regression achieved the highest predictive performance within the calibration dataset, although substantial uncertainty remained at the leaf level; a simple linear model was therefore adopted for cross-year projection due to its stability under extrapolation. Applying this calibration to daily 2022 CFR measurements generated a continuous "virtual acetone" trajectory, enabling qualitative comparison with weekly destructive extractions in 2023. Both years exhibited mid-season chlorophyll plateaus followed by late-summer declines; however, senescence, defined as the initiation of sustained post-peak decline, occurred earlier during the warmer and drier 2022 season. Mixed-effects modelling identified positive effects of temperature and wind speed on CFR in 2022, while generalised additive modelling of the 2023 dataset revealed a non-linear seasonal decline under comparatively mild conditions. Because cross-year projections rely on a low-fit linear calibration, interannual differences are interpreted primarily in terms of relative seasonal trajectory shape and timing rather than absolute chlorophyll magnitude.
Why it matches plant phenotyping methodsCCM-300による葉クロロフィル測定を破壊的測定と比較し、校正モデルの開発・性能評価と季節軌跡の再構築を行っており、植物表現型取得法が研究の中心である。
abstractThis study evaluates the performance of CCM-300 measurements and reconstructs seasonal chlorophyll dynamics in field maple (Acer campestre) across two contrasting summers in the United Kingdom.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicData Availability: Data used in the study can be accessed via https://zenodo.org/records/17475985.Open asset ↗zenodo · 17475985pdf-page:11 lines:1-44Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Early and accurate detection of apple leaf diseases is critical for sustainable agriculture, yet manual diagnosis remains time-consuming and error-prone. This study introduces a novel deep learning framework centered on a custom ContinuousLayer, a spatially adaptive convolutional layer designed to overcome the limitations of standard CNNs. This architecture automates the classification of apple leaf diseases Black rot, rust, scab, and healthy leaves with high precision. The model addresses dataset imbalance through strategic resampling, achieving uniform class distribution. The ContinuousLayer introduces spatial feature modulation using trainable Gaussian basis functions, enhancing feature extraction while penalising kernel irregularities through a hybrid composite loss function. Trained on a dataset of 3,164 images balanced via bicubic up-sampling, and evaluated on a held-out test set of 10% of the data, the model attains a 98.63% test accuracy, with F1-scores ranging from 0.98 to 1.00 across classes. Visual analysis of the confusion matrix reveals minimal misclassification, predominantly between rust and scab. Comparative evaluation against baseline architectures demonstrates the efficacy of the ContinuousLayer in capturing disease-specific spatial patterns. These results underscore the potential of integrating mathematically inspired layers into CNNs for plant pathology applications, offering a highly accurate tool for precision agriculture in controlled environments.
Why it matches plant phenotyping methodsリンゴ葉の病徴を画像から分類する新規深層学習層と解析手法を開発・比較評価しており、植物病害状態の表現型抽出が中心である。
abstractThis study introduces a novel deep learning framework centered on a custom ContinuousLayer, a spatially adaptive convolutional layer designed to overcome the limitations of standard CNNs.
Reproduction assets foundThe paper's apple leaf disease image dataset (3,164 images) is explicitly stated to be publicly available on Kaggle, matching an allowed URL. No author code or model checkpoints are reported as available.Dataset · publicThe datasets analysed during the current study is publicly available in the Kaggle repository at https://www.kaggle.com/datasets/mhantor/apple-leaf-diseases.Open asset ↗Kaggle · mhantor/apple-leaf-diseaseslines:595-613Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Aegle marmelos (bael) is a medicinally important tropical crop that remains severely underrepresented in computational plant pathology research. This study proposes AMnet, a deep learning framework integrating an InceptionV3 backbone with a fixed graph convolutional network to capture both within-region disease texture and between-region spatial propagation patterns for automated four-class classification of Aegle marmelos leaf diseases: Cercospora leaf, healthy leaf, leaf curl, and leaf spot. The graph module constructs a 25-node spatial graph directly from CNN feature maps, enabling end-to-end training without external preprocessing. Compared against MobileNetV2, InceptionV3, VGG19, and DenseNet201, AMnet achieved the best overall performance with 98.83% accuracy, 98.84% F1-score, 99.95% PR-AUC, and 98.45% MCC, alongside the fastest inference time of 1.83 ms. Robustness was confirmed through bootstrap confidence interval estimation and four-fold cross-validation. PCA-based clustering analysis with silhouette scoring and Davies–Bouldin indexing demonstrated clear class separability in the learned embeddings. Grad-CAM and LIME visualizations confirmed that predictions were grounded in biologically meaningful leaf regions rather than background artifacts. A Gradio-based prototype further demonstrated practical deployment potential. Although broader field validation remains necessary, AMnet provides an accurate, interpretable, and reproducible framework for diagnosing Aegle marmelos leaf disease.
Why it matches plant phenotyping methods葉画像から植物の病害状態を分類する説明可能な深層学習手法を開発・比較・検証しており、植物表現型（病徴・病害状態）の取得・推定が中心的です。
abstractThis study proposes AMnet, a deep learning framework integrating an InceptionV3 backbone with a fixed graph convolutional network to capture both within-region disease texture and between-region spatial propagation patterns for automated four-class classification of Aegle marmelos leaf diseases: Cercospora leaf, healthy leaf, leaf curl, and leaf spot.
Reproduction assets foundThe paper analyses a publicly available Aegle marmelos leaf disease image dataset deposited on Mendeley Data, explicitly linked in the Data availability statement and reference [36]. No author analysis code or trained model checkpoint is reported as publicly available.Dataset · publicThe dataset analysed of this study is publicly available in the Mendeley Data repository at
(https://data.mendeley.com/datasets/54r883j5zr/1).Open asset ↗Mendeley Datapdf-page:29 lines:1-43Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Nutrient deficiencies in coffee plants significantly impact bean quality and yield, making timely detection crucial for successful cultivation. Current assessment methods rely on manual inspection, which is labor-intensive and time-consuming, posing challenges for large-scale field management. This approach often results in inconsistent evaluations and delayed interventions. Methods This study presents CoNutriNet, an automated deep learning architecture that integrates DenseNet121 with a novel Graph-Enhanced Attention Feature Network (GEAFNet) for classifying nutrient deficiencies in coffee leaves. DenseNet121 provides deep hierarchical and regional feature representation, while GEAFNet captures local, fine-grained spatial features through Inception, Ghost, and Efficient Channel Attention (ECA) modules. Furthermore, a Graph Convolutional Network (GCN) is included to model spatial dependencies and structural variations between leaf regions. Feature representations from both pathways are concatenated and refined using a Coordinate Attention (CA) module to enhance discriminative capability. Results Evaluation on the CoLeaf dataset demonstrates that CoNutriNet achieves an accuracy of 94.5%. The integration of lightweight attention mechanisms, dense connectivity, and graph-based modeling improves both performance and computational efficiency. Conclusion These results indicate that CoNutriNet achieves and efficient performance in nutrient deficiency detection in coffee crops, highlighting its potential for deployment in agricultural environments to support precision farming and optimize yield.
Why it matches plant phenotyping methodsコーヒー葉の栄養欠乏という植物状態を画像から分類する深層学習手法を開発し、データセットで性能評価しており、表現型取得・推定が研究の中心である。
abstractThis study presents CoNutriNet, an automated deep learning architecture that integrates DenseNet121 with a novel Graph-Enhanced Attention Feature Network (GEAFNet) for classifying nutrient deficiencies in coffee leaves.
Reproduction assets foundThe paper's phenotyping analysis (coffee nutrient deficiency classification) is performed on publicly available leaf image datasets. The data availability statement links a Mendeley Data repository containing the analyzed data, which is an allowed URL. No author analysis code or trained model checkpoints are explicitlyDataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://data.mendeley.com/datasets/brfgw46wzb/1Open asset ↗brfgw46wzb/1lines:866-910Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Common beanLeafClassificationStress / disease detectionDisease symptoms / severity
Introduction Accurate disease diagnosis is crucial for enhancing agricultural productivity and reducing postharvest losses, directly impacting food quality and safety. Traditional detection methods often rely on extensive feature modeling and perform poorly in complex field environments. Methods This study proposes a deep learning model called ZDAM, based on an improved ZFNet integrated with a dual attention mechanism. The classical ZFNet is first optimized to improve feature extraction efficiency. A combined channel and spatial attention mechanism is then incorporated to refine feature representation for disease identification in key crops. Finally, a residual module is added to boost accuracy. Results Evaluated on a dataset of 11,903 bean leaf images covering healthy leaves and four disease types, including leaf mould, rust, mosaic, and white spot, the model achieves an average recognition accuracy of 99.02%, outperforming MobileMamba, Vision Transformer, and Chest- OMD. Discussion This approach offers a scalable solution for automated disease monitoring, supporting postharvest quality preservation and sustainable crop production.
Why it matches plant phenotyping methods豆葉の病害状態を画像から推定する深層学習モデルを開発し、複数モデルとの性能比較も行っており、植物フェノタイピング手法が研究の中心である。
abstractThis study proposes a deep learning model called ZDAM, based on an improved ZFNet integrated with a dual attention mechanism.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publictomato leaf disease data from the open-source dataset New Plant Disease Dataset ( https://www.kaggle.com/vipoooool/new-plant-diseases-dataset ) were also utilized. Both datasets include healthy samples and four disease categories: rust disease, mosaic disease, leaf mold disease, and white spot disease. ( https://pan.baidu.com/s/197Lyn2TGdIjLCE2gylsiHA?pwd=krpw )Open asset ↗pan.baidu.comlines:405-484Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
Abstract Background Sorghum ( Sorghum bicolor ) is a versatile C4 crop used for food and feed and as biomass for bioproducts and energy. Improving nitrogen use efficiency (NUE) in sorghum is important because fertilizer is costly and excessive fertilizer use has negative environmental impacts. Leaf senescence mediates nutrient recycling, but its dynamic progression is difficult to quantify at scale. We evaluated whether visible-near-infrared hyperspectral imaging can provide high-throughput measures of N-limitation-induced senescence in sorghum and link these phenotypes to gene expression. Sorghum Tx430 plants were grown under four N treatments (6, 9, 12, and 15 mM), imaged from vegetative growth through grain fill, and destructively sampled for RNA-seq at four developmental stages. Results A supervised support vector machine with a radial basis function kernel classified pixels from a hyperspectral image of sorghum plants grown under different N levels into green leaf, yellow leaf, dry leaf, stalk, panicle, and background classes with 0.93 accuracy. We defined the senescence ratio as the sum of yellow and dry leaf areas divided by the green leaf area and computed it across multiple growth stages and nitrogen levels. The senescence ratio did not differ among N treatments during vegetative growth, but it declined with increasing N during boot, anthesis, and grain fill, indicating earlier senescence under N limitation. Among the genes whose expression positively correlated with senescence ratio were 13 putative transcription factors, including SbiRTX430.02G247100, a WRKY1/ZAP1 homolog and a WRKY4 homolog. Gene regulatory network analysis of the top 1% of genes associated with SbiRTX430.02G247100 showed enrichment for processes associated with leaf senescence and chlorophyll catabolism. In contrast, the network associated with the WRKY4 homolog was enriched for autophagy-related terms. Conclusions Our study shows that automated hyperspectral imaging is highly effective for monitoring dynamic plant phenotypes, such as stress-induced senescence, that are difficult to visually score with the naked eye. Here, nitrogen deficiency served as the stress condition. Still, this approach supports large-scale phenotypic data collection for any such stressor and enables analyses with greater statistical power, yielding more robust conclusions and the potential for new insights that can be applied to engineering and breeding better crops.
Why it matches plant phenotyping methodsソルガムの動的な老化表現型を高スループットに取得する hyperspectral imaging と、SVMによる画像分類・senescence ratio算出が研究の中心であり、植物状態の定量化手法を実証している。
abstractWe evaluated whether visible-near-infrared hyperspectral imaging can provide high-throughput measures of N-limitation-induced senescence in sorghum
Reproduction assets foundThe paper's availability statement points to a public GitHub repository containing the authors' image-processing, machine-learning classification, transcriptomic analysis, and figure-generation scripts. The 148 GB hyperspectral image data is only promised 'upon acceptance' (not yet public), and the RNA-seq deposit is aCode · publicle in the NCBI SRA repository,
552
under BioProject PRJNA1452908 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1452908)
553
(RNA-seq raw reads SRR38119224 to SRR38119282). Scripts used for image processing,
554
machine-learning classification, transcriptomic analyses, and figure generation will be
555
accessible through GitHub (https://github.com/belafif2/TX430_Senescence). Image data (148
556
GB) will be made available in a data repository upon acceptance. Other relevant processed data
557
files and supporting figures are available as supplementary data documents.
558
559
Competing interests
560
The authors declare that they have no competing interests.
561
Funding
562
This work was funded Open asset ↗belafif2/TX430_Senescencepdf-raw-page:22 lines:1-54Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Rubber tree powdery mildew is a major foliar disease that threatens the yield and quality of natural rubber. Its lesions are typically small, irregular, and embedded in complex backgrounds, making accurate automated detection difficult. Methods To address this challenge, we propose RubberFormer, an end-to-end detection framework based on a refined Transformer architecture for detecting small powdery mildew lesions in complex scenarios. RubberFormer adopts MobileNetV4 as a lightweight backbone, introduces the Hierarchical Attention with Local-global Optimization (HALO) module for multiscale local-global feature fusion, incorporates the Unified Cross-Attention Network (UCAN) to enhance multidimensional feature interaction, and applies Normalized Wasserstein Distance (NWD) Loss to improve small-object localization. Results Extensive experiments were conducted on PM-Dataset-Plus, which contains 9,765 images, and PD-40, a large-scale plant disease dataset containing 80,369 images across 40 disease categories and 8 crops. RubberFormer achieved superior detection accuracy and generalization performance compared with existing methods, while maintaining computational efficiency suitable for practical agricultural monitoring. Discussion These results demonstrate that RubberFormer is effective for detecting small and irregular rubber tree powdery mildew lesions under complex conditions. The framework has practical value for rubber tree disease monitoring and provides a transferable design strategy for agricultural vision tasks involving small objects and complex backgrounds.
Why it matches plant phenotyping methodsゴム樹の病斑という植物の病害状態を画像から検出するTransformer手法を開発し、複数データセットで性能検証しており、植物表現型取得が中心である。
abstractwe propose RubberFormer, an end-to-end detection framework based on a refined Transformer architecture for detecting small powdery mildew lesions in complex scenarios.
Reproduction assets foundThe paper's authors publicly release both plant disease image datasets used in this study: PM-Dataset-Plus (9,765 rubber tree powdery mildew images) and PD-40 (80,369 images, 40 categories, 8 crops), each with an explicit availability statement and GitHub URL matching the allowed URLs. No analysis code or trained modelDataset · publicPM-Dataset-Plus is available at https://github.com/wfcyliyuheng-dev/PM-Dataset-PlusOpen asset ↗wfcyliyuheng-dev/PM-Dataset-Pluslines:1199-1255Dataset · publicPD-40 is available at https://github.com/wfcyliyuheng-dev/PD40-DatasetOpen asset ↗wfcyliyuheng-dev/PD40-Datasetlines:1199-1255Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Accurate crop disease detection supports precision agriculture, but field-deployable identification remains hindered by complex backgrounds, varying illumination, and heavy deep learning models. This work presents a lightweight visual detection approach for grape leaf diseases under unconstrained field conditions. Built on the YOLO11n backbone, the method integrates three customized modules: C3k2-UltraLightBlock for efficient feature representation, LeafRepFusionStem for low-level feature enhancement, and RCSA-HSFPN for refined multi-scale fusion with residual channel-spatial attention. A dedicated dataset with complex backgrounds is constructed via augmentation and background replacement. Experiments show the model achieves 92.0% precision, 92.9% recall, and 93.0% mAP@0.5, with only 2.9 GFLOPs and 1.73 M parameters, representing 54.7% and 33.2% reductions over the baseline. Heatmap visualization confirms improved lesion focusing and background suppression, while cross-crop tests validate strong generalization. This framework provides an efficient solution for real-time, edge-deployable plant disease monitoring, balancing accuracy and computational efficiency for practical agricultural visual computing applications.The implementation code for this study is available at:https://github.com/aitizc/Lightweight-Visual-Detection-Framework-for-Complex-Background-Grape-Leaf-Disease-Identification.git
Why it matches plant phenotyping methodsブドウ葉の病斑・病害状態を画像から推定する軽量な視覚検出手法を開発・評価しており、植物病害表現型の取得が中心である。
abstractThis work presents a lightweight visual detection approach for grape leaf diseases under unconstrained field conditions.
Reproduction assets foundThe authors explicitly state that the implementation code for this study is publicly available on their GitHub repository. The paper's grape leaf disease dataset itself is not stated as deposited (only the public PlantVillage source is cited), so only the authors' code qualifies as a paper-specific public asset.Code · publicThe implementation code for this study is avail-
able at:https://github.com/aitizc/Lightweight-Visual-Detection-Framework-for-
Complex-Background-Grape-Leaf-Disease-Identification.gitOpen asset ↗https://github.com/aitizc/Lightweight-Visual-Detection-Framework-for-pdf-page:2 lines:1-43Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published8 Jun 2026International Journal of Engineering and ManufacturingCited by 0 · OpenAlex ↗
Plant disease detection is vital for agricultural sustainability and food security. While Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) have achieved high accuracy in this domain, CNNs often require millions of parameters and substantial computation. ViTs suffer from the quadratic time and space complexity of self-attention (SA), limiting their use on resource-constrained devices. Although SA is capable of modelling long-range dependencies when symptoms are dispersed, many plant diseases exhibit small, localized lesions or texture changes; therefore, Neighborhood Attention (NA) offers a more efficient and targeted alternative by focusing on nearby regions rather than the entire image. This work proposes a custom Localized NA block implemented in TensorFlow/Keras that operates directly on CNN feature maps, bypassing patch embedding and transformer modules. A lightweight CNN is then developed by combining depth-wise separable convolutions with the proposed localized NA block. In addition, a 100-category plant disease dataset covering 16 crops is presented. The dataset is curated, class-balanced, and made publicly available to support reproducibility and encourage further research. The proposed 9-layer CNN, with just 1.7M parameters and a size of 6.74 MB, achieved a favorable balance between accuracy, model size, and computational efficiency, compared with MobileNetV1, MobileNetV2, DenseNet121, InceptionV3, MobileViT-XXS, and EfficientViT-M0, achieving 98.97%± 0.33% accuracy on PlantVillage and 93.36%± 0.28% on the proposed dataset. The ablation study showed that the NA block improved test accuracy by approximately 2–3%, while Grad-CAM visualizations indicated more precise targeting of diseased areas in the leaf image.
Why it matches plant phenotyping methods植物葉画像から病徴を推定する軽量CNNと注意機構を開発し、複数データセットで比較評価・アブレーションを行い、さらに100カテゴリの公開データセットを提示しているため、植物フェノタイピング手法が中心である。
abstractThis work proposes a custom Localized NA block implemented in TensorFlow/Keras that operates directly on CNN feature maps
Reproduction assets foundThe paper's authors curated a 100-category plant disease dataset and explicitly state it is publicly available on Kaggle in both augmented-train and raw split forms. These are paper-specific, public, actionable phenotype image datasets. The PlantVillage benchmark is a third-party dataset, not a paper-specific asset, soDataset · publicrs declare no conflict of interest
Funding Declaration
This research work was supported by KLE Technological University, Hubbali, India under the Ph.D. Fellowship
Program.
Data Availability Statement
The newly curated 100-category Plant Disease Dataset used in this study is publicly available on Kaggle.
Augmented Train Dataset: https://www.kaggle.com/datasets/rithambararajput/augmented-train
Raw Dataset: https://www.kaggle.com/datasets/rithambararajput/100-class-split-raw-dataset
The Plant Village dataset, used as a benchmark for comparative evaluation, is also publicly accessible at:
https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset.Ethical Declarations
This study does noOpen asset ↗Kaggle · rithambararajput/augmented-trainpdf-raw-page:19 lines:1-51Dataset · publicsupported by KLE Technological University, Hubbali, India under the Ph.D. Fellowship
Program.
Data Availability Statement
The newly curated 100-category Plant Disease Dataset used in this study is publicly available on Kaggle.
Augmented Train Dataset: https://www.kaggle.com/datasets/rithambararajput/augmented-train
Raw Dataset: https://www.kaggle.com/datasets/rithambararajput/100-class-split-raw-dataset
The Plant Village dataset, used as a benchmark for comparative evaluation, is also publicly accessible at:
https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset.Ethical Declarations
This study does not involve human participants or animals. Therefore, ethical approval was not rOpen asset ↗Kaggle · rithambararajput/100-class-split-raw-datasetpdf-raw-page:19 lines:1-51Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Convolutional Neural Networks (CNNs) are widely used for plant disease detection, yet their performance is strongly influenced by hyperparameter selection. Traditional manual tuning or random search approaches are inefficient and may lead to suboptimal solutions. In this study, we propose and evaluate four hybrid metaheuristic strategies, Ant Lion Optimizer combined with Whale Optimization Algorithm (ALO-WOA), Ant Lion Optimizer with Dragonfly Algorithm (ALO-DA), Ant Lion Optimizer with Particle Swarm Optimization (ALO-PSO), and Particle Swarm Optimization with Whale Optimization Algorithm (PSO-WOA), for automatic hyperparameter tuning of CNNs. The methods were applied to a tomato leaf disease dataset comprising 21,421 training, 4,586 validation, and 4,602 test images across 10 classes. The CNN architecture was fixed with three convolutional blocks and a tunable dropout and learning rate. The experimental results show that ALO-DA achieved the highest test accuracy of 97.83%, closely followed by ALO-WOA (97.67%) and PSO-WOA (97.52%), while ALO-PSO achieved 95.26%. These findings demonstrate that hybrid metaheuristics can effectively improve CNN hyperparameter search compared to single optimizers, balancing exploration and exploitation more efficiently. Limitations and future research directions are discussed.
Why it matches plant phenotyping methodsトマト葉の病害状態を画像から分類するCNNについて、ハイブリッドメタヒューリスティックによるハイパーパラメータ最適化手法を開発・評価しており、植物フェノタイピング手法が中心である。
abstractIn this study, we propose and evaluate four hybrid metaheuristic strategies, Ant Lion Optimizer combined with Whale Optimization Algorithm (ALO-WOA), Ant Lion Optimizer with Dragonfly Algorithm (ALO-DA), Ant Lion Optimizer with Particle Swarm Optimization (ALO-PSO), and Particle Swarm Optimization with Whale Optimization Algorithm (PSO-WOA), for automatic hyperparameter tuning of CNNs.
Reproduction assets foundThe paper uses the public Kaggle Tomato Leaf Disease Dataset V2 as its phenotyping image data and publishes the authors' complete hybrid metaheuristic CNN optimization pipeline (training, optimization, evaluation scripts) in a public GitHub repository. Trained models/logs are only available on request.Code · publicThe source code, trained model configurations, and experimental scripts used in this study are publicly available at: https://github.com/simarkalsi24/Hybrid-of-Optimization-Algorithm-.git The repository includes the complete training and optimization pipeline, implementations of all hybrid algorithms (ALO–DA, ALO–PSO, ALO–WOA, and PSO–WOA), as well as experiment configurations, logs, evaluation scripts, and visualization outputs to ensure full reproducibility of the reported results.Open asset ↗GitHub · simarkalsi24/Hybrid-of-Optimization-Algorithm-lines:579-588Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Apple leaf disease segmentation is critical for yield and quality preservation in what is globally one of the most economically significant fruit crops. Despite recent advances in deep learning, real-world orchard environments present three primary challenges: (1) low contrast between lesions and background textures, which hinders accurate localization; (2) leaf overlap and occlusion, leading to incomplete feature representation and increased false negatives; and (3) the inherent limitations of unimodal RGB imagery in capturing subtle pathological features, which constrains generalization and accuracy. To address these issues, we proposed Language-Infused Visual Mamba (LViM), a dual-path U-Net architecture that integrates Mamba and Transformer modules for semantic-visual feature fusion. LViM achieves robust segmentation in complex environments through three core innovations: (1) A U-shaped Multimodal Transformer (MTT) branch integrated with AMBERT, which leverages inter-modal semantic relationships to enhance textual feature extraction and provide high-level semantic cues, thereby improving lesion-background discriminability; (2) a U-shaped Visual State Space (VMamba) branch that employs 2D Selective Scanning (SS2D) and Visual State Space (VSS) blocks to capture global context and fine-grained details, mitigating the impact of occlusion; and (3) Cross-Attention Gate Fusion (CAGF) and Linguistic Cross-Nested (LCN) modules that facilitate efficient cross-modal alignment and hierarchical feature modeling to better identify subtle lesions. Experimental results demonstrate that LViM consistently outperforms the VM-UNet baseline, yielding improvements of 4.05% in Precision, 4.25% in Dice coefficient, 4.49% in mIoU, and 4.23% in Recall.
Why it matches plant phenotyping methodsリンゴ葉の病斑を画像から分割する手法を開発し、複雑な環境での性能を評価しており、植物病害状態の取得・推定が研究の中心である。
abstractApple leaf disease segmentation is critical for yield and quality preservation
Reproduction assets foundThe paper's curated multimodal apple leaf disease dataset (image-text pairs with pixel-level annotations for four disease types) is explicitly stated as publicly released in the authors' LViM GitHub repository. Code/models are only promised 'upon acceptance,' so the dataset asset qualifies as public, while the code is.Dataset · publicThe curated multimodal apple leaf disease dataset constructed in this study has been publicly released at https://github.com/csuft1906ll/LViMOpen asset ↗csuft1906ll/LViMlines:273-283Code / dataset availability confirmedOpenAlex · Europe PMC · checked 15 Sept 2026
Plant leaf disease classification and severity grading are essential for precision agriculture, enabling timely intervention and optimized management. Existing models often fail to recognize previously unseen disease categories due to rigid label spaces and limited representation of plant phenotypes. To address these challenges, we propose a knowledge-driven unified framework for plant disease classification and severity grading. A Common Knowledge Learner consolidates fundamental features of plant species, disease categories, and severity levels from labeled data, forming a transferable representation space. It employs a multi-level contrastive learning strategy to capture both global semantic representations and fine-grained lesion patterns. Building on these representations, a Cross-Domain Adaptation module leverages a teacher-student framework with Low-Rank Adaptation (LoRA) bridges in-domain and out-of-domain feature spaces using large-scale unlabeled data. Meanwhile, a contrastive feature library enables similarity-based reasoning and supports flexible label space expansion during inference without retraining. We evaluate our approach on Leaf-CG, a large-scale dataset comprising 441,448 images from 59 plant species, 373 disease categories, and four severity levels. Experiments demonstrate that our framework outperforms existing baselines, achieving 94.9% disease classification accuracy and 90.6% severity grading accuracy in-domain. Under out-of-domain conditions, the method achieves 82.1% true positive rate (TPR) in open-set settings, highlighting its strong generalization ability and potential for practical plant disease management. Code and dataset are available at https://www.uniplantcg.samlab.cn.
Why it matches plant phenotyping methods植物画像から病害分類と病害重症度を推定する計算・画像ベース手法を開発し、大規模データセットで評価しており、フェノタイピング手法が中心である。
abstractwe propose a knowledge-driven unified framework for plant disease classification and severity grading.
Reproduction assets foundThe paper's Leaf-CG dataset (test subset publicly available), analysis code, and trained model weights (plant.pth, disease.pth, severity.pth) are explicitly released at the authors' site https://www.uniplantcg.samlab.cn. Cited datasets (AI Challenger 2018, PlantVillage, etc.) are prior work, not paper-specific assets.Code · publicving 94.9% disease classification accuracy and 90.6% severity grading accuracy in-domain. Under out-of-domain conditions, the method achieves 82.1% true positive rate (TPR) in open-set settings, highlighting its strong generalization ability and potential for practical plant disease management. Code and dataset are available at https://www.uniplantcg.samlab.cn .
Keywords
Plant disease diagnosis Knowledge-driven learning Domain adaptation pmc-status-qastatus 0 pmc-status-live yes pmc-status-embargo no pmc-status-released yes pmc-prop-open-access yes pmc-prop-olf no pmc-prop-manuscript no pmc-prop-legally-suppressed no pmc-prop-has-pdf yes pmc-prop-has-supplement yes pmc-prop-pdf-onlyOpen asset ↗uniplantcg.samlab.cnlines:1-29Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Litchi is an important economic fruit in southern China, and its precision management relies on the rapid and accurate estimation of the Soil and Plant Analyzer Development (SPAD) values in leaves. Addressing the limitations of existing SPAD detection methods, such as limited rapid coverage, inadequate modeling of dynamic environmental interference, and shallow fusion of multi-source data, this study constructed an Internet of Things (IoT) system to collect real-time environmental data from a litchi orchard, combined with unmanned aerial vehicle (UAV) multispectral imagery to obtain canopy vegetation index and texture features. A Long Short-Term Memory (LSTM) network model integrated with a feature level attention mechanism (MLSTM) was proposed to fuse IoT time-series data, vegetation index, and high dimensional texture features for dynamic SPAD value prediction. The results indicate that multi-source feature fusion significantly improves SPAD estimation accuracy. The MLSTM model achieved optimal performance under the all-features situation, with a coefficient of determination (R²) of 0.897 and a root mean square error (RMSE) of 2.638, outperforming other comparative models. The attention mechanism effectively enhanced the model's focus on key features, improving feature utilization efficiency and model interpretability. The multi-source data fusion method and MLSTM model proposed in this study enable high precision, dynamic estimation of SPAD values in litchi leaves, providing reliable data support for precision fertilization, stress diagnosis, and yield prediction in litchi orchards, as well as theoretical support for promoting the practical application of this technology in smart agriculture.
Why it matches plant phenotyping methodsIoT・UAVマルチスペクトル画像から葉のSPAD値を推定するデータ融合システムとMLSTMモデルを開発・評価しており、植物形質取得手法が研究の中心です。
abstractthis study constructed an Internet of Things (IoT) system to collect real-time environmental data from a litchi orchard, combined with unmanned aerial vehicle (UAV) multispectral imagery to obtain canopy vegetation index and texture features.
Reproduction assets foundThe paper's data availability statement points to a public Zenodo repository containing the study's multi-source SPAD/IoT/multispectral dataset.Dataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://zenodo.org/records/18308090 .Open asset ↗zenodo · 18308090lines:427-441Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Soybean production is significantly affected by crop diseases and improper pesticide use, which hinder effective disease management and reduce yield. In this study, we propose an efficient multi-task convolutional neural network (CNN) framework for the simultaneous detection of soybean seed diseases and pesticide presence from seed images. The model leverages a shared feature extraction backbone with task-specific output heads to learn complementary features for both disease classification and pesticide detection. A dataset of 429 soybean leaf images was preprocessed using normalization and augmentation techniques and split into training, validation, and testing sets. We evaluated three backbone architectures VGG19, MobileNetV3, and ConvNeXt within the multi-task framework. Experimental results demonstrate that the approach maintains computational efficiency suitable for real-world deployment while achieving high performance, with accuracies of 95%, 96%, and 97% for MobileNetV3, VGG19, and ConvNeXt, respectively. Additionally, explainable AI methods, such as Grad-CAM, highlight regions of focus for both tasks, making the model's decision-making process interpretable. This framework provides a practical tool for informed crop management and agricultural monitoring.
Why it matches plant phenotyping methods植物画像から病害状態を推定するCNN手法の開発・評価が研究の中心であり、病害表現型の画像ベース推定に該当する。
abstractwe propose an efficient multi-task convolutional neural network (CNN) framework for the simultaneous detection of soybean seed diseases and pesticide presence from seed images.
Reproduction assets foundThe paper's authors publicly released their custom analysis code (preprocessing, training, evaluation) on GitHub, matching an allowed URL. The enriched Kaggle image/annotation dataset is also public but its URL is not among the allowed URLs, so it is not listed.Code · publicThe custom code developed for this study is publicly available on GitHub at https://github.com/fikaduberie/Soybean-Disease-and-Pest (version v1.0).Open asset ↗fikaduberie/Soybean-Disease-and-Pesthtml-lines:1281-1329Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
WheatGrowth chamberLeafGrowth / time-series analysisGrowth / development / phenology
Disentangling genotype × environment (G×E) controls of flowering time requires phenotypes that link molecular regulation, developmental physiology and environment. Here, we integrated time-resolved measurements of apical development, final leaf number (FLN), and expression of the flowering-time genes VRN1, VRN2 and VRN3 across contrasting temperature and photoperiod regimes in six wheat genotypes spanning a wide range of developmental sensitivities. By combining controlled-environment phenotyping with concurrent gene-expression profiling, we show that environmentally driven variation in FLN is coherently explained by shifts in the timing of key apical transitions and associated VRN gene-expression dynamics. These integrated datasets were used to parameterise and interrogate the Cereal Anthesis Molecular Phenology (CAMP) model, enabling direct comparison between observed foliar gene-expression time courses and modelled gene activity. While overall developmental responses were well captured by the model, systematic differences between observed and modelled gene-expression patterns highlight the importance of distinguishing foliar expression from apical regulatory activity, as well as differences in temporal scaling. Building on this framework, we present a phenotyping protocol based on FLN responses to defined temperature and photoperiod treatments that delivers unconfounded developmental phenotypes explicitly linked to underlying genetic regulation.
Why it matches plant phenotyping methodsFLN応答に基づくフェノタイピングプロトコルを提示し、温度・光周期処理下で遺伝的に解釈可能な発育表現型を取得する方法が中心的に扱われている。
abstractBuilding on this framework, we present a phenotyping protocol based on FLN responses to defined temperature and photoperiod treatments that delivers unconfounded developmental phenotypes explicitly linked to underlying genetic regulation.
Reproduction assets foundThe paper's CAMP model code and the analysis scripts producing its figures are explicitly stated as publicly available on the authors' GitHub repository, directly reproducing this paper's phenotyping analysis.Code · publicwere also validated and the best-performing sets selected. A
348 description of each of the primers used in this study is given in the supplementary material
349 (Table SA1).
350 2.9 Verification of CAMP predictions
351 2.9.1 Model set-up and operation.
352 The CAMP model was coded into a Python script which is available at
353 https://github.com/HamishBrownPFR/CAMP/blob/master/CAMP.ipynb. A formal
354 description of the code and parameterisation scheme is given in the supplementary material.
355 The FLN developmental phenotypes measured for each genotype (Section 3.1) were used to
356 derive the Vrn expression parameters needed for CAMP. Each of the treatments was
357 simulated using CAMP wOpen asset ↗https://github.com/HamishBrownPFR/CAMP/pdf-layout-page:14 lines:1-49Code · publicpression parameters needed for CAMP. Each of the treatments was
357 simulated using CAMP with its corresponding daily temperature and Pp, so its predictions of
358 Vrn gene expression could be compared with those observed. The script running the CAMP
359 code and producing the graphs displayed in this paper can be viewed at
360 https://github.com/HamishBrownPFR/CAMP/blob/master/Tests/CAMPCETests.py.
14
UNOFFICIALOpen asset ↗https://github.com/HamishBrownPFR/CAMP/pdf-layout-page:14 lines:1-49Code · publicnd testing of the model in
690 broader contexts. EW contributed substantially to the improvement of model concepts and the
691 manuscript and all authors provided final checking.
692 8. Data Availability
693 All the data and scripts used to analyse data and produce graphs as well as CAMP model code are
694 publicly available at https://github.com/HamishBrownPFR/CAMP/
695 9. References
696 Allard V, Otto V, Bela K, Rousset M, Le Gouis J, Martre P. 2012. The quantitative
697 response of wheat vernalization to environmental variables indicates that vernalization is not
698 a response to cold temperature. Journal of Experimental Botany 63: 847–857.
699 Baumont M, Parent B, Manceau L, Brown HE,Open asset ↗https://github.com/HamishBrownPFR/CAMP/pdf-layout-page:31 lines:1-60Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Published5 Jun 2026Proceedings of the National Academy of Sciences of the United States of AmericaCited by 0 · OpenAlex ↗
To improve crop yield and resilience, it is essential to identify the steps limiting [Formula: see text] assimilation rate in plant leaves. The combined effect of multiple traits can be resolved by mechanistic models of the underlying diffusion, biochemistry, and geometry. Yet the widely used simple serial resistance models overlook tissue geometry, and detailed anatomical models are computationally heavy and rely on parameters that are difficult to measure. Here, we develop a framework for systematic species and model comparison, and find that the necessary level of model resolution is species-specific. We apply a minimal reaction-diffusion model and reduce [Formula: see text] fixation in leaves to two key parameters. These parameters comprise a compact phase space in which three rate-limiting regimes emerge naturally: stomatal uptake, intercellular diffusion, and intracellular processes. Mapping diverse plant species into this phase space reveals: 1) dominant colimitations by stomatal and intracellular processes, 2) an equal partition between species that require spatially resolved leaf-scale models and species where intracellular models suffice. Taken together, we present a scalable path for interpreting complex trait data and bridging between models.
Why it matches plant phenotyping methods葉のCO2固定を機構モデルで2パラメータに縮約し、複数種の生理的制限状態と複雑な形質データを解釈・比較する計算フレームワークが中心であるため、植物生理形質の推定・解析手法として含める。
abstractHere, we develop a framework for systematic species and model comparison
Reproduction assets foundThe paper deposits its analysis code/scripts publicly on Zenodo (DOI 10.5281/zenodo.19087524) and GitHub (andreas-stillits/CarbonFixationModel), and uses the publicly deposited Knauer et al. leaf-trait/mesophyll-conductance dataset on Figshare (10.6084/m9.figshare.19681410) to map species into (τ, γ) space. All three, Code · publicCode and Scripts. All code is readily available at our github and at a public
repository (DOI: 10.5281/zenodo.19087524).Open asset ↗Zenodo · 10.5281/zenodo.19087524pdf-raw-page:8 lines:1-60Code / dataset availability confirmedOpenAlex · Europe PMC · bioRxiv · Crossref · checked 15 Sept 2026
BarleyCommon beanCowpeaGrowth chamberMesh / voxelLiDAR / point cloudMultispectral / hyperspectralLeafWhole plant / canopy / plot / fieldAnnotation / quality control
Abstract High-throughput 3D multispectral plant phenotyping platforms generate large volumes of point cloud files, but trait extraction is typically performed by sensor-bundled software whose internal algorithms are not publicly documented, which limits reproducibility and integration into custom research pipelines. Here we present PhytoScan3D, an open-source Python pipeline that extracts morphological and spectral phenotypic traits, spanning plant height, 3D leaf area, digital biomass, convex hull volume, leaf inclination, canopy geometry, NDVI, hue, and vegetation indices, from both PLY and PCD point cloud files generated by Phenospex PlantEye F500 and F600 sensors, and is portable to point clouds from any acquisition platform. PhytoScan3D was validated against HortControl (PhenoSpex) ground-truth measurements on 936 barley ( Hordeum vulgare ) pot-date observations from the growth chamber trial (20 Norwegian cultivars, 12 scan dates, Septemenr 2025 to January 2026), achieving Pearson r = 0.913 to 0.999 and ratio approximately 1.000 for Plant Height Max, 3D Leaf Area, and NDVI Average. A vectorised mesh face filtering implementation achieved a 120x speed improvement, increasing valid 3D Leaf Area coverage from 0.6% to 100% of files. Cross-format validation on 223 PlantEye F600 PCD files from the ICRISAT LeasyScan platform (four legume species: mungbean, cowpea, lima bean, and common bean; 1,523 plant observations) yielded r = 0.884 against independent cuboid annotation heights. The systematic positive bias (mean +27.2 mm, ratio = 1.44) is attributable to PhytoScan3D computing height from raw point cloud Z-range while cuboid annotations are fitted to segmented plant points only, with the offset consistent across all four species (per-species r = 0.880 to 0.888). Cross-dataset processing of 1,180 PLY files from the Crops3D benchmark (8 species, 3 acquisition methods) confirmed zero extraction errors. PhytoScan3D is available at “github.com/kovimallik/phytoscan3d” under the MIT licence and processes 1,651 files across three independent datasets in under 12 minutes on GPU hardware. Highlights PhytoScan3D is the first open-source Python pipeline for batch extraction of phenotypic traits, including plant height, 3D leaf area, digital biomass, convex hull volume, leaf inclination, NDVI, and excess green index, from both PLY and PCD point cloud files generated by Phenospex PlantEye sensors. Primary validation against HortControl ground-truth measurements on 936 barley pot-date observations achieved Pearson r = 0.913-0.999 for Plant Height Max, 3D Leaf Area, and NDVI Average. A 120x computational speedup in mesh face filtering (vectorised NumPy vs. set-based loop) increased the coverage of valid 3D Leaf Area extraction from 0.6% to 100% of files. Cross-format validation on 223 PlantEye F600 PCD files from ICRISAT LeasyScan (four legume species, 1,523 plants) achieved r = 0.884 against independent cuboid annotation heights. The systematic +27.2 mm bias reflects a methodological difference (raw Z-range vs. soil-segmented annotations), is consistent and predictable across all four species (per-species r = 0.880-0.888), and is correctable by a single linear factor. Cross-dataset processing of 1,180 PLY files from the Crops3D benchmark (8 species, 3 acquisition methods) confirmed zero extraction errors. Significant scan-unit variation was detected for Plant Height Max (F = 5.71, p < 0.001, η 2 = 0.138) and Canopy Width X (F = 6.32, p < 0.001, η 2 = 0.150), demonstrating the biological utility of extracted traits.
Why it matches plant phenotyping methods植物の3D点群・マルチスペクトルデータから形態・スペクトル形質を抽出するオープンソース手法を開発し、複数データセットで技術検証・ベンチマークしているため、植物フェノタイピング手法が中心である。
abstractHere we present PhytoScan3D, an open-source Python pipeline that extracts morphological and spectral phenotypic traits
Reproduction assets foundThe paper's own analysis code (PhytoScan3D pipeline) is publicly released on GitHub under the MIT licence, and the two external 3D point cloud datasets used for validation (Crops3D and ICRISAT LeasyScan) are publicly available on figshare. The primary barley PLY dataset is not yet public (to be deposited in NVA upon).Code · publicditing, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
Data Availability
PhytoScan3D source code, documentation, and example datasets are available at
https://github.com/kovimallik/phytoscan3d under the MIT licence. The barley PLY dataset
will be deposited in the Norwegian Research Information Repository (NVA) upon acceptance.
The Crops3D benchmark dataset is publicly available at
https://doi.org/10.6084/m9.figshare.27313272 (Zhu et al. 2024). The ICRISAT LeasyScan
dataset is publicly available at https://doi.org/10Open asset ↗github.com/kovimallik/phytoscan3dpdf-raw-page:15 lines:1-36Dataset · publicData Availability
PhytoScan3D source code, documentation, and example datasets are available at
https://github.com/kovimallik/phytoscan3d under the MIT licence. The barley PLY dataset
will be deposited in the Norwegian Research Information Repository (NVA) upon acceptance.
The Crops3D benchmark dataset is publicly available at
https://doi.org/10.6084/m9.figshare.27313272 (Zhu et al. 2024). The ICRISAT LeasyScan
dataset is publicly available at https://doi.org/10.6084/m9.figshare.28270742 (Galba et al.
2025).
Acknowledgements
This work was supported by the PheNo, DLT-Farming and Soil2Milk from Research Council
of Norway and TWIN-NUE from Norwegian University of Life Sciences (NMBU). The
authoOpen asset ↗figshare · 10.6084/m9.figshare.27313272pdf-raw-page:15 lines:1-36Dataset · publicimallik/phytoscan3d under the MIT licence. The barley PLY dataset
will be deposited in the Norwegian Research Information Repository (NVA) upon acceptance.
The Crops3D benchmark dataset is publicly available at
https://doi.org/10.6084/m9.figshare.27313272 (Zhu et al. 2024). The ICRISAT LeasyScan
dataset is publicly available at https://doi.org/10.6084/m9.figshare.28270742 (Galba et al.
2025).
Acknowledgements
This work was supported by the PheNo, DLT-Farming and Soil2Milk from Research Council
of Norway and TWIN-NUE from Norwegian University of Life Sciences (NMBU). The
authors thank Sara Catarina Costa Laranjeira, Min Lin and other NMBU growth facility staff
for plant care and scanning operOpen asset ↗figshare · 10.6084/m9.figshare.28270742pdf-raw-page:15 lines:1-36Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Quantitative disease resistance in plants emerges from complex interactions between host tissues and pathogen growth dynamics, producing a spectrum of phenotypic responses. In plant-fungal interactions, disease is most visibly expressed through lesions that vary in number, size, shape, and color, collectively defining a lesion profile. For Cochliobolus heterostrophus, a fungus causing Southern Corn Leaf Blight of maize (Zea mays ssp. mays), we show that infection on different maize genotypes produces strikingly different lesion profiles. However, it remains unclear whether such macroscopic variation in lesion profiles corresponds to consistent differences in the three-dimensional organization of pathogen colonization within host tissue. We therefore examined variation in the three-dimensional structure of C. heterostrophus-infection networks across host genotypes representing four lesion-profile classes. Using light-sheet microscopy and filament-tracing methods adapted from neuroscience, we developed quantitative metrics to characterize infection network organization, including depth, density, shape, and spatial association with host vascular tissue. In this dataset, network depth was similar across genotypes, whereas network morphology (shape and density), spatial association with vascular bundles, hyphal segment length, and branching frequency varied. Notably, genotypes with similar quantitative resistance levels sometimes exhibited distinct patterns of fungal colonization, suggesting that comparable resistance can arise from different underlying infection dynamics. These findings indicate that lesion profiles may not uniquely predict infection network structure and highlight the utility of three-dimensional network metrics for describing variation that likely reflects multiple underlying host and pathogen processes. This multi-scale framework provides tools for linking macroscopic disease phenotypes with microscopic infection processes in quantitative disease resistance.
Why it matches plant phenotyping methods植物病斑と病原菌感染ネットワークを対象に、ライトシート顕微鏡とトレーシング法を適応し、感染構造を定量化する指標を開発・適用しており、表現型取得法が研究の中心である。
abstractUsing light-sheet microscopy and filament-tracing methods adapted from neuroscience, we developed quantitative metrics to characterize infection network organization, including depth, density, shape, and spatial association with host vascular tissue.
Reproduction assets foundThe paper's Data availability statement explicitly deposits metadata, data, and computer code as Supplementary Files accompanying the open-access publication (Supplementary Materials 1-5, including XLSX datasets and an untyped Supplementary Material 5 likely holding code). These are paper-specific phenotyping assets (eCode · publicMetadata, data, and computer code from this study are available in Supplementary Files included with the publication.Open asset ↗lines:147-204Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
We introduce the Grapevine (Vitis vinifera) Leaf Image Dataset (GVLiD), a carefully curated set of 3,477 annotated images of Grapevine (Vitis vinifera) leaves to catalyze research in computer vision and plant pathology. Whereas the PlantVillage and Hermos datasets, for instance, contain mainly scanned or laboratory-acquired leaves, GVLiD features vineyard in situ images along with detailed metadata (GPS, lighting, weather, and device model) and expert-verified annotations. To measure the reliability of the annotation, label consistency was very high (κ = 0.86-0.92; 95% CI) as assessed by inter- and intra-rater agreement. Besides Indian viticulture, the dataset also aims to support the field of foliar disease detection in precision agriculture and ML benchmarking, which face significant challenges due to variable illumination and natural leaf backgrounds under field conditions. GVLiD is intended to enable worldwide, reproducible, real-world testing of AI systems for crop disease monitoring.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像で注釈化したデータセットであり、植物病害状態の画像ベース表現型測定と再現可能なベンチマークが中心です。
abstractWe introduce the Grapevine (Vitis vinifera) Leaf Image Dataset (GVLiD), a carefully curated set of 3,477 annotated images of Grapevine (Vitis vinifera) leaves to catalyze research in computer vision and plant pathology.
Reproduction assets foundThe paper's grapevine leaf image dataset (GVLiD) is deposited on Mendeley Data, but that URL is not among the allowed URLs. The authors' validation/analysis code (image-quality metrics, metadata-completeness checks, annotation-reliability calculations) is publicly available on GitHub at the allowed URL, with explicit 'Code · publicAll validation scripts (image-quality metrics, metadata-completeness checks, and annotation-reliability calculations) are publicly available in the GVLiD GitHub repository. This ensures full reproducibility of all validation results reported here.
git clone https://github.com/MilindGayakwad/DNNOpen asset ↗https://github.com/MilindGayakwad/DNNhtml-lines:255-349Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Potatoes are one of the major crops eaten in developing countries; however, their production is falling due to various diseases. Early identification and detection of potato leaf diseases play a vital role in improving potato quality and quantity. Existing methods are either computationally resource intensive or lack trust in their decision-making process, which makes them difficult to deploy for real-time potato disease classification and limits its accessibility. To mitigate these limitations, this study proposed an attention-enhanced MobileNetV2 with a squeeze-and-excitation architecture, which balances high accuracy with low computational resources. This method incorporates the strength of MobileNetV2 and Squeeze-and-excitation networks. A total of 2152 images of early blight, late blight, and healthy leafs were obtained from the Kaggle public repository, which are partitioned into 70% training, 20% validation, and 10% testing and were utilized to train, validate, and test the proposed model. The MobileNetV2 backbone is utilized for feature extraction, and then a squeeze-and-attention block is used to recalibrate the feature maps by focusing on important features and suppressing irrelevant ones. Gradient-weighted Class Activation Mapping (Grad-CAM) was implemented to visualize the most relevant region of the leaf for decision-making, which increases model interpretability and user trust. The proposed model achieves a remarkable performance of 99% testing accuracy with 9.41 MB total parameters. The proposed model is suitable for real-time potato leaf disease detection and classification, which can be easily accessible to agricultural stakeholders, including farmers, and contributes to food security.
Why it matches plant phenotyping methodsジャガイモ葉の病害状態を画像から分類する深層学習手法を提案・評価しており、植物表現型（病害状態）の取得・推定が中心である。
abstractthis study proposed an attention-enhanced MobileNetV2 with a squeeze-and-excitation architecture, which balances high accuracy with low computational resources.
Reproduction assets foundThe paper's phenotyping inputs are 2152 potato leaf images (early blight, late blight, healthy) obtained from a public Kaggle repository, explicitly stated as publicly available in the Declarations. No author analysis code is shared (Code Availability: Not applicable), and no trained model checkpoints are released.Dataset · publicAvailability of Data: The datasets generated during and/or analyzed during the current study are
publicly available at https://www.kaggle.com/datasets/faysalmiah1721758/potato-dataset.Open asset ↗Kaggle · faysalmiah1721758/potato-datasetpdf-page:23 lines:1-35Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published1 Jun 2026International Journal of Electrical and Computer Engineering (IJECE)Cited by 0 · OpenAlex ↗
Plant diseases remain a critical challenge in agriculture, causing substantial yield losses and threatening food security. In this work, we propose a hybrid deep feature engineering framework that integrates deep learning-based feature extraction with classical machine learning for accurate plant disease detection. A pretrained vision transformer (ViT) model is employed to extract discriminative features from leaf images, effectively capturing complex spatial relationships. To address the curse of dimensionality, principal component analysis (PCA) is applied, retaining 98% of the variance while reducing feature space complexity. The refined features are then classified using a support vector machine (SVM) optimized through hyperparameter tuning. Experimental results on the bean leaf lesions dataset demonstrate strong performance, achieving 92% accuracy and a weighted F1-score of 0.92. The proposed ViT–PCA–SVM pipeline effectively balances accuracy, computational efficiency, and generalization, making it a promising solution for real-time smart farming applications.
Why it matches plant phenotyping methods葉画像から植物病害状態を推定するViT–PCA–SVM解析パイプラインが研究の中心であり、植物表現型（病斑・病害状態）の画像ベース推定手法に該当する。
titleTransformer-based hybrid classification for plant leaf disease detection using vision transformer, principal component analysis, and support vector machine
Reproduction assets foundThe paper's only qualifying asset is the public Bean Leaf Lesions dataset (leaf images used as phenotyping input for disease classification), explicitly declared in the DATA AVAILABILITY section with a Kaggle URL. No author analysis code, trained models, or checkpoints are released.Dataset · publicI R D O E Vi Su P Fu
Vijayalakshmi S. Abbigeri ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Geetha D. Devanagavi ✓ ✓
CONFLICT OF INTEREST STATEMENT
All authors declare that they have no conflicts of interest.
DATA AVAILABILITY
The data that support the findings of this study are openly available in Kaggle, "Bean leaf lesions
dataset," [Online] at https://www.kaggle.com/datasets/advayprasad/bean-leaf-lesions-dataset.
REFERENCES
[1] Food and Agriculture Organization (FAO), “Climate change fans spread of pests and threatens plants and crops, new FAO study,”
Food and Agriculture Organization (FAO), 2021. https://www.fao.org/newsroom/detail/Climate-change-fans-spread-of-pests-
and-threatens-plants-and-crops-new-FAOOpen asset ↗Kagglepdf-layout-page:7 lines:1-70Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Plant leaf diseases must be detected and treated early to improve crop yield and reduce agricultural losses. However, pixel-level representations and the inability to be read limit the applicability of existing deep learning approaches to the agricultural sector. A graph neural network termed the Attention-Enhanced Graph Neural Network (AE-GNN) may explain and diagnose multi-plant leaf disease. The proposed framework models leaf pictures as a graph with nodes representing discriminative leaf areas and edges representing their spatial connection. Before creating the global context vector and classification, graph features are aggregated, and an attention weighting method is applied to refocus on disease-relevant nodes obscured by less informative background characteristics. Final disease prediction uses a multilayer perceptron classifier. A curated dataset of half-spinach and curry leaf pictures is used to assess the proposed method for fifteen illnesses and their healthy classifications. Grad-CAM-based explainable AI methods make the model predictions' most important areas clearer. The dataset and source code from this work are available on GitHub for reproducibility and openness. Experimental results reveal that the proposed AE-GNN outperforms convolutional neural networks and graph-based models in classification. Graph-structured learning, attention enhancement, and explainability create a robust and interpretable framework for multi-plant leaf disease diagnosis.
Why it matches plant phenotyping methods葉画像から植物病害状態を分類・診断する画像解析手法を提案し、既存モデルとの比較評価と説明可能性解析を行っているため、植物フェノタイピング手法が中心である。
abstractA graph neural network termed the Attention-Enhanced Graph Neural Network (AE-GNN) may explain and diagnose multi-plant leaf disease.
Reproduction assets foundThe paper's Data availability section explicitly links a public GitHub repository containing the paper's spinach/curry leaf fungal disease image dataset used for the AE-GNN phenotyping/classification analysis.Dataset · publicData availability
The dataset is available at the link below. https://github.com/MeganathanE1990/FINAL-DISEASE-DATA-SET/tree/mainOpen asset ↗MeganathanE1990/FINAL-DISEASE-DATA-SETlines:413-463Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
This article presents a multispectral imaging dataset dedicated to training a machine learning algorithm for the in situ detection of Huanglongbing (HLB). HLB, also known as citrus greening disease, is a major pathology caused by the bacterial pathogen Candidatus Liberibacter asiaticus , particularly in species of the citrus genus. The dataset is constituted of terrestrial images acquired in a commercial sweet orange orchard of the variety Pera Rio ( Citrus sinensis (L.) Osbeck). The images describe large portions of canopy, with healthy leaves and sections infected by HLB as well as some confounding factors naturally present in orchards. Multispectral images were acquired with a multi-lens camera within the visible-near-infrared domain, resulting in 14 narrow spectral bands. The image acquisition was conducted during two field campaigns in 2023 and 2024. In total, the dataset contains 2,978 images divided into two classes HLB (1,681) and non-HLB (1,297). Originally, data are stored in TIFF format as 14 monochromatic images, organised by spectra band. Additionally, an HDF5-format version is provided, where images are stored as 3D arrays with spectral bands in ascending order. This format is compatible with various programming languages, enables efficient data handling, and is optimised for machine learning and image processing applications, supporting reproducible and portable analysis. This dataset is a valuable resource for the development and benchmarking of classification models, including deep learning approaches, aimed at the detection of HLB. Phytopathology imaging datasets are scarce yet essential for advancing digital agriculture and the development of robust tools for crop disease detection worldwide.
Why it matches plant phenotyping methods柑橘葉・樹冠のマルチスペクトル画像からHLB感染状態を推定するデータセットであり、植物病害状態の表現型取得と機械学習ベンチマークを中心とする。
abstractThis article presents a multispectral imaging dataset dedicated to training a machine learning algorithm for the in situ detection of Huanglongbing (HLB).
Reproduction assets foundThe article is a Data in Brief describing a public multispectral HLB citrus image dataset deposited on Data INRAE (Recherche Data Gouv, DOI 10.57745/054NAB), plus an authors' GitHub repository with preprocessing, registration, and model training scripts. Both are paper-specific, public, and directly actionable.Dataset · publicData accessibility
Repository name: Data INRAE
Data access link: https://doi.org/10.57745/054NABOpen asset ↗Data INRAE · 10.57745/054NABhtml-lines:89-123Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
MaizeTomatoLeafPhysiological trait estimationCalibration / preprocessingWater status / transpiration
Within the soil-plant-atmosphere continuum, water movement is driven by the water potential gradients between these three domains. To have a comprehensive understanding of such water relations, an examination of how plants respond to variations in soil water availability is required. The methodologies employed for measuring water potential in leaf (Ψ leaf ) and soil (Ψ soil ) have undergone a significant evolution; transitioning from qualitative assessments to the use of high-precision digital sensors over the past few decades. The present protocol aims to provide a comprehensive, step-by-step guide from the germination phase of maize and tomato plants to the installation of two sensors that continuously monitor water potential in the leaf (PSY1 psychrometer) and in the soil (TEROS 21 matric potential sensor). Additionally, we present the code for processing the raw data files in RStudio.
Why it matches plant phenotyping methods葉の水ポテンシャルを連続測定するセンサー設置、データ処理コード、手順を中心とした植物生理形質の測定プロトコルであり、方法論的貢献が明確。
abstractThe present protocol aims to provide a comprehensive, step-by-step guide from the germination phase of maize and tomato plants to the installation of two sensors that continuously monitor water potential in the leaf (PSY1 psychrometer) and in the soil (TEROS 21 matric potential sensor).
Reproduction assets foundThe paper deposits its authors' R analysis notebook with an example water-potential dataset, the CR800 datalogger program, and an installation video on Zenodo, all publicly accessible.Code · publicthat were missing, zero, or otherwise aberrant. It was also programmed to identify and remove inverted day-night cycle patterns, as well as values that were statistically insignificant.
Figure 9 shows applications of data cleaning on the example dataset. For more details, please check codes that have been deposited on Zenodo (
https://doi.org/10.5281/zenodo.20080750 ,
D’Agostino, 2026 ).
Figure 9.
Example of data cleaning using the algorithm.
Green is kept data and red is discarded data.
Conclusion
In summary, the present protocol is not confined to the descriptive monitoring of Ψ
soil
and Ψ
leafOpen asset ↗Zenodo · 10.5281/zenodo.20080750lines:452-504Code · public(1) the address of each Teros 21; (2) the data transporting port (“C1” or “C3”); (3) the creation of dataset files to store the recorded soil matric potential and temperature, as well as the voltage of the battery for power supply; (4) the time interval for the data recording.
An example of the program was deposited on Zenodo (
https://doi.org/10.5281/zenodo.17158115 ), with the document name of “Program-CR800”). Before starting, install the software of “Device Configuration Utility” and “PC400” from Campbell Scientific (
https://www.campbellsci.com/devconfig ;
https://www.campbellsci.com/pc400 ). “CRBasic Editor” is integrated inside PC400. For more details about the programming, please reOpen asset ↗Zenodo · 10.5281/zenodo.17158115lines:321-378Dataset · publiculic limitation, soil-root disconnection, and recovery. Consequently, this linkage of the protocol to mechanistic analyses of water transport in the SPAC is more direct.
Ethics and consent
Ethical approval and consent were not required.
Data availability
The datasets and codes to analyze the data have been deposited on Zenodo (
https://doi.org/10.5281/zenodo.20080750 ,
D’Agostino (2026) ).
Data are available under the terms of the Creative Commons Zero v1.0 Universal.
An additional explicative video for the psychrometer installation on leaves is available on Zenodo (
https://doi.org/10.5281/zenodo.17510720 ,
Degand
et al. (2025) ).
The author(s) declare that this video is released under theOpen asset ↗Zenodo · 10.5281/zenodo.20080750lines:505-651Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Plant leaf disease detection is a critical task in precision agriculture, where reliable diagnosis under real-world conditions is essential for reducing crop losses and supporting timely intervention. Although deep learning models have achieved high classification accuracy, their performance often degrades under domain shift between controlled laboratory datasets and real-field environments, while predictive uncertainty and confidence calibration remain largely unaddressed.This study presents an uncertainty-aware cross-domain evaluation framework based on a Hierarchical Vision Transformer (HViT) for plant leaf disease classification. The framework integrates multi-scale feature learning with Monte Carlo Dropout-based predictive uncertainty estimation and temperature-based calibration to systematically analyze model behavior in terms of accuracy, reliability, and robustness. Experiments were conducted on two complementary datasets: the New Plant Diseases Dataset (controlled conditions) and the PlantDoc dataset (field conditions), enabling bidirectional cross-domain evaluation. Results demonstrate that the proposed framework achieves superior performance, attaining 97.8% accuracy on controlled data and 93.6% on field data, while significantly improving calibration with lower Expected Calibration Error (ECE = 0.032 / 0.041), reduced Negative Log-Likelihood, and lower Brier score compared to baseline CNN and transformer models. Furthermore, the framework exhibits improved robustness under domain shift, with reduced performance degradation and stable uncertainty behavior. Overall, this study highlights the importance of integrating uncertainty estimation and calibration within a hierarchical transformer-based framework, providing a more reliable and deployment-ready solution for real-world agricultural disease diagnosis.
Why it matches plant phenotyping methods植物葉の病害状態を直接推定する不確実性-aware分類フレームワークの開発・評価が中心であり、異なる条件のデータセット間で精度、校正、頑健性を検証している。
abstractThis study presents an uncertainty-aware cross-domain evaluation framework based on a Hierarchical Vision Transformer (HViT) for plant leaf disease classification.
Reproduction assets foundThe paper's Data availability statement explicitly links the two public image datasets used for its cross-domain plant leaf disease classification experiments: the New Plant Diseases Dataset on Kaggle and the PlantDoc dataset on Dataset Ninja. No author analysis code, models, or checkpoints are reported as available.Dataset · publicThe New Plant Diseases Dataset can be obtained from Kaggle at [https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset]Open asset ↗Kaggle · vipoooool/new-plant-diseases-datasetlines:360-398Dataset · publicThe PlantDoc dataset is available for download at [https://datasetninja.com/plantdoc#download]Open asset ↗plantdoclines:360-398Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Cotton production is highly vulnerable to foliar diseases and pest-induced damage, which significantly reduce yield and compromise fiber quality. Rapid, reliable, and automated disease identification is therefore essential for supporting sustainable crop management. In this study, we propose a hybrid deep learning framework integrating a ResNet50 backbone with Squeeze-and-Excitation (SE) channel attention modules to enhance discriminative feature representation for cotton leaf disease classification. The model is trained on a publicly available disease dataset comprising six classes and optimized using Weighted CrossEntropyLoss, Adam optimization, ReduceLROnPlateau scheduling, and Early Stopping to ensure stable convergence and robust generalization. Experimental results demonstrate outstanding performance, achieving 99.72% training accuracy and 99.31% validation accuracy, with convergence at the 14th epoch. Visualization through Grad-CAM reveals that the model focuses on biologically relevant symptom regions, thereby enhancing interpretability and supporting expert validation. Comparative analysis with state-of-the-art methods shows that the proposed model surpasses existing CNN, transfer learning, and hybrid architectures in both accuracy and model transparency. These results indicate that the proposed SE-ResNet50 framework offers a highly accurate, interpretable, and computationally efficient solution suitable for real-world cotton disease monitoring and precision agriculture applications.Clinical trial registrationThis study is not a clinical trial; therefore, clinical trial registration is not applicable.
Why it matches plant phenotyping methods綿花葉の病徴を画像から分類する深層学習フレームワークの開発が中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として適格です。
titleA hybrid SE-ResNet50 deep learning framework for high-accuracy and explainable cotton leaf disease classification.
Reproduction assets foundThe paper's Data Availability statement points to the public Kaggle cotton plant disease dataset used for training the SE-ResNet50 model. No author analysis code or trained model checkpoint is explicitly deposited.Dataset · public“Cotton plant disease.” Accessed: Nov. 21, 2025. [Online]. Available: https://www.kaggle.com/datasets/dhamur/cotton-plant-disease.Open asset ↗Kaggle · dhamur/cotton-plant-diseasehtml-lines:458-493Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
Plant diseases pose a major threat to global food security, significantly reducing agricultural yields. Therefore, timely diagnosis of plant diseases can help prevent food losses and support economic stability. This study explores the use of eight Convolutional Neural Networks and two Vision Transformers for apple leaf disease diagnosis. The feature extraction layers of each model were modified to incorporate DropBlock layers, while preserving pretrained weights from the ImageNet dataset. Images from the Plant Pathology 2021 dataset were used to fine-tune the models for multi-label classification, targeting five disease categories and a healthy label. Three experiments were conducted to evaluate model performance on the test set. First, the ResNet50 model was used to determine optimal Dropout and DropBlock probabilities. Second, these parameters were applied across all models to identify those with the best performance. Finally, twenty-three Swarm Optimization Algorithms were used to optimize classifier thresholds, improving accuracy and F1-scores. A DropBlock probability of 0.05 and a Dropout probability of 0.2 yielded superior results. Among the models, SwinV2T attained an accuracy of 90.7%, while SwinV2S achieved the highest F1-score of 91.7%, slightly outperforming the ConvNeXtT and ConvNeXtS architectures. The results demonstrated the effectiveness of DropBlock regularization and optimized classifier thresholds, highlighting the superior performance of recent architectures and optimization algorithms over their older counterparts. These findings suggest that such networks hold substantial promise for accurately identifying and diagnosing apple leaf diseases.
Why it matches plant phenotyping methodsリンゴ葉画像から病害状態を分類・診断する深層学習手法を比較評価し、正則化や閾値最適化による性能改善も検証しており、植物フェノタイピング手法が研究の中心である。
abstractThis study explores the use of eight Convolutional Neural Networks and two Vision Transformers for apple leaf disease diagnosis.
Reproduction assets foundThe paper's apple leaf disease phenotyping is based on the public Plant Pathology 2021 (FGVC8) Kaggle image dataset and an authors' reorganized multi-label version publicly deposited on GitHub; both are explicitly linked in the Data availability statement. No author analysis code or trained model checkpoints are statedDataset · publicFor this research, the dataset was reorganized and extended into a multi-label format. The complete modified dataset is publicly available at: https://github.com/soroushtou/Plant-Pathology-2021---MultiLabel-Dataset.Open asset ↗Plant-Pathology-2021---MultiLabel-Datasetlines:239-262Dataset · publicThe original dataset used in this study is the publicly available Plant Pathology 2021 dataset from the FGVC8 competition available at: https://www.kaggle.com/competitions/plant-pathology-2021-fgvc8.Open asset ↗lines:239-262Code / dataset availability confirmedEurope PMC · OpenAlex · checked 13 Sept 2026
Among carnivorous plants, the Venus flytrap (Dionaea muscipula) is known for its rapid (<1 s) trap closure. Although buckling instability, hydrostatic pressure, and hydroelastic coupling have all been proposed to be involved, the nature of this process and the relationship between trap size and curvature remain elusive. Here, we monitored the closure of Venus flytraps and performed micro-CT scanning and 3D reconstruction, revealing that increasing angular velocity was correlated with higher values of a non-dimensional shape index. Based on these experimental data, we constructed a geometric model of the trap that takes leaf orientation into account. We found that leaf curvature is dependent on leaf size, a relationship we denote as a size-curvature constraint. We further propose a curvature design derived from differential deformations of a two-layer model of the leaf, which could be a powerful tool to control the curvatures of soft and bending surface structures in the field of biomimetics.
Why it matches plant phenotyping methodsマイクロCTと3D再構成で葉の閉鎖運動・曲率を定量化し、幾何モデルでサイズ–曲率関係を推定することが研究の中心であり、植物形態・運動状態のフェノタイピング手法に該当する。
abstractHere, we monitored the closure of Venus flytraps and performed micro-CT scanning and 3D reconstruction, revealing that increasing angular velocity was correlated with higher values of a non-dimensional shape index.
Reproduction assets foundThe paper's Data Availability statement points to an authors' GitHub page hosting all data files and related rendering files for the Venus flytrap closure measurements and 3D reconstructions, matching an allowed URL.Dataset · publicAll data files and related rendering files are available from the github ( https://satorutsugawa.github.io/flytrap_geometric_model_datashare/) .Open asset ↗githublines:105-144Code / dataset availability confirmedEurope PMC · OpenAlex · checked 5 Sept 2026
The vertical heterogeneity of rice canopy structure limits the accuracy of inverting leaf physicochemical parameters using traditional radiative transfer models, while LiDAR-based 3D reconstruction remains costly for large-scale applications. To address these challenges, this study proposes a method for constructing 3D rice canopy scenes using "Precision Mode" and "Rapid Mode" strategies. The Precision Mode builds detailed structural models based on measured morphological parameters, validated via the LESS 3D radiative transfer model. To overcome the limitations of obtaining detailed morphology via UAV remote sensing, the Rapid Mode employs machine learning algorithms-specifically Support Vector Machine (SVM), Random Forest (RF), and XGBoost-to map easily accessible parameters (LAI, Above-ground Biomass, Plant Height, and Transplanting Date) to detailed 3D structural parameters. Results indicate that XGBoost achieves the highest accuracy in the Rapid Mode. Furthermore, simulated spectra under both modes showed high consistency with measured spectra, yielding average RMSE values of 0.0104 (R 2 = 0.9965) for the Precision Mode and 0.0307 (R 2 = 0.9694) for the Rapid Mode. Although the spectral accuracy of the Rapid Mode is slightly lower, its modeling efficiency is significantly enhanced, retaining a strong capability to reproduce spectral response characteristics across growth stages. This approach provides an effective tool for analyzing vertical spectral response mechanisms and offers an efficient data simulation scheme for UAV remote sensing parameter inversion based on 3D radiative transfer models.
Why it matches plant phenotyping methodsイネ群落の3D構造を構築・推定する手法を開発し、放射伝達モデルと実測スペクトルで検証しており、植物形質の取得・再現が研究の中心です。
abstractthis study proposes a method for constructing 3D rice canopy scenes using "Precision Mode" and "Rapid Mode" strategies.
Reproduction assets foundThe paper's Data Availability statement says the collected phenotype/structural/spectral data are publicly available on the authors' GitHub repository (allowed URL), while the analysis code is only available from the corresponding author upon request (request_only, no public URL).Dataset · publicThe data collected and used in this study are publicly available at: https://github.com/baijc4095-code/2024data . The code used for analysis can be obtained from the corresponding author upon reasonable request.Open asset ↗baijc4095-code/2024datalines:240-256Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published23 May 2026Informatika: Jurnal Teknik Informatika dan MultimediaCited by 1 · OpenAlex ↗
Early identification of Rice leaf diseases remains a challenge in agricultural practices, as detection is commonly performed through manual visual observation that is time-consuming and prone to misclassification. Diseases such as blast, Bacterial Leaf Blight, tungro, and Brown Spot often exhibit similar visual characteristics, particularly at early stages. To address this problem, an Android-based application was developed to classify Rice leaf diseases using a Convolutional Neural Network (CNN) with a transfer learning approach based on the MobileNetV3 architecture. The model was trained using a labeled Rice leaf image Dataset obtained from Hugging Face, with preprocessing and data augmentation applied to improve generalization performance. The trained model was deployed through Hugging Face Space using an API-based architecture, allowing image classification to be performed without heavy computational requirements on mobile devices. Experimental results demonstrate that the proposed model achieved an accuracy of approximately 90% on the testing Dataset, exceeding the predefined minimum target accuracy of 85%, with precision and recall values above 80% across all disease classes based on confusion matrix evaluation. These results indicate that the MobileNetV3-based transfer learning approach provides reliable classification performance with good computational efficiency, making it suitable for mobile-based Rice leaf disease detection applications.
Why it matches plant phenotyping methodsイネ葉の画像から病害状態を分類するCNN手法を開発・評価し、モバイル実装まで行っているため、植物病害フェノタイピング手法が中心です。
abstractan Android-based application was developed to classify Rice leaf diseases using a Convolutional Neural Network (CNN) with a transfer learning approach based on the MobileNetV3 architecture
Reproduction assets foundThe paper's plant-phenotyping input is a public labeled rice leaf disease image dataset obtained from Hugging Face (girish787/riceLeafDataset), used to train the MobileNetV3 classifier. No author analysis code or trained model checkpoint is explicitly deposited.Dataset · public[12] G. Kumar, “riceLeafDataset.” Apr. 25, 2024. Accessed: Oct. 20, 2025. [Online]. Available:
https://huggingface.co/Datasets/girish787/riceLeafDatasetOpen asset ↗huggingface.co/Datasets/girish787/riceLeafDataset · girish787/riceLeafDatasetpdf-page:10 lines:1-44Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
Abstract Chlorophyll estimation is fundamental in plant physiology, crop management, and ecological studies; however, destructive and non-destructive methods are often interpreted interchangeably despite differing measurement principles. The present study compared four chlorophyll estimation approaches—two non-destructive (SPAD meter and GreenSeeker) and two destructive (80% acetone and DMSO extraction)—across eight crop species under uniform field conditions. Significant interspecific variation was observed for all methods. Correlation and regression analyses revealed generally weak relationships among methods, particularly between leaf-level (SPAD, solvent extraction) and canopy-level (GreenSeeker) measurements, reflecting scale-dependent behavior and methodological differences. Moderate associations were observed between SPAD and acetone-extracted chlorophyll for certain traits, whereas GreenSeeker showed poor agreement with solvent-based estimates. Differences between DMSO and acetone extraction further highlighted solvent-specific extraction efficiency. The results demonstrate that chlorophyll estimation methods are not directly interchangeable and should be selected based on study objectives, biological scale, and leaf anatomical characteristics. Species-specific calibration and integration of canopy structural parameters are required to improve cross-method interpretability.
Why it matches plant phenotyping methods複数の葉・キャノピーのクロロフィル推定法を作物種間で比較し、相関、回帰、スケール依存性、互換性を評価しており、植物表現型測定法の技術的検証が中心である。
abstractThe present study compared four chlorophyll estimation approaches—two non-destructive (SPAD meter and GreenSeeker) and two destructive (80% acetone and DMSO extraction)—across eight crop species under uniform field conditions.
Reproduction assets foundThe preprint declares that the datasets generated in this chlorophyll-method comparison study (SPAD, GreenSeeker, acetone and DMSO measurements across eight crop species) are publicly deposited in Figshare under DOI 10.6084/m9.figshare.31817989. This is a paper-specific, publicly actionable phenotype dataset. No authorDataset · publicThe datasets generated during the current study are available in the Figshare repository, https://doi.org/10.6084/m9.figshare.31817989Open asset ↗Figshare · 10.6084/m9.figshare.31817989lines:163-185Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Tomato diseases pose a significant threat to global agricultural production, often leading to substantial yield loss and major economic damage. Traditional disease detection methods rely on manual inspection, which is not only time-consuming and labor-intensive but also difficult to implement for real-time monitoring. While deep learning-based object detection techniques offer a potential alternative to manual inspection, existing models still face challenges in extracting subtle disease features, suppressing complex background interference, and in handling multi-scale disease representations in complex agricultural environments, limiting detection performance. To address these limitations, this paper proposes a novel TDD-YOLO model for precise tomato-disease detection (TDD) in complex agricultural settings. The proposed model is based on YOLOv11 with the following three main improvements: (1) a feature enhancement module is added to improve the backbone's ability to extract disease spot textures; (2) a joint attention mechanism is introduced to explicitly model cross-dimensional dependencies, effectively suppressing background interference; and (3) a feature fusion module is added to retain disease information across different scales while reducing computational costs. Experimental results, obtained on the Tomato-Village dataset (containing field-acquired images of tomato leaves with six diseases, collected in real agricultural environments, featuring complex backgrounds and varying illumination conditions) and Tomato-Disease dataset (emphasizing a greater diversity in tomato disease types along with healthy leaf samples), demonstrate that the proposed TDD-YOLO model outperforms the baseline in detection of tomato diseases (e.g., by improving mAP@50 and mAP@50:95, averaged across disease categories, by 4.1% and 6.0% on Tomato-Village and by 3.6% and 3.9% on Tomato-Disease, respectively) and state-of-the-art models (e.g., by improving the average mAP@50 and mAP@50:95, compared to the first runner-up, by 3.2% and 4.7% on Tomato-Village and by 2.4% and 2.1% on Tomato-Disease, respectively), while maintaining good parameter count and computational complexity, confirming its effectiveness and potential for practical usage in complex agricultural environments. The author-generated code and weight files are publicly available at https://github.com/LingShaQ/TDD-YOLOCode.
Why it matches plant phenotyping methodsトマト葉の病斑・病害状態を画像から検出するYOLOモデルを開発し、複数データセットでベースラインおよび既存モデルと比較検証しており、植物病害フェノタイピング手法が中心である。
abstractExperimental results, obtained on the Tomato-Village dataset
Reproduction assets foundThe paper's tomato-disease detection experiments rely on two public image/annotation datasets (Tomato-Village on GitHub, Tomato-Disease on Zenodo), and the authors explicitly state their generated code and weight files are publicly available on GitHub. The Ultralytics YOLO repositories are generic third-party librariesCode · publicThe author-generated code and weight files are publicly available at https://github.com/LingShaQ/TDD-YOLOCode.Open asset ↗LingShaQ/TDD-YOLOCodehtml-lines:110-113Dataset · publicAll data used in this article are obtained from the publicly available Tomato-Village dataset (https://github.com/mamta-joshi-gehlot/Tomato-Village)Open asset ↗mamta-joshi-gehlot/Tomato-Villagehtml-lines:1159-1171Dataset · publicthe publicly available Tomato-Disease dataset (https://zenodo.org/records/15868289).Open asset ↗html-lines:1159-1171Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Background Leaf-level biogenic volatile organic compounds (BVOCs) emissions represent a major source of organic gases in the atmosphere, influencing both climate and air quality. These emissions are strongly driven by environmental perturbations, which affect individual plant- to ecosystem-level processes. Uncovering all the BVOCs and understanding how their emissions respond to altered environmental conditions provide critical insights into vegetation-driven changes in atmospheric chemistry. We developed a tandem instrumentation setup that integrates a proton transfer reaction time-of-flight mass spectrometer (PTR-ToF-MS) with parts-per-trillion detection limits and a photosynthetic infrared gas exchange system for the untargeted survey of all the BVOCs. This novel system enables simultaneous, real-time monitoring of BVOC emissions and photosynthetic parameters at the leaf level, offering new opportunities to disentangle the physiological and environmental drivers of VOC release. Furthermore, we established the VOC Analysis and Processing Optimization Resource (VAPOR), an open-access software tool designed for rapid data post-processing and the analysis of the variability of hundreds of BVOCs. We assessed the performance of the tandem system under varying background conditions, using standard gas mixtures and a range of environmental factors. Results Blank emissions were substantially lower for major BVOCs (e.g., isoprene) compared to those observed in plant emissions. Despite this, the observation of background-level VOCs highlights the importance of routinely acquiring and accounting for blank measurements in analyses using the coupled instrumentation. Introduction of known VOC concentrations to the system demonstrated a linear response across different compounds with varying molecular compositions, indicating minimal gas loss regardless of chemical moieties within the coupled instrumentation. We applied the optimized system to investigate the physiological mechanisms driving BVOC emissions across different genotypes of poplar and pennycress. The high mass resolution capabilities of the PTR-ToF-MS, coupled with comprehensive VAPOR-driven data analysis, enabled the identification of several important BVOCs, including methanol and methanethiol; these BVOCs displayed substantial variation across pennycress genotypes and showed concentrations ~ 100-350% higher than the blank. Moreover, isoprene emissions varied significantly among poplar genotypes grown in different potting media. Conclusions Tandem instrumentation offers a powerful tool for profiling volatile molecular markers and elucidating their genetic and environmental underpinnings. This approach enhances our ability to predict BVOC emissions in response to genotype by environmental interactions and contributes to a deeper understanding of vegetation responses to environmental changes.
Why it matches plant phenotyping methods葉レベルの植物揮発性物質排出と光合成パラメータを取得するタンデム計測系を開発・検証し、解析ソフトウェアも提供しているため、植物表現型取得法が中心である。
abstractWe developed a tandem instrumentation setup that integrates a proton transfer reaction time-of-flight mass spectrometer (PTR-ToF-MS) with parts-per-trillion detection limits and a photosynthetic infrared gas exchange system for the untargeted survey of all the BVOCs.
Reproduction assets foundThe paper's authors developed VAPOR, an open-access software tool used to post-process and analyze the paper's leaf VOC emission measurements, with explicit public availability at the authors' GitHub repository.Code · publicThe open-source code for VAPOR is accessible at https://github.com/INTERSECT-BESS/ORNL-VOC . In this study, VAPOR was used to post-process the VOC results generated from the offline collection of gases from poplars with different soil media.Open asset ↗INTERSECT-BESS/ORNL-VOClines:127-146Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Introduction Confidence calibration, selective prediction, out-of-distribution scoring, and deep ensembles are mature techniques in machine learning, yet their efficacy under the severe domain shift encountered when plant disease classifiers move from controlled laboratory imagery to heterogeneous field photographs has not been systematically benchmarked. Methods Models trained on PlantVillage were evaluated on PlantDoc leaf-level crop images under a parent-image-aware split protocol, and a suite of standard mitigation techniques was applied to characterize the reliability gap. Analyses included temperature scaling and selective prediction for a fine-tuned ResNet-50, quantitative image-level shift analysis, Grad-CAM visualization, simple target-aware adaptation baselines, frozen-feature backbone comparisons, and ensemble baselines. Results In the primary case study, a fine-tuned ResNet-50 suffered a 67.7-percentage-point accuracy collapse upon cross-domain transfer, while mean predicted confidence remained at 79.76%. Post-hoc temperature scaling reduced calibrated ECE to 0.3645 but left selective risk at 80% coverage at 64.30%. Quantitative image-level shift analysis confirmed large-effect-size differences in saturation ( d = 3.90), border edge density ( d = 3.33), and foreground-occupancy proxy ( d = 2.48) between the two domains, while Grad-CAM visualizations showed that the model shifts attention from lesion-centered regions in PlantVillage to background-dominated areas in PlantDoc. Simple target-aware mitigations, including adaptive batch normalization and feature moment matching, improved accuracy from 0.321 to 0.343 and 0.366, respectively, whereas DANN-style adversarial adaptation degraded performance to 0.252. A frozen-feature backbone comparison across five backbones showed that, within the energy-scoring frozen-backbone comparison, DINOv2-S/14 achieved the highest unknown-detection AUROC (0.764) and the lowest selective risk at 80% coverage (0.520), with paired Wilcoxon tests confirming statistically significant accuracy and macro-F1 differences across backbones. Two ensemble baselines were evaluated: a warm-start end-to-end ResNet-50 ensemble reduced calibrated ECE to 0.063 but achieved only 0.666 AUROC, while a lightweight DINOv2 linear-probe ensemble achieved 0.779 AUROC after calibration but under limited epistemic diversity. Discussion Neither ensemble established deployment-grade reliability: the best selective risk at 80% coverage across all configurations remained above 0.51. The principal contribution is a reproducible, deployment-oriented reliability characterization showing that standard post-hoc and lightweight adaptation techniques reduce but do not eliminate the severe reliability gap under controlled-to-field transfer in agricultural computer vision.
Why it matches plant phenotyping methods植物病害画像分類の信頼性・ドメインシフト・校正・選択的予測を体系的にベンチマークしており、病害状態を画像から推定する方法の技術評価が中心である。
abstracttheir efficacy under the severe domain shift encountered when plant disease classifiers move from controlled laboratory imagery to heterogeneous field photographs has not been systematically benchmarked.
Reproduction assets found本文中に内容が明示された植物フェノタイピング関連の補足表と、その公開リンクを確認しました。Supplement · publicSupplementary Table 1 ) was therefore constructed by normalizing all labels to a canonical Crop_Disease format and retaining only those categories for which an unambiguous semantic match existed in both datasets.Open asset ↗lines:335-337Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published22 May 2026JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer)Cited by 0 · OpenAlex ↗
Plant leaf diseases are a major factor in reducing agricultural productivity, particularly for local commodities that often lack adequate artificial intelligence-based disease detection systems. This study aims to optimize the performance of a Convolutional Neural Network (CNN) model using the Inception V3 architecture through the application of multi-stage data augmentation to improve the classification accuracy of local plant leaf diseases. The dataset used is PlantifyDR from Kaggle, which has limited data volume and visual variation, requiring an effective augmentation strategy to improve the model's generalization ability. The proposed multi-stage augmentation approach consists of three stages—geometric, photometric, and texture-noise augmentation—that systematically enrich the diversity of training images. Evaluation results show that the proposed model provides significant performance improvements compared to the baseline model. The Inception V3 model with multi-stage augmentation achieved an accuracy of 0.762, an F1-score of 0.727, and a perfect AUC (1.00) across all classes, while the baseline model only achieved an accuracy of 0.595 and an average AUC of 0.877. Accuracy, loss, ROC curve, and confusion matrix analyses confirmed that multi-stage augmentation reduced overfitting and enhanced the model's ability to differentiate disease symptoms across leaf types. Therefore, this study concludes that multi-stage data augmentation is an effective approach for optimizing deep learning models on small and complex datasets, while also providing a significant contribution to the development of more accurate and reliable AI-based plant disease detection systems.
Why it matches plant phenotyping methods葉画像から植物病害症状を分類するCNNと、性能改善のための多段階データ拡張を開発・評価しており、植物の病害状態を推定する画像ベースのフェノタイピング手法が中心です。
abstractThis study aims to optimize the performance of a Convolutional Neural Network (CNN) model using the Inception V3 architecture through the application of multi-stage data augmentation to improve the classification accuracy of local plant leaf diseases.
Reproduction assets foundThe paper's plant leaf disease classification experiments use the publicly available PlantifyDR Kaggle dataset (Apple, Berry, Guava leaf images), which is the image input for the study's phenotyping measurements. No author analysis code or trained model checkpoints are reported as publicly deposited.Dataset · publice: (Research Results, 2025)
Figure 1. Sample Image from the Research Dataset
Figure 1 shows a sample of the research data.
The dataset used in this study was obtained from
the open dataset platform Kaggle under the title
"PlantifyDR Dataset," provided by Lavaman151.
The dataset is publicly accessible through the
following link:
https://www.kaggle.com/datasets/lavaman151/pl
antifydr-dataset. This dataset is a collection of plant
leaf images from several species, including Apple,
Berry, and Guava, categorized based on leaf health,
making it relevant for plant disease classification
research.
Comparison of the Baseline Model with the
Proposed Model (Multi-Stage Data
Augmentation)
This subsectioOpen asset ↗lavaman151pdf-raw-page:4 lines:1-150Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
The research proposes Cross Disease Similarity Awareness Learning (CDSAL), a robust multiclass tomato leaf disease detection framework based on high-quality and explainable deep learning. The approach solves the problem of superimposed patterns of disease especially Leaf Miner, Tomato Spotted Wilt Virus (TSWV), and nutrient deficiencies through the combination of multi-domain feature learning and inter-disease similarity modeling. In contrast to conventional metric learning or contrastive learning methods that function on pairwise or triplet sample associations, CDSAL develops a class-level Cross Disease Similarity Matrix that represents structured inter-disease proximity within the embedding space. Moreover, rather than employing episodic prototype construction typical of few-shot learning, the proposed system persistently updates centroid representations throughout supervised training and incorporates similarity-aware regularization directly into the loss function. This facilitates structural embedding reshaping specifically designed for visually overlapping illness categories, beyond traditional prototype-based learning methodologies. The input images are processed through HSV based green masking, morphological cleaning, extraction of leaf contours and resizing, and using a large amount of geometric and color-space augmentation to reduce the imbalance among the classes. DenseNet121 and EfficientNet-B0 are used to obtain feature representations and class-separated centroid of latent embedding's to form a Cross Disease Similarity Matrix, where similarity-aware optimization is possible during training. Grad-CAM on the target layers offers decipherable disease-specific activation signatures. The findings of the experiments show that classification accuracy at unseen samples is 99.77% with high resilience to visual confounding. The predictions, proximity of diseases that are similar and explainable features are provided by CDSAL, thereby facilitating reliable decision-making in agricultural diagnostics.
Why it matches plant phenotyping methodsトマト葉の病害状態を画像から推定する深層学習手法を提案し、前処理・特徴抽出・類似度学習・説明可能性を技術的中心として評価しているため。
abstractThe research proposes Cross Disease Similarity Awareness Learning (CDSAL), a robust multiclass tomato leaf disease detection framework based on high-quality and explainable deep learning.
Reproduction assets foundThe paper's plant-phenotyping inputs are two publicly available Kaggle image datasets explicitly named in the Data Availability statement: PlantVillage (emmarex/plantdisease) used as the main dataset and TomatoVillage (mamtag/tomato-village) used for ablation/field-condition experiments. No author analysis code, modelsDataset · publicThe datasets analyzed during the current study are available in the Kaggle repository. [https://www.kaggle.com/datasets/emmarex/plantdisease]Open asset ↗Kaggle · emmarex/plantdiseasehtml-lines:605-624Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Accurately estimating leaf area index (LAI) is vital for evaluating crop growth and predicting yields. Conventional approaches, however, often struggle due to the limited representativeness of available data and the complex structure of plant canopies, which reduce their reliability across diverse canopy architectures and observation conditions. To overcome these challenges, this work introduces an LAI retrieval framework that combines a three-dimensional radiative transfer model (3D RTM) with deep learning techniques. Representative 3D maize canopy scenarios were generated using the LESS model, producing synthetic LiDAR point clouds constrained by realistic structural parameters. A deep learning model based on PointNet++ was trained, and transfer learning (TL) was employed to facilitate knowledge transfer from simulated to actual measured data. The TL-enhanced model demonstrated significant improvement, with R2 rising from 0.537 to 0.842 and RMSE dropping from 0.541 to 0.288 m2·m−2. Moreover, retrieval performance was notably affected by scanning mode, angle, and stem diameter, achieving optimal results under TLS acquisition, moderate scanning angles, and intermediate stem widths. These findings suggest that integrating 3D RTM-generated synthetic point clouds with transfer learning is an effective strategy for enhancing the robustness and generalization of LiDAR-based LAI retrieval.
Why it matches plant phenotyping methodsLiDAR点群からトウモロコシのLAIを推定する手法を、3D放射伝達モデル、PointNet++、転移学習で開発・検証しており、植物形態形質の取得・推定が研究の中心です。
abstractthis work introduces an LAI retrieval framework that combines a three-dimensional radiative transfer model (3D RTM) with deep learning techniques.
Reproduction assets foundThe paper's field-measured LiDAR point cloud and LAI data (Yingke Oasis and Huazhaizi sites) come from a publicly accessible TPDC dataset with an explicit URL in the Data Availability Statement. No author analysis code, trained models, or synthetic dataset deposit is stated.Dataset · public2024WX06.
Data Availability Statement: The dataset used in this study was obtained from the National Tibetan
Plateau Data Center (TPDC, https://www.tpdc.ac.cn/ (accessed on 6 September 2025)), a publicly
accessible scientific data platform providing multi-source geoscientific datasets. The specific dataset
can be accessed via: https://www.tpdc.ac.cn/zh-hans/data/4d60d570-0aa9-417b-8a9d-c32b73b564
(accessed on 6 September 2025). The TPDC database integrates long-term observational and remote
sensing data with standardized quality control, ensuring the reliability and consistency of the datasets
for scientific research.
Acknowledgments: The authors would like to acknowledge the National TibetaOpen asset ↗4d60d570-0aa9-417b-8a9d-c32b73b564pdf-raw-page:19 lines:1-51Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Chili peppers (Capsicum annuum L.) are a strategic horticultural commodity in Indonesia, but their productivity is often hampered by pathogen infections that cause leaf diseases such as anthracnose, leaf spot, and yellow virus. Early detection by farmers is still dominated by subjective visual observation and prone to misdiagnosis due to the similarity of symptoms between diseases. Although Deep Learning technology through Convolutional Neural Networks (CNN) offers an automated solution, implementation in real-world conditions still faces significant challenges such as lighting variations, complex backgrounds, and limited local datasets. This often leads to a drastic decrease in model performance compared to testing in a controlled environment. To address these issues, this study proposes an optimization of the transfer learning strategy on the MobileNetV2 architecture by integrating progressive layer-wise fine-tuning and adaptive data augmentation techniques. The fine-tuning method is carried out gradually on the pre-trained model layers, while adaptive augmentation dynamically manipulates images based on environmental characteristics to improve model robustness. The results of this study, which include multi-class classification on cross-location image data, are projected to be able to boost the accuracy and generalization ability of the model in heterogeneous field conditions. Practically, this research provides a framework for a more precise and robust disease detection system to accelerate the implementation of precision agriculture in the future.
Why it matches plant phenotyping methods植物葉の病徴を画像から分類するCNN手法の改良が研究の中心であり、病害状態の表現型推定に該当する。転移学習のファインチューニングと適応的画像拡張、異なる圃場条件での頑健性評価を扱っている。
abstractThe results of this study, which include multi-class classification on cross-location image data, are projected to be able to boost the accuracy and generalization ability of the model in heterogeneous field conditions.
Reproduction assets foundThe paper states its chili leaf image dataset was supplemented with data from a supporting repository, cited as a public Mendeley Data deposit (reference [2]). This is a public plant-image dataset directly used for the paper's disease-classification phenotyping. No authors' analysis code or trained model checkpoint is,Dataset · public[2] F. Wajidi and N. Arifin, “Deteksi Penyakit Daun Cabai Menggunakan Kombinasi GLCM dan HSV dengan
Klasifikasi SVM,” vol. 11, no. 02, 2025. [Online]. Available:
https://data.mendeley.com/datasets/w9mr3vf56s/1Open asset ↗w9mr3vf56s/1pdf-page:9 lines:1-56Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Abstract Plant diseases are a serious danger to the world’s food security, because they lower agricultural output and increase economic losses. Due to subjectivity, fluctuating lighting, and environmental unpredictability, traditional visual examination techniques are frequently incorrect. The Excess Green (ExG) vegetation index and pseudo-thermal representations produced from RGB pictures are two synthetically developed complementary representations that are integrated with RGB imagery in this study’s lightweight multimodal deep learning system to address these issues. Histogram shifting and pseudo-infrared color mapping are used in a reproducible picture alteration pipeline to create the pseudo-thermal modality, which allows for extra visual signals without the need for specific thermal sensors. In order to classify plant diseases while preserving computational efficiency, the suggested framework uses MobileNetV3-Small backbones to extract modality-specific characteristics. This is followed by feature-level fusion. The publicly accessible Ginger Leaf Dataset, which includes RGB pictures of ginger leaves in four different conditions—Damage-Pest, Dehydrated, Healthy, and Leaf-blight—was used for the experiments. For training, validation, and testing, the dataset was split using a stratified 70:15:15 split. Python-based preprocessing procedures were used to create the extra modalities (ExG and pseudo-thermal representations) from the original RGB images. The experimental results show that the combination of the representations with RGB images can enhance the classification performance compared with the unimodal RGB-based models. Ablation experiments are also conducted to examine the contributions of different modalities to the overall categorization accuracy. The experimental results show that plant disease recognition can be improved with the help of efficient computing by combining lightweight convolutional neural networks with computationally generated visual representations.
Why it matches plant phenotyping methodsRGB画像からExG・疑似熱画像を生成し、植物葉の病害状態を分類するマルチモーダル手法が研究の中心であり、アブレーション評価も実施している。
titleHybrid deep learning-based multimodal framework for plant leaf disease classification using RGB, Excess Green (ExG), and pseudo-thermal representations with MobileNetV2
Reproduction assets foundThe paper's phenotyping experiments use the publicly available Ginger Leaf Dataset (RGB leaf images of four ginger leaf conditions), with a public GitHub repository and dataset website. The authors' derived ExG/pseudo-thermal representations and preprocessing scripts are only available upon request, so they do not yetDataset · publicor multispectral images
IEEE Geosci. Remote Sens. Lett. 2025
10.1109/LGRS.2025.XXXXXXX
Ulku, I., Tanriover, O. O. & Akagündüz, E. Cross-band correlation-aware interactive fusion for multispectral images. IEEE Geosci. Remote Sens. Lett.
10.1109/LGRS.2025.XXXXXXX
(2025).
10. Wong, J. Ginger Leaf Dataset. GitHub Repository (2023). https://github.com/wongjay1941/Ginger-Leaf-Dataset
11.
Bhakta I
A novel plant disease prediction model based on thermal images using modified deep convolutional neural network
Precis. Agric. 2023 24 23 39
10.1007/s11119-022-09927-x
Bhakta, I. et al. A novel plant disease prediction model based on thermal images using modified deep convolutional neural network. Precis.Open asset ↗https://github.com/wongjay1941/Ginger-Leaf-Datasetlines:580-681Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Three-dimensional (3D) procedural plant architecture models have emerged as an important tool for simulation-based studies of plant structure and function, extracting plant architectural parameters from field measurements, and for generating realistic plants in computer graphics. However, measuring the architectural parameters for these models at the field and population scales remains prohibitively labor-intensive. We present a novel algorithm that generates the 3D plant architecture from an image, to create a functional structural plant model from an image that reflects organ-level geometric and topological parameters, providing a more comprehensive representation of the plant’s architecture. Instead of using 3D sensors or processing multi-view images with computer vision to obtain the 3D structure of plants, we propose a method that generates token sequences containing a procedural definition of the plant architecture. This work uses only synthetic images for training and testing, where “exact” architectural parameters were known, which allowed for testing of the hypothesis that organ-level architectural parameters could be extracted from imagery data using a vision language model (VLM). A synthetic dataset of cowpea plant images was generated using the Helios 3D plant simulator, with the detailed plant architecture encoded in XML files. We developed a plant architecture tokenizer for the XML file defining plant architecture, converting it into a token sequence that a language model can predict. Then, a VLM was trained to predict plant architecture token sequences from images. Our results demonstrate that the model can predict plant architecture tokens with an F1 score of 0.73 in a teacher-forcing method. Evaluation of the model was performed through autoregressive generation, achieving a BLEU-4 score of 94.00% and a ROUGE-L score of 0.5182. Our model achieves lower MAPE than feature regression-based methods in estimating bulk plant-level traits that require understanding of the occluded 3D structure of the plant, such as leaf count and leaf area. We conclude that generating plant architecture and parameter extraction from synthetic imagery are feasible using a VLM approach, supporting future extension to real imagery.
Why it matches plant phenotyping methods画像から器官レベルの植物構造と形態形質を抽出するVLM手法の開発・評価が中心であり、植物フェノタイピング手法に該当する。
abstractWe present a novel algorithm that generates the 3D plant architecture from an image, to create a functional structural plant model from an image that reflects organ-level geometric and topological parameters
Reproduction assets foundThe paper's footnotes explicitly state that the authors' code is available on GitHub and the synthetic cowpea image/XML dataset is available on Hugging Face, both paper-specific and publicly actionable. The Helios URL is a generic third-party simulator library, not a paper-specific asset.Code · public1. ^ Code is available at: https://github.com/GEMINI-Breeding/Image2PlantArchitecture .Open asset ↗GEMINI-Breeding/Image2PlantArchitecturelines:600-676Dataset · public2. ^ Dataset is available at: https://huggingface.co/datasets/heesup/Cowpea-Architecture-XML .Open asset ↗heesup/Cowpea-Architecture-XMLlines:600-676Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Many countries greatly rely on agriculture as a means of livelihood and economic growth. Even the most industrialized countries need food, medicine, clothing, and shelter produced by crops. Rice is one of the most significant and widely grown crops worldwide. Nonetheless, the severely impacted crops in rice production are those of bacteria, fungi, and viruses, which decrease yield and quality. Manual disease detection is hectic, challenging, and, in most cases, inaccurate. Recent advances in deep learning and computer vision have demonstrated significant potential to improve the detection and classification of diseases. This study proposes a deep learning hybrid model for the automated detection and classification of rice leaf diseases. This method consists of five key stages: image preprocessing, segmentation, augmentation, multi-feature extraction via adaptive fusion, and classification. There are five rice leaf diseases to discuss and recognize: Blight, brown spot, sheath blight, tungro, and leaf blast. The first step is global contrast enhancement, which improves image quality. After that, the segmentation is performed using Otsu's Thresholding to extract the leaf area. Then, the modified VGG16 and modified ResNet50 networks are used in parallel to extract features using a transfer-learning approach. The adaptive fusion technique combines these features to obtain a dominant, proper feature representation. Lastly, the classification is done using an adaptive fusion score technique. Experimental results show excellent performance, with class-wise Precision in the range of 95.5-100%, class-wise recall in the range of 97.4-100%, and overall test accuracy of 98.5%.
Why it matches plant phenotyping methodsイネ葉の病害状態を画像から自動検出・分類する深層学習ワークフローが研究の中心であり、葉領域抽出、特徴抽出、分類性能まで評価しているため、植物病害フェノタイピング手法に該当する。
abstractThis study proposes a deep learning hybrid model for the automated detection and classification of rice leaf diseases.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe dataset employed in this study is accessible online at https://www.kaggle.com/datasets/rajeshbhattacharjee/rice-diseases-using-cnn-and-svm.Open asset ↗Kaggle · rajeshbhattacharjee/rice-diseases-using-cnn-and-svmhtml-lines:929-951Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Today, the intelligent automation of agriculture has received much attention from researchers. One of the important factors for the success of this automation is the timely diagnosis of plant disease and making a decision appropriate to the existing conditions of the plant. Since the progress of the disease is a determining factor in the type of treatment method, the diagnosis of the severity of the disease is of particular importance. However, accurate diagnosis of plant disease progression depends on various factors, including the availability of appropriate and well-annotated training datasets for designing an efficient diagnostic system. On the other hand, the similarity of the complications of different diseases has made this work challenging. In this study, two tomato diseases, namely Bacterial Spot and Mosaic Virus, are investigated using images collected from the PlantVillage, Taiwan tomato leaves, Field-PlantVillage, and Syn-PlantVillage datasets. The disease severity levels are divided into six stages for Bacterial Spot and four stages for Mosaic Virus, and a specifically designed deep convolutional neural network is proposed for severity classification. Experimental results demonstrate that the proposed method achieves high accuracy under challenging field conditions and outperforms several state-of-the-art methods.
Why it matches plant phenotyping methodsトマト葉の病徴・病害重症度を画像から段階分類するCNN、背景除去、病斑セグメンテーションを開発しており、植物状態の取得・推定手法が中心である。
titleA parallel convolutional neural network with background removal and lesion segmentation for field plant disease severity classification
Reproduction assets foundThe paper's own severity-annotated datasets are explicitly restricted (available only on request), so no public paper-specific data asset qualifies. The authors do provide an explicit public code availability link for their proposed BaSPaC model. The Mendeley and Drive links are pre-existing external datasets cited as,Code · publicCode availability
https://github.com/m-hasheminejad/BaSPaC.Open asset ↗m-hasheminejad/BaSPaChtml-lines:878-908Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Under small-sample conditions, hyperspectral leaf chlorophyll estimation is affected by high-dimensional collinearity, measurement noise, and cross-source acquisition discrepancies. Existing studies often treat training-distribution expansion and model-error complementarity separately. This study proposed a physically constrained composite spectral augmentation-weighted ensemble framework for reproducible small-sample chlorophyll estimation. Methods Using 1,113 valid spectrum-label pairs from the leaf subset of the GreenHySpectra dataset in the 400-1000 nm range, spectra and chlorophyll reference values were matched by sample identifiers and divided into training and validation sets. Low-magnitude Gaussian noise and smooth wavelength warping were applied only to the training set. XGBoost, partial least squares regression, and ridge regression were optimized with Optuna using a CMA-ES sampler, and ensemble weights were calibrated by Bayesian optimization. An independent external set of 90 tomato leaf samples was used to evaluate transferability. Results Composite augmentation improved model stability and reduced validation error relative to the non-augmented baseline. The weighted ensemble model achieved the best internal performance, with R² = 0.6392 and RMSE = 8.8883. On the external samples, the model achieved R² = 0.498 and RMSE = 9.801. Discussion The proposed workflow integrates physically plausible augmentation, heterogeneous learner complementarity, and independent external validation. The external results indicate partial cross-source transferability while highlighting distributional and measurement-chain discrepancies that still limit absolute generalization.
Why it matches plant phenotyping methods葉のクロロフィル量という植物形質をハイパースペクトルから推定する手法を開発し、外部データで転移性を検証しており、表現型取得・推定が研究の中心である。
titleHyperspectral estimation of leaf chlorophyll under small-sample conditions via spectral augmentation and weighted ensemble learning.
Reproduction assets foundThe paper's phenotyping analysis is built on the public GreenHySpectra hyperspectral dataset (leaf subset, 1,113 spectrum–chlorophyll pairs), which is a paper-specific, publicly available input with an authors' cited URL matching the allowed list. No author analysis code, trained models, or public deposit of the 90-solDataset · publicAvatarr05 ( 2023 ). GreenHySpectra/GreenHyperSpectra dataset (Hugging Face Datasets) [WWW document] . Available online at: https://huggingface.co/datasets/Avatarr05/GreenHySpectra (Accessed May 15, 2026).Open asset ↗Hugging Face Datasets · Avatarr05/GreenHySpectralines:749-785Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
The citrus diseases are affecting the fruit production worldwide thereby posing an economical burden. Major research is moving towards finding solutions using Artificial Intelligence (AI) and Image processing methods. Due to factors like illumination variations, leaf form, and disease symptoms, image data has intrinsic uncertainties that are typically difficult for traditional machine learning techniques to handle. In this paper, the interpretability of fuzzy logic is combined with the resilience of deep learning to propose a novel Fuzzy Convolutional Neural Network (Fuzzy-CNN) architecture for the automated diagnosis of citrus leaf diseases. The hybrid method uses a Convolutional Neural Network (CNN) to obtain complex features of citrus images, and a Fuzzy Inference System (FIS) to improve the classification results. The proposed approach encodes accurate data into fuzzy sets and applies linguistic concepts to determine the severity of a disease, which will contribute to the further development of the decision. In order to test and verify the proposed approach, several experiments were carried out, which proved that Fuzzy-CNN is more effective than regular CNN models with the approximate accuracy difference approximately 1.8, and especially in cases when the symptoms of disease are not clear. To strengthen experimental validation, the proposed method is evaluated on two independent datasets, including an external benchmark dataset, imbalance-aware evaluation metrics are employed to ensure robustness and generalizability. Experimental results demonstrate consistent and statistically significant improvements over existing neuro-fuzzy and machine learning approaches. This research contributes to early detection by collaborating the potential of fuzzy neural networks and offering a flexible solution for real-time disease detection in citrus crops.
Why it matches plant phenotyping methods柑橘葉画像から病害および重症度を推定するFuzzy-CNN手法を開発し、独立データセットとベンチマークで検証しており、植物フェノタイピング手法が中心である。
abstractpropose a novel Fuzzy Convolutional Neural Network (Fuzzy-CNN) architecture for the automated diagnosis of citrus leaf diseases.
Reproduction assets foundThe paper's Data Availability statement links two public image datasets used for the citrus disease phenotyping/classification experiments (a Mendeley citrus leaves dataset and a Kaggle orange fruit dataset), and a third public Kaggle citrus disease dataset is cited as the external benchmark dataset used for validationDataset · publicThe data used in the current study is publicly available from the following links. [https://data.mendeley.com/datasets/3f83gxmv57/2]Open asset ↗data.mendeley.com · 3f83gxmv57/2html-lines:525-539Dataset · publicThe data used in the current study is publicly available from the following links. [https://data.mendeley.com/datasets/3f83gxmv57/2] [https://www.kaggle.com/datasets/sgandhi2003/orange-fruit-dataset]Open asset ↗www.kaggle.com · sgandhi2003/orange-fruit-datasethtml-lines:525-539Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Understanding cellular growth dynamics in plants requires precise, long-term imaging of developing tissues. Cauline leaves are produced during the transition from vegetative to reproductive development and provide a useful system for studying how laminar organs diversify in form and function. While other laminar organs, such as rosette leaves and sepals, have been extensively studied, early cauline leaf development remains technically challenging to capture due to their concealed position, curved morphology, and the presence of dense trichomes. Here, we provide a complete pipeline for the dissection, confocal imaging, 2.5D segmentation, and image analysis of initiating cauline leaves in Arabidopsis thaliana . This method enables reproducible, high-resolution imaging of cauline leaves, supporting robust quantitative analysis of growth across developmental stages at cellular scale resolution. Key features • Fine dissection method for exposing initiating cauline leaves in Arabidopsis thaliana . • Long-term confocal live imaging of cauline leaf development at cellular resolution. • Optimized imaging parameters for high-fidelity 2.5D segmentation and growth analysis in MorphoGraphX.
Why it matches plant phenotyping methodsカウリン葉の成長を細胞レベルで定量化するための解剖、共焦点イメージング、2.5Dセグメンテーション、画像解析パイプラインが中心的に開発・提示されている。
abstractHere, we provide a complete pipeline for the dissection, confocal imaging, 2.5D segmentation, and image analysis of initiating cauline leaves in Arabidopsis thaliana .
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · public2. MorphoGraphX 2.0.1 ( https://morphographx.org/software/ ) (access date, 2026-02-26) [10–11]
3. All codes have been deposited to OSF: https://osf.io/uth78/ (access date, 2026-02-26)
Procedure
A. Plant growth
1. Sow the seeds in pots filled with moist, room-temperature soil. Add a layer of water to the bottom of the tray and cover with a lid to maintain high humidity.
Note: Space seeds sufficiently to avoid contact between the developing plants and to prevent leaf damage; typicallyOpen asset ↗OSFlines:109-143Code / dataset availability confirmedCrossref · checked 5 Sept 2026
Early disease diagnosis plays a key role in grape production for minimizing crop risk and maximizing yield. Downy Mildew, Powdery Mildew, and Bacterial Leaf Spot are some of the major diseases that threaten productivity and require timely and accurate diagnosis. This research introduces a new multi-model framework that integrates AI-based image segmentation triggered by Environmental Susceptibility Conditions to inform precision grape farming. The proposed method combines a soft-voting ensemble of the DeepLabV3+, U-Net, and FCN-8’s models for segmentation of diseased and healthy leaf areas with high accuracy, by understanding environment data to evaluate the risk of disease propagation. Major contributions of the study are the understanding of environmental conditions for context-aware disease propagation, an efficient ensemble segmentation method for accurate leaf disease segmentation and severity analysis, performed on a self-collected dataset from a grape farm in Nashik, Maharashtra, India. The system enables early warning and decision support mechanisms to promote sustainable disease management in grape cultivation, with potential implications for reducing unnecessary pesticide usage. Experimental results show the efficacy of the proposed method, with segmentation accuracy of 96.81% and precision of 99.09%, with a Dice score of 0.95 and a mean Intersection over Union (mIoU) of 0.91, demonstrating excellent robustness under noise conditions. Unlike existing studies either image or sensor-approaches, this work introduces the integration of image data and knowledge of environmental insights offers a scalable, reliable, and real-time disease monitoring solution aligned with the goals of smart and sustainable farming.
Why it matches plant phenotyping methodsブドウ葉の病斑領域を画像分割し、病害の重症度を推定する手法を開発・評価しており、植物の病害状態の取得が研究の中心です。
abstractThe proposed method combines a soft-voting ensemble of the DeepLabV3+, U-Net, and FCN-8’s models for segmentation of diseased and healthy leaf areas with high accuracy
Reproduction assets foundThe paper's grape leaf disease image dataset (NGLDD/NGLD) used for segmentation phenotyping is publicly deposited on Mendeley Data by the authors. No code or model checkpoints are reported as publicly available.Dataset · publicThe dataset used in this study is publicly available in the Mendeley Data repository as the Niphad Grape Leaf Disease
Dataset (NGLD) (DOI: https://doi.org/10.17632/8nnd2ypcv3.5).Open asset ↗Mendeley Data · 10.17632/8nnd2ypcv3.5pdf-page:25 lines:1-65Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
Northern Corn Leaf Blight (NCLB; also, Turcicum Leaf Blight, TLB), caused by Exserohilum turcicum (teleomorph: Setosphaeria turcica), is one of the most destructive foliar diseases of maize worldwide, often causing severe yield losses under favorable conditions. We developed a maize-specific, web-based Decision Support System (DSS) for real-time NCLB detection and management ( https://maize-nclb.streamlit.app/ ), integrating advanced deep-learning for automated diagnosis and fungicide advisory. Among thirteen Machine-learning and deep-learning models evaluated for classification, the Visual Geometry Group 16-layer convolutional neural network (VGG16) outperformed all others, achieving 94.0% accuracy, with balanced precision, recall, and F1-score of 0.94, and an AUC-ROC of 0.93. Confusion matrix analysis revealed minimal misclassification, with only 12 errors out of 357 samples, confirming the model's high reliability in distinguishing healthy and infected plants, while Grad-CAM visualizations consistently highlighted biologically meaningful lesion regions, supporting the model's interpretability and alignment with plant pathological symptoms. Field validation of DSS-guided fungicide recommendations (Azoxystrobin 18.2% + Difenoconazole 11.4% w/w SC) demonstrated significant benefits, reducing disease incidence to 6.8% compared with 67.4% in controls, achieving 90% disease reduction, and enhancing grain yield by 35.4% (8.55 t/ha), with a favorable cost-benefit ratio of 1:2.49. Seasonal disease progression analysis further confirmed DSS effectiveness, with cumulative disease burden reduced by approximately 85% compared with untreated control. These results highlight the potential of integrating deep-learning with field-validated management strategies into a practical DSS, demonstrating its potential for precision disease management in maize.
Why it matches plant phenotyping methods葉の病斑を画像から分類・可視化する深層学習法を開発し、野外で検証した研究であり、植物病害状態のフェノタイピング手法が中心です。
abstractintegrating advanced deep-learning for automated diagnosis and fungicide advisory
Reproduction assets foundThe paper explicitly states that the complete implementation (model training, preprocessing, evaluation, Grad-CAM visualization) and the final trained VGG16 model are publicly available on GitHub, and the deployed Streamlit DSS is publicly accessible. The Scribd link is a cited prior-work bulletin, not a paper-specificCode · publicthe complete implementation, including model training, preprocessing, evaluation, and Grad-CAM visualization, along with deployment instructions, is publicly available at: https://github.com/anuragd02/NCLB-VGG16-Detection.Open asset ↗anuragd02/NCLB-VGG16-Detectionhtml-lines:133-143Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
This paper explores the use of optimized convolutional neural networks (CNNs) to classify diseases affecting potato leaves using TensorFlow-2. The dataset, sourced from Kaggle's Plant Village repository, includes 152 images of healthy potato leaves and 1000 images each of early and late blight. The methodology covers data preparation, model architecture design, training, evaluation, and deployment. During data preparation, the data set was split into training sets (80%) and testing sets (20%), with images resized to 128x128 pixels. The Deep Learning (DL) models built using CNN with 4 different optimizers (ADAM, SGD, RMSPROP, and ADAMAX) and trained using a sparse categorical cross-entropy loss function, include multiple convolutional and pooling layers for feature extraction, and fully connected layers for classification. Early stopping was used to prevent overfitting. Model performance was assessed using accuracy, loss curves, confusion matrix, ROC curve, precision recall curve, classification report, and F1 score. In addition, we have used data augmentation to balance the dataset by increasing healthy potato leaves 6 times and the use of Ensemble Deep Learning (EDL). EDL10 which contains DL1 (CNN + ADAM), DL2 (CNN + SGD), DL3 (CNN + RMSPROP) and DL4 (CNN + ADAMX) performs best with a accuracy score of 97.0%. This highlights the importance of data balancing and the use of the ensemble classification approach for the detection of blight in Potato Leaves.
Why it matches plant phenotyping methodsジャガイモ葉の病害状態を画像から分類するCNN・アンサンブル手法の設計、評価、データ拡張が研究の中心であり、植物病害フェノタイピング手法に該当する。
abstractThis paper explores the use of optimized convolutional neural networks (CNNs) to classify diseases affecting potato leaves using TensorFlow-2.
Reproduction assets foundThe paper uses the public Kaggle PlantVillage potato leaf image dataset and archives its complete analysis source code on Zenodo with explicit availability statements and URLs.Code · publicThe complete source code is hosted in a DOI-minting repository and has been archived on Zenodo to ensure long-term accessibility and reproducibility. The code is released under an open-source license. The archived version corresponding to this publication is available at : https://doi.org/10.5281/zenodo.19624017Open asset ↗Zenodo · 10.5281/zenodo.19624017lines:252-314Dataset · publicThe datasets generated and/or analysed during the current study are available at : PlantVillage Dataset, accessed from https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset.Open asset ↗Kaggle · plantvillage-datasetlines:252-314Code / dataset availability confirmedEurope PMC · bioRxiv · Crossref · checked 15 Sept 2026
Plant disease detection using deep learning is essential for precision agriculture, enabling early and automated crop health monitoring. This study proposes an end-to-end transfer learning pipeline, LeafyVGG-16, for multi-class classification of plant diseases and nutrient deficiencies using a tomato leaf dataset. The framework integrates data preprocessing, augmentation, and a VGG-16 backbone with a two-stage fine-tuning strategy. The proposed model is evaluated against CNN, DenseNet-121, Inception-V3, EfficientNetB0, and ResNet-50, achieving an accuracy of 0.93 with precision, recall, and F1-scores of 0.93, 0.90, and 0.92, respectively. These results demonstrate the effectiveness of transfer learning for fine-grained plant disease recognition. We further evaluate model robustness under adversarial cyber attacks to assess deployment reliability in agricultural systems. Under Fast Gradient Sign Method (FGSM) attacks ( ϵ = 0.01– 0.05), the model shows an accuracy drop of 1%–7.5%, while Projected Gradient Descent (PGD) attacks ( ϵ = 0.05, step size = 0.005, 10 iterations) produce similar degradation, highlighting the model’s vulnerability to adversarial perturbations. These findings highlight potential security and reliability risks in AI-based agricultural decision-making systems. Future work will focus on improving robustness and cyber-resilience and extending this framework to other crops for secure and context-aware deployment in resource-constrained environments.
Why it matches plant phenotyping methods植物葉の画像から病害・栄養欠乏状態を分類する深層学習パイプラインが研究の中心であり、複数モデルとの比較評価と敵対的攻撃下での頑健性検証も実施しているため。
abstractThis study proposes an end-to-end transfer learning pipeline, LeafyVGG-16, for multi-class classification of plant diseases and nutrient deficiencies using a tomato leaf dataset.
Reproduction assets foundThe paper's plant-phenotyping input is the publicly available Tomato-Village Variant-a dataset (4,525 tomato leaf images across 8 disease/deficiency classes), which the authors explicitly cite with a public Kaggle URL. No author analysis code, trained model checkpoints, or supplementary data deposits are mentioned. TheDataset · publicThis study uses the publicly available Tomato-Village
dataset [18], which is designed for real-world tomato disease
detection in agricultural environments.Open asset ↗pdf-raw-page:3 lines:1-59Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Crop diseases pose a significant threat to agricultural productivity and global food security. Timely and accurate detection of such diseases is crucial for improving both crop yield and quality. While numerous deep learning approaches rely solely on image data for disease identification, they often overlook the complementary value of textual information in enhancing visual analysis. To address this limitation and effectively fuse features from different modalities, we propose a Cross-Model fusion framework based on a vision-language model that integrates cross-attention and gated fusion mechanisms for crop disease recognition. Our approach utilizes the Zhipu.ai multi-modal model to generate comprehensive textual descriptions of diseased crop leaves, including global description, local lesion description, and color-texture description. These textual descriptions are then encoded into feature embeddings, while visual features are extracted using the ShuffleNet-v2 model as the image encoder. Subsequently, a cross-attention module aligns and fuses the two modalities, and a gated fusion module enables dynamic feature selection during the fusion process. Extensive evaluations on the Soybean Disease and PlantVillage datasets demonstrate that our method outperforms existing image-based models in terms of accuracy. Specifically, our model achieves recognition accuracies of 99.04% and 99.12% on the respective datasets, surpassing the ShuffleNet-V2 model by 1.09% and 2.53%, respectively. These results highlight the effectiveness of Cross-Model learning in integrating visual and textual cues for accurate and efficient disease recognition, offering a scalable solution for crop disease diagnosis.
Why it matches plant phenotyping methods植物葉の病徴を画像と言語情報から認識する融合フレームワークを開発し、複数データセットで既存手法と比較評価しているため、植物フェノタイピング手法が中心である。
abstractwe propose a Cross-Model fusion framework based on a vision-language model that integrates cross-attention and gated fusion mechanisms for crop disease recognition.
Reproduction assets foundThe paper's crop disease recognition experiments use two openly available image datasets, both with explicit public availability statements in the Data Availability section: the Soybean Disease dataset (Dryad DOI) and the PlantVillage dataset (Kaggle). No author analysis code, trained models, or generated text-annotaitDataset · publicThe datasets utilized in this study are openly accessible. The soybean dataset is available at https://doi.org/10.5061/dryad.41ns1rnj3.Open asset ↗Dryad · 10.5061/dryad.41ns1rnj3html-lines:403-424Dataset · publicThe plantvillage dataset is available at https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset.Open asset ↗Kaggle · plantvillage-datasethtml-lines:403-424Code / dataset availability confirmedEurope PMC · bioRxiv · checked 15 Sept 2026
Accurate and reproducible assessment of foliar disease severity is essential for evaluating the performance of heterogeneous plant communities and understanding host-pathogen interactions. However, traditional visual scoring methods remain subjective, with limited precision, and difficult to scale in large phenotyping experiments. Here, we present a semi-automated image analysis workflow designed to quantify multiple foliar disease symptoms simultaneously on wheat flag leaves sampled from varietal mixtures. The workflow combines three methodological components: (i) a standardized protocol for leaf sampling and imaging, (ii) supervised machine learning segmentation using Random Forest implemented in Ilastik to classify multiple symptoms (powdery mildew and yellow rust), and (iii) a graphical user interface facilitating pipeline deployment by non-specialist operators. To evaluate the influence of image representation on classification performance, four color spaces (RGB, HSV, HLS, LAB) were systematically compared. The approach was validated using images of durum wheat flag leaves collected from a field experiment assessing eight-way varietal mixtures under natural fungal pressure. Cross-validation against manually annotated images demonstrated high segmentation accuracy across all symptom. Comparison among color spaces revealed only minor differences in performance. Overall, this workflow offers a cost-effective, annotation-efficient and reproducible alternative to deep learning approaches, leveraging open-source and actively maintained tools while requiring limited training data and enabling objective, reproducible and scalable disease phenotyping.
Why it matches plant phenotyping methods葉の病害症状を画像解析で定量化するワークフローを開発し、色空間比較と手動アノテーションによる検証を行っており、植物表現型取得法が中心である。
abstractwe present a semi-automated image analysis workflow designed to quantify multiple foliar disease symptoms simultaneously
Reproduction assets foundThe paper's authors explicitly state that all code implementing the leaf disease quantification workflow (SegLeaf, including the graphical interface and documentation) is hosted in a public GitHub repository. No separate public phenotype dataset or trained model checkpoint is described in the supplied blocks.Code · publicted by the Agence Nationale de
la Recherche (ANR) (project SCOOP, grant no. ANR-19-CE32-0011; and project MOBIDIV,
grant no. ANR-20-PCPA-0006).
Code and Data Availability
The method and associated scripts developed in this work are freely available to the re-
search community. All code is hosted in a public GitHub repository at https://github.com/titouanlegourrierec/SegLeaf, which includes the full implementation of the method includ-
ing the graphical interface and documentation to guide users through the analysis pipeline.
15
.
CC-BY 4.0 International license
made available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display tOpen asset ↗titouanlegourrierec/SegLeafpdf-raw-page:15 lines:1-39Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Tea (Camellia sinensis) is the world's second most consumed beverage, enjoyed daily by more than two billion people. In Bangladesh, it serves as a cornerstone agricultural export and a major sector of the domestic economy. However, commercial tea cultivation remains highly vulnerable to fungal and pest-related diseases such as Blight, Red Rust, and Helopeltis which severely reduce crop yield and compromise leaf quality. While early detection is critical to preventing widespread outbreaks, traditional manual inspection is slow, subjective, and highly error-prone. Deep learning provides a scalable alternative, yet single-branch networks often struggle to capture both minute disease lesions and broader structural degradation simultaneously. To address this, we propose a Hybrid Feature Fusion architecture that runs two highly efficient feature extractors in parallel: EfficientNetV2-Small to isolate fine-grained local textures, and MobileNetV3-Small to capture the global structural context of the leaf. The models were trained and evaluated on a real-world dataset of 2,000 annotated images, evenly distributed across the four target classes (Blight, Red Rust, Helopeltis, and Healthy). Before training, the images underwent a standardized preprocessing pipeline including resizing to 224 × 224 pixels and normalization, supplemented by a dynamic augmentation strategy featuring random rotations, horizontal flips, and brightness adjustments to improve model robustness. The proposed hybrid framework achieved an outstanding peak classification accuracy of 96.80% alongside a macro Area Under the Curve (AUC) of 0.9980. To rigorously validate its performance, the hybrid model was benchmarked against six diverse architectures: a Vision Transformer (ViT-B16 at 76.40%), a Custom CNN (89.60%), MobileNetV3 (94.40%), ResNet50 (95.60%), DenseNet121 (96.40%), and EfficientNetV2-B3 (97.60%). Although EfficientNetV2-B3 achieved a marginally higher raw accuracy, the proposed dual-branch framework delivered a superior precision-recall balance and faster convergence stability. These findings demonstrate that the proposed hybrid methodology is highly reliable and computationally balanced, making it an ideal candidate for integration into Internet of Things (IoT) edge devices for real-time disease monitoring in precision agriculture.
Why it matches plant phenotyping methods茶葉の病徴を画像から分類する深層学習手法の開発と、注釈付きデータセットおよび複数モデルとのベンチマーク検証が中心であり、植物病害状態の表現型推定に該当する。
abstractwe propose a Hybrid Feature Fusion architecture
Reproduction assets foundThe paper's Data Availability statement explicitly deposits the curated 2000-image tea leaf dataset on Mendeley Data and the analysis code on GitHub, both with public URLs matching allowed_urls.Dataset · publicThe dataset comprising 2000 annotated tea leaf images was curated under real-world field conditions. It has been made available at https://data.mendeley.com/datasets/3x42rbj8yv/1.Open asset ↗3x42rbj8yv/1html-lines:465-480Code · publicThe computational code supporting the findings of this study is publicly accessible on GitHub: https://github.com/rayhankhan2192/Tea_Leaf_Disease_Model.Open asset ↗GitHub · rayhankhan2192/Tea_Leaf_Disease_Modelhtml-lines:465-480Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
Quantifying the kinetics of net CO2 assimilation (A) and stomatal conductance (gs) under fluctuating light typically relies on gas exchange measurements, which are slow and thus unsuited for high-throughput phenotyping. As a result, faster, non-invasive phenotyping methods are needed to further evaluate these traits at a larger scale. However, first the relationship between non-steady-state parameters must be examined in greater detail. In this study, we aimed to determine whether variations in non-steady-state values of chlorophyll fluorescence and leaf temperature reflect differences in key gas exchange traits under fluctuating light conditions. Here, the correlations between the times required for a change in non-steady-state A, gs, operating efficiency of PSII (ΦPSII), and leaf temperature (Tleaf) during stepwise changes in light intensity were evaluated across nine plant species. Both steady-state and non-steady-state photosynthetic traits varied significantly among species. Overall, we found significant positive correlations between non-steady-state A and ΦPSII for time to 50% and 90% of final steady-state values (t50; r2 = 0.70) and (t90; r2 = 0.33). The t90 of gs and that of Tleaf were also significantly correlated after both increases (r2 = 0.45) and decreases (r2 = 0.61) in light intensity. Our findings suggest that the times required for a change in ΦPSII (particularly t50) and Tleaf (particularly t90) can be used as indicators of dynamic A and gs, respectively, facilitating faster phenotyping of the complex processes of photosynthesis and stomatal conductance kinetics in the future.
Why it matches plant phenotyping methods非定常クロロフィル蛍光と葉温を用いて光合成・気孔コンダクタンス動態を推定する高速フェノタイピング手法を評価しており、相関検証が研究の中心である。
abstractfaster, non-invasive phenotyping methods are needed to further evaluate these traits at a larger scale.
Reproduction assets foundThe paper's primary gas exchange, chlorophyll fluorescence, and leaf temperature phenotyping data are explicitly deposited in the WUR data repository (DOI 10.17887/WUR01-TMWYJN), stated in the Data availability section. No author analysis code repository is stated; the agricolae R package is a generic library, not a论文-Dataset · publicThe primary data and associated metadata are publicly available through the WUR data repository at https://doi.org/10.17887/WUR01-TMWYJN .Open asset ↗WUR data repository · 10.17887/WUR01-TMWYJNlines:406-446Code / dataset availability confirmedCrossref · Europe PMC · checked 14 Sept 2026
LiDAR / point cloudFlowerLeafStem / branchSegmentationGrowth / development / phenology
The segmentation of 3D point clouds of plant organs, such as leaves and stems, helps to monitor plant growth and is a key step in plant growth phenotype analysis. Compared to point cloud segmentation tasks in other fields, plant point cloud segmentation is more challenging due to the interwoven distribution of various parts such as stems, leaves, and flowers. In this paper, we propose a universal point cloud segmentation network PlantEFRSegnet that can be used for multi-species of plants. The proposed PlantEFRSegnet utilizes a newly designed edge point preservation downsampling module to identify and preserve the points at the edges of plant organs during the downsampling process, in order to assist the segmentation network in learning the contours of various plant organs. PlantEFRSegnet performs supervised feature repair on the point cloud features obtained through downsampling to mitigate the impact of feature loss on segmentation performance during feature embedding. The encoder of the segmentation network is composed of four local feature extraction modules. These four modules can not only extract features but also enhance the features corresponding to points with high contributions in local regions based on point attention mechanism. We evaluated the proposed PlantEFRSegnet on a laser-scanned plant point cloud dataset. Compared with the state-of-the-art approaches, the proposed PlantEFRSegnet achieved better segmentation results.
Why it matches plant phenotyping methods植物器官の3D点群を対象に、器官分割と植物成長フェノタイプ解析を行う新規ネットワークを開発・評価しており、フェノタイプ取得の計算手法が中心である。
abstractThe segmentation of 3D point clouds of plant organs, such as leaves and stems, helps to monitor plant growth and is a key step in plant growth phenotype analysis.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe experimental dataset used in this paper can be obtained through the following link: https://github.com/dllab23/PlantPointCloud (accessed on 11 May 2026).Open asset ↗dllab23/PlantPointCloudhtml-lines:785-806Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Background High-throughput automated image analysis holds great promise for plant breeding by enabling faster, more accurate assessment of traits relevant to crop improvement. Imaging-based systems, such as the CropReporter, allow automated quantification of photosynthetic parameters like PSII efficiency under ambient light from a top-down 2D perspective. However, standard analysis tools average values across the 2D top view, overrepresenting upper leaves and underrepresenting those in the lower canopy. Upper leaves may occlude lower ones, and due to the pinhole projection of the camera, lower leaves of the same size appear smaller in the image. Consequently, vertical heterogeneity in PSII efficiency within the canopy cannot be resolved using a single 2D image. Results To address these issues, we integrated top-view PSII efficiency data (by CropReporter) with 3D structural data from RGB point clouds (by MaxiMarvin). Alignment accuracy between MaxiMarvin and CropReporter was high, with R² ≥ 0.98 for the x-axis and R² ≥ 0.99 for the y-axis. The method was tested using Chenopodium quinoa, Glycine max, and Solanum tuberosum, exposed to salinity, waterlogging and drought stress respectively. In Chenopodium quinoa, it allowed precise determination of when senescence began in the lower leaves. In Solanum tuberosum, the reduction in PSII efficiency by drought was the same for all leaf layers, while in Glycine max, waterlogging stress most strongly affected the middle layer of the canopy. Conclusions This framework enables the 3D mapping of PSII efficiency across the vertical plant profile by combining top-view chlorophyll fluorescence imaging (CropReporter) with 3D structural data (MaxiMarvin). It reveals vertical variation in photosynthetic activity across canopy layers. With standard 2D chlorophyll fluorescence imaging it is difficult to distinguish between non-photosynthetic tissues like flower heads and lower layers of leaves, that might have the same PSII values. Using height-based filtering, taking data from the 3D mapping, such distinction can be made with the method presented in this paper. This allows estimating the PSII efficiencies of leaves only. By capturing layer-specific responses to abiotic stress and developmental changes, the method provides physiologically relevant input for crop growth modelling and highlights the importance of accounting for canopy structure in photosynthetic analyses.
Why it matches plant phenotyping methods2Dクロロフィル蛍光によるPSII効率を3D植物構造へ投影し、群落層別の葉の生理形質を推定する手法の開発・検証が中心である。
abstractTo address these issues, we integrated top-view PSII efficiency data (by CropReporter) with 3D structural data from RGB point clouds (by MaxiMarvin).
Reproduction assets foundThe authors state that the analysis scripts (2D–3D alignment pipeline) and the phenotyping data used in the study are included with the publication as supplementary material, accessible via the article DOI. This is a paper-specific, publicly available asset containing the authors' analysis code and data.Dataset · publicThe scripts and the data that were used in the current study are available and added to this publication.Open asset ↗lines:143-180Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Given the increasing frequency, severity, and socioecological impacts of wildfires, there is an urgent need for robust frameworks to better characterize fire behavior and flammability patterns across ecosystems to support early warning, mitigation, and management strategies. However, flammability remains difficult to quantify and scale, as it involves multiple interacting components that are typically measured at the bench scale. This study aimed to establish empirical links between spectral information, plant traits, and flammability metrics, and to scale these relationships to satellite imagery to translate these metrics into a spatial context. We combined laboratory spectroscopy, plant trait measurements including leaf mass per area, carbon, and cellulose, and combustion experiments using a simple and reproducible burning device. In total, 84 samples were collected and analysed, allowing us to characterise how spectral signatures relate to vegetation traits and fire behaviour. Spectral indices were developed to estimate plant traits, which were subsequently used as predictors in flammability models. These models were then transferred to Environmental Mapping and Analysis Program (EnMAP) hyperspectral imagery to derive spatial estimates across eucalypt forests and grasslands of the Australian Capital Territory (ACT). Spectral information distinguished fuel types and captured variability of the plant traits, while these traits showed associations with combustion behaviour. Based on these links, the best-performing model predicted the rate of temperature increase, a combustibility metric, in eucalypt forests (R2 = 0.70; Root Mean Square Error = 32.48 °C/s). In contrast, grassland models showed limited predictive performance, likely due to weaker relationships between plant traits and flammability metrics. Overall, this study demonstrates a practical and scalable approach for deriving flammability maps from hyperspectral and in situ data, highlighting the potential of plant-trait-based remote sensing. The resulting maps should not be interpreted as standalone fire risk products, but rather as a characterization of the structural and biochemical drivers of flammability. The main constraint of this work is the limited sample size. Future research should expand spatial and temporal coverage to better capture vegetation variability and enable the inclusion of independent validation datasets. Exploring alternative combustion protocols and testing more advanced spectral modelling approaches for trait estimation would provide additional insights.
Why it matches plant phenotyping methods植物形質を分光情報から推定し、ハイパースペクトル画像へ展開して可燃性関連の植物状態を評価する手法が研究の中心であり、モデル性能も検証しているため。
abstractSpectral indices were developed to estimate plant traits, which were subsequently used as predictors in flammability models.
Reproduction assets foundThe paper's supplementary materials (hosted publicly by MDPI) contain the paper-specific plant phenotype measurements: sampled species lists, fractional cover, and measured vegetation traits across dates and plots, plus combustion replicate variability and trait–flammability relationship data. The raw underlying data,谱Supplement · publicbroader environmental coverage, improved plant trait retrieval meth-
ods, and independent validation. Future work should also explore non-linear modelling
frameworks to better capture the complexity of vegetation flammability across ecosystems.
Supplementary Materials: The following supporting information can be downloaded at:
https://www.mdpi.com/article/10.3390/rs18101546/s1, Supplementary Table S1 provides the list of
sampled plant species and their percentage cover across sites, paddocks, plots, and fuel types; Table
S2 presents the fractional cover of each species and litter component; Figure S1 shows the study-site
vegetation map; Figures S2–S6 show the measured vegetation traits acrosOpen asset ↗pdf-raw-page:22 lines:1-49Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published12 May 2026Journal of Advanced College of Engineering and ManagementCited by 0 · OpenAlex ↗
Apple cultivation is a crucial agricultural activity in various mountainous regions, playing a vital role in supporting the local economy and sustaining the livelihoods of farmers. Several prominent mountain districts are known for leading apple production. However, apple orchards in these areas are often threatened by numerous diseases that reduce fruit yield and quality. In this research, we suggest a machine learning-based technique to automate the detection and classification of common apple diseases based on images of apple leaves collected from various regions. Through the use of Convolutional Neural Networks (CNN), the system can classify diseases with 97.36% precision. For post hoc explainability, Grad-CAM is used, which highlights the important regions that influenced CNN’s decision. The automated disease detection tool provides farmers in Nepal’s rural mountain areas with an affordable real time solution to monitor orchard health, minimize crop loss, and improve apple production. The dataset used in this study is originally derived from the United States based PlantVillage dataset, which is widely used for apple leaf disease classification research. Although the dataset is not collected from Nepal, the visual characteristics of apple leaf diseases remain largely consistent across regions due to similar biological infection patterns. Therefore, the model trained on this dataset is applicable to Nepali apple cultivation environments as well. At present, a publicly available or annotated Nepali specific apple leaf disease dataset is not available, which limits region-specific training and evaluation.
Why it matches plant phenotyping methodsリンゴ葉画像から病害状態を分類するCNNベースの手法とGrad-CAMによる解釈を中心に扱うため、植物病害フェノタイピング手法として該当する。
abstractwe suggest a machine learning-based technique to automate the detection and classification of common apple diseases based on images of apple leaves collected from various regions.
Reproduction assets foundThe paper's apple leaf disease image dataset (9,696 images, four classes) is publicly available on Kaggle and explicitly cited by the authors as the dataset used for training and evaluation. No author code, trained model, or other paper-specific assets are reported.Dataset · publicIn this study, the dataset used for apple leaf disease classification was obtained from
Kaggle [20]. The dataset contains a total of 9,696 images of apple leaves, which include
both diseased and healthy samples.Open asset ↗Kagglepdf-raw-page:4 lines:1-39Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Common beanRiceWheatLeafClassificationDisease symptoms / severity
In Bangladesh, crop leaf diseases create a serious risk to food security and production from agriculture. Timely identification of leaf diseases in rice, wheat, and bean crops is considered crucial for the implementation of effective disease detection and classification strategies. To address this challenge, a MobilenetV2-based disease identification and classification system is proposed in this research. Previous studies focus on classifying diseases of a single species, leaving the need to train models separately for each species. This research focuses on forming a single standard model to perform leaf disease classification for multiple crop species including rice, wheat, and beans. The approach makes use of transfer learning with the MobilenetV2 model, which is fine-tuned using a dataset of annotated crop leaf images specific to Bangladesh. Following a comprehensive evaluation, an overall accuracy of 97.87% was achieved in the classification of crop leaf diseases, which surpasses the accuracy of a number of previous studies focusing on leaf disease detection of a single crop. The system demonstrates the capability to rapidly diagnose diseases in real time by enabling the users to prompt intervention to mitigate potential crop losses, ultimately leading to amplified crop yield and food security. Overall, the research highlights the promise of AI-powered solutions in tackling crop leaf disease detection, which in turn encourages greater research and technology adoption to support sustainable farming methods especially in the crop disease classification domain in Bangladesh and throughout the world. Received: 24 May 2025 | Revised: 9 March 2026 | Accepted: 14 April 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The data that support the findings of this study are openly available in the Bangladeshi Crops Disease Dataset at https://www.kaggle.com/datasets/nafishamoin/bangladeshi-crops-disease-dataset and the Bean Disease Dataset at https://www.kaggle.com/datasets/therealoise/bean-disease-dataset. Author Contribution Statement Md. Mahmudul Hasan: Conceptualization, Methodology, Visualization, Supervision. Md. Omar Faruq: Software, Validation, Writing – original draft. Mahadi Hasan Musa: Formal analysis, Investigation. Mohammad Mamunur Rashid: Resources, Data curation, Writing – review & editing. Khandaker Mohammad Mohi Uddin: Writing – review & editing, Project administration, Supervision.
Why it matches plant phenotyping methods葉画像から作物の病害状態を推定する深層学習手法を開発・評価しており、植物病害フェノタイピングが中心的な技術貢献である。
abstracta MobilenetV2-based disease identification and classification system is proposed in this research.
Reproduction assets foundThe paper's Data Availability Statement openly provides the Bean Disease Dataset on Kaggle, which is one of the two public image datasets used to train the multi-crop leaf disease classification model. The Bangladeshi Crops Disease Dataset URL is not among the allowed URLs, so only the bean dataset is reported. No codeDataset · publict
The authors declare that they have no conflicts of interest to
this work.
Data Availability Statement
The data that support the findings of this study are openly
available in the Bangladeshi Crops Disease Dataset at https://
www.kaggle.com/datasets/nafishamoin/bangladeshi-crops-disease-
dataset and the Bean Disease Dataset at https://www.kaggle.com/datasets/therealoise/bean-disease-dataset.Author Contribution Statement
Md. Mahmudul Hasan: Conceptualization, Methodology,
Visualization, Supervision. Md. Omar Faruq: Software, Valida-
tion, Writing – original draft. Mahadi Hasan Musa: Formal
analysis, Investigation. Mohammad Mamunur Rashid: Resources,
Data curation, Writing – review & editing.Open asset ↗Kaggle · therealoise/bean-disease-datasetpdf-raw-page:11 lines:1-83Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Tomato leaf diseases pose significant threats to crop yield and food security. However, in real-world cultivation environments, factors such as fluctuating illumination, varying leaf occlusion, and ambiguous lesion morphology often compromise detection accuracy. This paper presents the Gradient-aware Bidirectional Retentive Detection Transformer (GBR-DETR), a model designed for high-precision, real-time disease detection. This model is composed of two network structures and a retentive feature aggregation module: (1) a Multi-scale Gradient-Aware Transfer Network (MGAT-Net) is designed to encode gradient information through the Sobel operator, thereby enhancing the localization stability for small and blurry lesions; (2) a Bidirectional Context Pyramid Network (BCPN) is proposed to enable bidirectional interactions among multi-level features through a top-down and a bottom-up pathway, thereby generating multi-scale lesion features and bridging cross-scale semantic gaps; and (3) a Retentive Feature Aggregation Module (RFAM) is used to suppress background noise and establish global feature correlations, thereby enhancing the overall representation capability for lesion recognition. Experiments on the Multi-scenario Tomato Leaf Disease (M-TLD) dataset show that GBR-DETR yields gains of 3.12, 4.88, and 3.41 percentage points in mAP 50-95 , mAP 50 , and mAP 75 , respectively, over the baseline RT-DETR, while also outperforming representative DETR-based and CNN-based detectors. The model demonstrates robust generalization on the PlantDoc cross-domain benchmark, achieving a 2.11% improvement in mAP 50 over the baseline. Deployed on the NVIDIA Jetson Orin Nano with TensorRT FP16, it achieves 54 ms latency, enabling real-time disease monitoring on edge devices. This solution provides effective technical support for real-time disease monitoring in smart agriculture.
Why it matches plant phenotyping methodsトマト葉の病斑・病害状態を画像から検出するモデルを開発し、複数データセットで比較検証、エッジデバイス実装まで評価しており、植物病害表現型の取得手法が中心である。
abstractThis paper presents the Gradient-aware Bidirectional Retentive Detection Transformer (GBR-DETR), a model designed for high-precision, real-time disease detection.
Reproduction assets foundThe paper's M-TLD tomato leaf disease dataset (2212 images, 6581 annotations) and the GBR-DETR implementation/training code are explicitly stated to be publicly available at the authors' GitHub repository.Dataset · publicThe M-TLD dataset and all annotation files are publicly available at https://github.com/zhuojiaxiong6/DETR (accessed on 29 April 2026) to facilitate reproducibility and future research.Open asset ↗zhuojiaxiong6/DETRlines:38-108Code · publicThe code and dataset used in this study are publicly available at the following GitHub repository: https://github.com/zhuojiaxiong6/DETR (accessed on 29 April 2026). This repository contains the implementation of GBR-DETR, a Detection Transformer variant developed for detecting tomato leaf diseases and pests. All relevant training scripts, configuration files, and instructions for dataset usage are provided in the repository.Open asset ↗zhuojiaxiong6/DETRlines:673-675Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Accurate plant disease segmentation is often constrained by the availability of large, finely annotated datasets, particularly for rare diseases. This work presents a synthetic data generation pipeline that combines 3D leaf modelling with diffusion-based disease synthesis to address this limitation. Procedurally-generated leaf geometries are built in the 3D modelling package Blender to provide exact ground-truth masks, after which style-transfer is applied using Stable Diffusion, fine-tuned with Low-Rank Adaptation (LoRA) and guided by ControlNet conditioning to both preserve leaf structure and enforce correct lesion placement. The approach is evaluated on apple leaf diseases using a deliberately restricted subset of the PlantVillage dataset, simulating a controlled low-data-resource environment. Downstream task effectiveness is measured through leaf disease segmentation. The results show that combining data from the pipeline with limited real data leads to consistent improvements in segmentation performance.
Why it matches plant phenotyping methods植物病斑の画像セグメンテーション性能向上を目的に、3D葉モデルと拡散モデルによる合成データ生成パイプラインを開発・評価しており、植物病害状態の画像ベース推定が中心である。
abstractThis work presents a synthetic data generation pipeline that combines 3D leaf modelling with diffusion-based disease synthesis to address this limitation.
Reproduction assets foundThe authors publicly deposited the paper's annotated PlantVillage subset (75 images with segmentation masks) plus 300 synthetic images with ground-truth masks on Zenodo, directly reproducing this paper's phenotyping/segmentation data.Dataset · publicThis annotated subset of PlantVillage is available at https://doi.org/10.5281/zenodo.18659728 . The repository contains the 75 images from the restricted dataset with the corresponding segmentation masks along with 100 synthetic images per disease generated using Blender and Stable Diffusion, each with corresponding ground truth masks.Open asset ↗zenodo · 10.5281/zenodo.18659728lines:314-325Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Accurate and real-time detection of maize foliar diseases is important for field disease monitoring and yield protection. However, in complex natural field environments, different diseases often exhibit high visual similarity, and early weak lesions are easily confused with background elements such as dry leaves, soil, and shadows, leading to false positives and missed detections in existing models. To address these challenges, this study proposes an improved lightweight maize foliar disease detection model based on YOLO11, termed CKM-YOLO11. First, a mixed local channel attention mechanism is introduced and adapted to the task in the backbone to construct the C3k2-MLCA module, thereby enhancing joint modeling of local lesion textures, edge details, and global contextual information. Second, a lightweight residual attention module, named MLCA-HeadLite, is designed at the P5 layer of the neck/head to alleviate the suppression of weak lesion responses during deep feature fusion. Experimental results demonstrate that the proposed model achieves an mAP@50 of 81.5% on a self-constructed maize disease dataset with complex field backgrounds, improving mAP@50 and mAP@50-95 by 3.2 and 3.4 percentage points, respectively, compared with the baseline YOLO11, while maintaining a low parameter count and computational cost. Further analyses based on the confusion matrix, comparisons of detection results, and Grad-CAM visualizations indicate that the proposed model performs better in background suppression, retention of weak lesion responses, and robustness in complex scenes. This study provides a reference for the lightweight design of maize foliar disease detection models in complex field environments and their deployment on agricultural edge devices.
Why it matches plant phenotyping methodsトウモロコシ葉の病斑・病害状態を画像から検出する軽量モデルを開発し、データセット上で性能評価しているため、植物病害表現型の取得手法が中心である。
abstractthis study proposes an improved lightweight maize foliar disease detection model based on YOLO11, termed CKM-YOLO11.
Reproduction assets foundThe paper's maize foliar disease detection dataset is built from public image sources (CD&S Dataset from OpenDataLab and PlantDoc-Dataset corn rust leaf folders) that are explicitly cited with public URLs, qualifying as paper-specific public phenotype image inputs. The self-collected images and the authors' code/tranedDataset · publict was constructed using three public-data components together with a small number of self-collected maize leaf images. First, field-acquired maize disease images were obtained from the Corn Disease and Severity (CD&S) Dataset downloaded from OpenDataLab, and only the Dataset_Original folder in the raw dataset package was used ( https://opendatalab.com/OpenDataLab/CD_and_S/tree/main , accessed on 3 May 2026). Second, to supplement the leaf rust category, additional images were collected from the train/Corn rust leaf folder of the PlantDoc-Dataset GitHub repository ( https://github.com/pratikkayal/PlantDoc-Dataset/tree/master/train/Corn%20rust%20leaf , accessed on 3 May 2026). Third, leaf rustOpen asset ↗OpenDataLab/CD_and_Slines:38-47Dataset · publicOpenDataLab, and only the Dataset_Original folder in the raw dataset package was used ( https://opendatalab.com/OpenDataLab/CD_and_S/tree/main , accessed on 3 May 2026). Second, to supplement the leaf rust category, additional images were collected from the train/Corn rust leaf folder of the PlantDoc-Dataset GitHub repository ( https://github.com/pratikkayal/PlantDoc-Dataset/tree/master/train/Corn%20rust%20leaf , accessed on 3 May 2026). Third, leaf rust images from the test folder of the same PlantDoc-Dataset repository were also used ( https://github.com/pratikkayal/PlantDoc-Dataset/tree/master/test , accessed on 3 May 2026). In addition, a small number of self-collected maize leaf images Open asset ↗pratikkayal/PlantDoc-Datasetlines:38-47Dataset · public, additional images were collected from the train/Corn rust leaf folder of the PlantDoc-Dataset GitHub repository ( https://github.com/pratikkayal/PlantDoc-Dataset/tree/master/train/Corn%20rust%20leaf , accessed on 3 May 2026). Third, leaf rust images from the test folder of the same PlantDoc-Dataset repository were also used ( https://github.com/pratikkayal/PlantDoc-Dataset/tree/master/test , accessed on 3 May 2026). In addition, a small number of self-collected maize leaf images were included as negative samples and field-background supplements. Considering that the present study focuses on object detection under complex backgrounds rather than image-level classification under simple-backgOpen asset ↗pratikkayal/PlantDoc-Datasetlines:38-47Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Pakistan is the fourth-largest rice producer and the fifth-largest exporter worldwide. Timely disease detection remains challenging due to the scale of cultivation and reliance on manual monitoring. Developing reliable, ongoing computerized systems for plant health management is essential for efficient disease control. A deep learning approach is used as the core method to identify diseases in rice leaves. This methodology employs a range of advanced deep learning architectures to achieve top-tier feature extraction and classification. The publicly available rice leaf disease dataset on Zenodo supports research reproducibility and data transparency. We systematically process a balanced dataset of 1914 image samples using Python with TensorFlow and a GPU to enable high-speed computation for large-scale image processing. This study conducts a systematic comparative evaluation of five deep transfer learning architectures (InceptionV3, DenseNet201, ResNet152V2, EfficientNetV2L and MobileNetV2) trained independently. The base backbone models are then integrated with guided GrabCut segmentation with contour-detection method for interpretable disease localization. In this work, the methods of segmentation by GrabCut and contour detection are introduced to make the results of the study easier to interpret and explain the disease areas, but the final classification outcomes are obtained only on the basis of the underlying deep transfer learning models. As a result, infected leaf areas can be identified more effectively, allowing for better understanding and explainable of the disease.To enhance interpretability, GrabCut segmentation and contour detection are applied as post-hoc visualization techniques to highlight diseased regions corresponding to CNN predictions. These techniques do not influence the classification training process. All five models InceptionV3, DenseNet201,ResNet152V2,EfficientNetV2L and MobileNetV2 demonstrated their effectiveness in detecting rice diseases during training, validation, and testing phases, with models trained over 30 epochs. The training methods and accuracy rates of the models were compared during validation and final testing. InceptionV3 demonstrated the most moderate performance of 98.80% training, 98.44% validation, and 98.43% test accuracy, which means that it has strong generalization and consistent learning behavior. The performance of very high-density networks such as DenseNet201 (98.72% train, 98.43% val, 98.43% test), ResNet152V2 (99.02% train, 99.22% val, 97.39% test), EfficientNetV2L model accuracies (39.01% train, 48.70% val, 44.50% test) also showed competitive results, which validated the effectiveness of deep transfer learning in the classification of rice leaf disease, while MobileNetV2 model accuracies (98.09% train, 98.18% val, 96.87% test) indicate that a lightweight model can still achieve reliable classification performance with lower computational complexity. In general, the comparative analysis defines InceptionV3 as the most stable and efficient model in the framework proposed. These results illustrate InceptionV3 superior generalization ability, supported by explainable methods for improved feature localization, confirming the viability of transfer learning for accurate and practical rice disease detection using GrabCut segmentation and contour detection technique. The complete implementation code and data used for the research experimentation is publicly available at https://github.com/ummershakeel03/Rice-Leaf-Diseases-Classification for reproducibility and reuse.
Why it matches plant phenotyping methodsイネ葉の病徴領域を画像から分類・局在化する深層学習ワークフローが研究の中心であり、GrabCut・輪郭検出と複数モデルの比較評価を含むため、植物病害状態の画像ベース表現型計測として採用。
abstractA deep learning approach is used as the core method to identify diseases in rice leaves.
Reproduction assets foundThe paper explicitly states that the complete implementation code and the rice leaf disease image dataset (1914 samples) used in this study are publicly available: code on the authors' GitHub repository and the dataset on Zenodo (DOI 10.5281/zenodo.15817084). Both are paper-specific, public, and actionable.Code · publicThe complete implementation code and data used for the research experimentation is publicly available at https://github.com/ummershakeel03/Rice-Leaf-Diseases-Classification for reproducibility and reuse.Open asset ↗ummershakeel03/Rice-Leaf-Diseases-Classificationhtml-lines:1357-1368Dataset · publicThe dataset for this research study is available at: https://doi.org/10.5281/zenodo.15817084.Open asset ↗10.5281/zenodo.15817084html-lines:1357-1368Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Climate change-induced drought increasingly constrains water management in mixed-species urban gardens, requiring scalable and non-destructive approaches. This study proposes an integrated framework combining chlorophyll fluorescence, RGB image indices, and machine learning to classify plant drought response patterns. Ten garden plant species were evaluated under varying soil moisture conditions. Hierarchical cluster analysis integrating fluorescence parameters and RGB indices identified three physiologically defined response clusters, and their reproducibility using RGB indices alone was assessed. A total of 1,629 samples were augmented to 1,881 using the synthetic minority over-sampling technique (SMOTE) to address class imbalance. A support vector machine (SVM) model with a radial basis function kernel, using green leaf index (GLI), normalized green-red difference index (NGRDI), blue-green pigment index (BGI), and soil moisture (%) as predictors, achieved an accuracy of 0.91 and a Kappa coefficient of 0.84. In contrast, PLS-DA showed lower performance (accuracy 0.79, Kappa 0.65), indicating limited separability under linear assumptions. These results demonstrate that RGB indices combined with nonlinear models were able to reproduce physiologically defined drought response patterns under the given conditions. As a proof of concept, this study demonstrates the potential of the proposed framework; however, its generalizability is limited by the controlled greenhouse setting, the relatively small number of species, and the lack of external validation in heterogeneous field environments. The framework may provide a cost-effective approach for classifying plant drought responses and has the potential to support the grouping of plants with similar water requirements, which could contribute to improved irrigation management in mixed-species gardens under further validation.
Why it matches plant phenotyping methodsRGB画像指標と機械学習により、植物の干ばつ応答パターンという生理状態を分類し、蛍光測定との再現性を評価しているため、表現型取得・抽出手法が中心です。
abstractThis study proposes an integrated framework combining chlorophyll fluorescence, RGB image indices, and machine learning to classify plant drought response patterns.
Reproduction assets foundThe paper explicitly states that the authors' analysis code (data processing, feature extraction, SVM/PLS-DA modeling) is publicly deposited on Zenodo with a DOI matching an allowed URL. The phenotype datasets are only said to be in the manuscript/supplementary files, so the code deposit is the qualifying paperSpecificCode · publicThe code supporting the findings of this study, including data processing, feature extraction, and machine learning modeling is available at Zenodo: https://doi.org/10.5281/zenodo.19127295 .Open asset ↗Zenodo · 10.5281/zenodo.19127295lines:98-116Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Background Recently, plant disease detection and classification have become major concerns in agriculture. Early detection of plant diseases supports farmers to take precautionary actions to prevent the spread of infections across different parts of the plant. However, detecting and classifying plant leaf diseases remain challenging tasks due to the overlapping characteristics of different diseases. Methods To mitigate these limitations, this research developed a Multi-FusNet–convolutional neural network (Multi-FusNet–CNN) with an improved Huber loss function to classify multiple classes of plant leaf diseases. Here, a multipath residual network (Multi-RG) with cross-filtering fusion is integrated, and the pixel shuffling fusion method is developed for fusing low-level to up-sampled features. An improved Huber loss function is incorporated into the Multi-FusNet–CNN to effectively handle outliers and enhance the model’s generalization capability during training. Results The developed Multi-FusNet–CNN with improved Huber loss function achieved 99.95% accuracy, 99.13% F1-score, 99.87% recall, 99.27% precision, and 99.93% specificity, thereby outperforming existing conventional techniques. Conclusion The proposed Multi-FusNet–CNN model improved the generalization capability of the method during the training process on plant leaf disease detection and classification.
Why it matches plant phenotyping methods植物葉の病徴を画像から検出・分類するCNN手法の開発と性能評価が研究の中心であり、植物の病害状態を推定するフェノタイピング手法に該当する。
abstractthis research developed a Multi-FusNet–convolutional neural network (Multi-FusNet–CNN) with an improved Huber loss function to classify multiple classes of plant leaf diseases.
Reproduction assets foundThe paper uses two public plant leaf image datasets directly in its analysis: the Plant Village dataset (Kaggle) as the primary training/classification dataset and the RoCoLe dataset (datasetninja) for coffee leaf disease samples. Both have explicit public URLs in the references. No author code, models, or checkpoints,Dataset · public26
Plant village dataset . Available online at: https://www.kaggle.com/datasets/emmarex/plantdisease (Accessed February 03, 2026 ).Open asset ↗Kaggle · emmarex/plantdiseaselines:1010-1131Dataset · public31
RoCoLe dataset . Available online at: https://datasetninja.com/rocoleing (Accessed February 03, 2026 ).Open asset ↗datasetninja · rocoleinglines:1010-1131Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Field / plotLeafClassificationSegmentationStress / disease detectionDisease symptoms / severity
Early and accurate detection of plant leaf diseases is an essential requirement for precision agriculture, given their severe impact on global food security. While much has been done recently, many deep learning-based approaches will still fail in real-world tests because of challenges such as background clutter, differences in illumination, occlusion, or the fact that visual symptoms for these diseases can be very subtle early on. Traditional CNN- and Transformer-based architectures generally lack accurate lesion localisation and interpretability, hindering their practical deployment in agricultural decision-support tools. To address these issues, we present LDDHybridNet, a region-based, explanation-friendly deep learning framework that can identify leaf disease at an early, accurate stage. It then applies preprocessing steps guided by ROI, based on leaf segmentation from the U-Net, followed by a compact CNN-based spatial feature-extraction framework. We arrange spatial feature embeddings extracted from lesion regions into an ordered sequence and employ a Bi-LSTM with attention to model structured contextual dependencies, allowing progression-aware feature learning without requiring actual temporal image sequences. Lastly, Grad-CAM-based post-hoc explainability is employed to interpret model decisions, enabling transparent visualisation of disease-relevant regions. We conduct extensive experiments on the PlantVillage benchmark and the FieldPlant dataset and show that LDDHybridNet consistently outperforms representative CNN, transformer, and hybrid baselines across multiple evaluation metrics. Although the near-ceiling performance on PlantVillage reveals the dataset's artificial nature, the proposed framework achieves 95.37% accuracy under real-world field conditions and 92.84% on weak-lesion early-stage samples, demonstrating the method's robustness and early-stage detection potential. The performance boosts are statistically significant (P < 0.01). In general, LDDHybridNet is an interpretable and robust deep learning framework for leaf disease detection, which can support data-driven crop protection and precision agriculture applications.
Why it matches plant phenotyping methods葉の病害症状を画像から検出・局在化する深層学習手法の開発とベンチマーク評価が中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として収録する。
abstractwe present LDDHybridNet, a region-based, explanation-friendly deep learning framework that can identify leaf disease at an early, accurate stage.
Reproduction assets foundThe paper's phenotyping measurements are leaf disease detection experiments on two public image datasets: PlantVillage (Kaggle) and FieldPlant (IEEE Dataport), both cited with explicit public URLs. The authors' code, trained weights, and scripts are not publicly released and are available only on request, so no code/模型Dataset · public43.Hughes, D. P. & Mohanty, S. P. PlantVillage Dataset. [online] (2015). Available at: https://www.kaggle.com/datasets/emmarex/plantdiseaseOpen asset ↗PlantVillage Datasethtml-lines:657-726Dataset · public44.Moupojou, R. K., Bouachir, W., Ahamed, T. & Taki, A. H. FieldPlant: A Real-World Dataset for Leaf Disease Detection in Field Conditions. IEEE Dataport. [online] (2021). Available at: https://ieee-dataport.org/documents/fieldplant-datasetOpen asset ↗FieldPlanthtml-lines:657-726Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Multispectral / hyperspectralLeafRootMorphology / geometry measurementLeaf traitsRoot system architecture
Premise Selective breeding over thousands of years has prioritized aboveground yield, with little regard for changes belowground. Roots underpin plant growth and resilience, but our knowledge of these critical structures lags behind that of aboveground structures. Accurately phenotyping root traits is labor-intensive, expensive, and often destructive. High-throughput, nondestructive methods are required to advance understanding of the fundamental biology of root systems and to integrate hard-to-measure root traits into breeding programs. Methods We used American licorice (Glycyrrhiza lepidota Pursh.), a perennial legume with a rich ethnobotanical history, as a model to investigate root system phenotypes. We assessed root traits across multiple populations, analyzed relationships between above- and belowground phenotypes, and tested the use of multidimensional leaf traits, including spectral reflectance, in predicting root traits. Results Root traits of American licorice varied significantly across source populations. Root traits were strongly intercorrelated and each root trait correlated with an aboveground phenotype. Leaf spectral reflectance and elemental composition predicted belowground traits; however, interpretation of some trait-specific signals were complicated by isometric scaling between plant size and root traits. Conclusions These findings demonstrate the use of high-dimensional leaf traits as a proxy for root traits, with potential applications for understanding foundational questions in plant biology and in breeding programs targeting belowground structures of perennial herbaceous species. Further optimization and larger studies are needed to improve predictive models.
Why it matches plant phenotyping methods葉の高次元形質とスペクトル反射を用いて、測定困難な根形質を非破壊・高スループットに推定する方法が研究の中心である。
abstractHigh-throughput, nondestructive methods are required to advance understanding of the fundamental biology of root systems and to integrate hard-to-measure root traits into breeding programs.
Reproduction assets foundThe paper's data availability statement points to two public, paper-specific assets: raw root scans on Zenodo and a Figshare deposit containing RhizoVision Explorer output features, CropReporter data and metadata, spectral reflectance data, elemental composition data, and all R code needed to reproduce the analyses. NoDataset · publich Center Bioanalytical Chemistry Facility (RRID:SCR_001047). Finally, we thank the reviewers for their careful evaluation of our manuscript and constructive comments, which helped us clarify the conceptual framing and strengthen the overall quality of the work.
DATA AVAILABILITY STATEMENT
Raw root scans can be found on Zenodo ( https://zenodo.org/records/18852041 ). RhizoVision Explorer output features, CropReporter and associated metadata, spectral reflectance data, elemental composition data, and all R code needed to reproduce the analyses presented in this manuscript can be found on Figshare ( https://doi.org/10.6084/m9.figshare.28742870 ).
REFERENCES
Alahmad , S.
,
D.
Smith
,
C.
KatOpen asset ↗Zenodo · 18852041lines:173-419Dataset · publicILITY STATEMENT
Raw root scans can be found on Zenodo ( https://zenodo.org/records/18852041 ). RhizoVision Explorer output features, CropReporter and associated metadata, spectral reflectance data, elemental composition data, and all R code needed to reproduce the analyses presented in this manuscript can be found on Figshare ( https://doi.org/10.6084/m9.figshare.28742870 ).
REFERENCES
Alahmad , S.
,
D.
Smith
,
C.
Katsikis
,
Z.
Aldiss
,
S. M.
Brunner
,
S. V.
Meer
,
L.
Meijer
, et al. 2025 .
Phenotyping the hidden half: combining UAV phenotyping and machine learning to predict barley root traits in the field
. Journal of Experimental Botany
76 : 5161 ‐ 5178 .
40580084
10.1093/jxb/eraf268
PMC1Open asset ↗Figshare · 10.6084/m9.figshare.28742870lines:173-419Code / dataset availability confirmedCrossref · checked 5 Sept 2026
Disease prevention and water management are important to all the crops, particularly rice and sugarcane production in India. The article proposes a reinforcement learning (RL) based intelligent irrigation management system that is capable of optimising water consumption and crop nutrition in response to the changing agricultural climatic conditions. Decentralised reinforcement learning (RL) is used in a network of irrigation agents that utilise soil and microclimate sensor networks to set the terms of water allocation, water use efficiency (WUE) and crop health. At the same time, deep convolutional networks can be used to differentiate between plant stress/disease and leaf images and take applicable proactive actions. It is a framework that incorporates satellite-derived indices (NDVI, EVI, land surface temperature) with local sensor measurements and image-based health measurements through multimodal deep learning. Far-reaching simulations (including Indian climate and crop calendars) demonstrate that the multi-agent system lowers water consumption and preserves the yields and properly notifies stressed plants. The scores of disease detection with plantvillage-based fine-tuned on rice (120 (3 disease types) and 3829 (5 disease types) and sugarcane (2569 images for all disease types, Convolutional Neural Network (CNN) yield results of >98 % accuracy. Crop mapping (rice/sugarcane) Satellite/LSTM-based crop mapping (with Sentinel-1 / Sentinel-2) achieves more than 97 % accuracy. The suggested structure provides a data-driven, scalable system for precision agriculture to enhance the management of irrigation periods and crop health. Simulation experiments show that the RL-based controller can reduce water consumption while preserving optimal soil moisture levels when compared to rule-based irrigation strategies.
Why it matches plant phenotyping methods画像・衛星・センサーを統合して植物ストレス／病害状態を推定するマルチモーダル基盤が提案され、病害検出性能も評価されているため、植物表現型推定が実質的な構成要素である。
abstractdeep convolutional networks can be used to differentiate between plant stress/disease and leaf images
Reproduction assets foundThe paper reports simulation-based experiments using public leaf-image datasets. The only paper-specific public asset explicitly identified is the Kaggle rice leaf diseases dataset (vbookshelf/rice-leaf-diseases) cited as a data source for the rice disease fine-tuning set. No authors' code, trained models, or data dépDataset · publicConflict of interest: Authors do not have any conflict of interest 2026 Mar 31). Available from: https://www.kaggle.com/datasets/Open asset ↗Kagglepdf-page:16 lines:1-58Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Common beanLeafClassificationStress / disease detectionDisease symptoms / severity
Context Plant diseases are a serious danger to the world's food security since they drastically lower crop output. Traditional manual plant leaf inspection is time-consuming, labor-intensive, and frequently subjective. Recent developments in deep learning provide effective and scalable methods for image-based analysis-based automated plant disease identification. Techniques Three deep learning architectures-a proprietary Convolutional Neural Network (CNN), ResNet18, and Vision Transformer (ViT)-are used in this study to examine automated bean leaf disease identification. The Augmented iBean dataset, which has three classes-angular leaf spot, bean rust, and healthy leaves-was used to train and assess the models. Every model was trained using the same preprocessing and training settings to provide fair benchmarking. Receiver Operating Characteristic (ROC) curves, accuracy, precision, and confusion matrices were used to assess the model's performance. Outcomes ResNet18 fared better than CNN and Vision Transformer models, according to a comparative analysis. ResNet18 maintained a high level of computing efficiency while achieving 99% accuracy and 99.01% precision. Its better categorisation capacity across all disease categories was validated using confusion matrix and ROC analysis. In conclusion The study shows that ResNet18 offers the optimal trade-off between accuracy and efficiency and creates a standard benchmarking framework for bean leaf disease identification. The results demonstrate its applicability for real-time deployment in precision agricultural systems for better crop management and early disease identification.
Why it matches plant phenotyping methods豆葉の病徴を画像から認識する深層学習手法を比較・ベンチマークしており、植物病害状態の取得手法が研究の中心である。
abstractThree deep learning architectures-a proprietary Convolutional Neural Network (CNN), ResNet18, and Vision Transformer (ViT)-are used in this study to examine automated bean leaf disease identification.
Reproduction assets foundThe paper's data availability statement points to the Augmented iBean dataset on IEEE DataPort, the public bean leaf image dataset used for all phenotyping/classification experiments in this study. No author analysis code or trained model checkpoints are shared.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://ieee-dataport.org/documents/bean-leaf-disease-augmented-ibean-dataset.Open asset ↗ieee-dataport · bean-leaf-disease-augmented-ibean-datasethtml-lines:446-496Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Plant diseases threaten global agriculture, and deep learning-based disease recognition has become crucial for addressing this challenge. While DenseNet excels in plant disease classification due to its dense connectivity, its large size limits deployment on resource-constrained edge devices. This paper proposes Connection-Aware DenseNet Pruning (CADP), achieving efficient compression through three collaborative modules. First, the EdgePrune module explicitly models inter-channel feature flows via an edge weight network, using dual-channel importance scoring that fuses activation correlation and gradient information to remove redundant connections while preserving critical propagation paths. Second, connection-guided CP decomposition leverages EdgePrune's importance information, adaptively assigning differentiated ranks through the Connection Importance Index (CII) to balance preservation of critical layers with deep compression of secondary layers. Third, dual-stream knowledge distillation integrates throughout post-pruning and post-decomposition fine-tuning, combining output-level soft labels and intermediate spatial attention transfer to recover compression losses. CADP achieves 88% parameter reduction and 89% computational savings on DenseNet-121, maintaining 99.67% and 99.66% accuracy on PlantVillage and RiceLeaf datasets, achieving competitive accuracy with significantly fewer parameters. This provides a promising approach for resource-constrained deployment with potential generalizability and practical value.
Why it matches plant phenotyping methods植物画像から病害状態を推定する分類モデルの軽量化手法を開発し、PlantVillageおよびRiceLeafで性能を評価しているため、病害フェノタイピング手法が中心である。
titleCADP: Connection-Aware DenseNet Pruning for lightweight plant disease classification.
Reproduction assets foundThe paper uses two publicly available plant image datasets (PlantVillage and Rice Leaf Disease) hosted on Mendeley Data, explicitly linked in the Data Availability statement. No author analysis code, models, or checkpoints are shared.Dataset · publicThe datasets used in this study are publicly available. The PlantVillage dataset can be accessed at https://data.mendeley.com/datasets/tywbtsjrjv, and the Rice Leaf Disease dataset is available at https://data.mendeley.com/datasets/fwcj7stb8r/1.Open asset ↗fwcj7stb8rhtml-lines:698-748Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
This study proposes RoViT-KAN, a multi-task deep learning framework for plant disease classification, severity estimation, and uncertainty quantification. The architecture integrates a DeiT-Tiny Vision Transformer backbone with task-specific heads for disease classification, ordinal severity prediction, and heteroscedastic uncertainty estimation. To enhance interpretability, a Kolmogorov–Arnold Network (KAN) module is introduced to model continuous disease severity through learnable spline-based transformations. A four-stage curriculum learning strategy is employed to stabilize multi-task optimization by progressively activating prediction objectives. The model is evaluated on a rose leaf disease dataset comprising 3,113 original images and 10,000 augmented samples across four classes: healthy leaf, leaf holes, black spot disease, and dry leaf condition. Experimental results demonstrate a classification accuracy of 99.70%, with calibrated uncertainty estimates (Brier score = 0.0914) and reliable severity prediction. Ablation studies validate the contribution of each architectural component. The model highlights the potential of combining transformer-based architectures with uncertainty-aware learning and interpretable neural representations for robust plant disease analysis.
Why it matches plant phenotyping methodsバラ葉の画像から病害分類と病害重症度を推定する手法を開発・評価しており、植物の病態を対象とした画像ベース表現型解析が研究の中心である。
abstractThis study proposes RoViT-KAN, a multi-task deep learning framework for plant disease classification, severity estimation, and uncertainty quantification.
Reproduction assets foundThe paper's rose leaf disease dataset (RoseLeafSet) is publicly deposited on Mendeley Data, and the authors' RoViT-KAN implementation code is publicly available on GitHub, both with explicit availability statements and URLs matching allowed entries.Dataset · publicThe datasets analyzed during the current study are publicly available in the Mendeley Data repository at: https://data.mendeley.com/datasets/9g668bfhy5/3.Open asset ↗Mendeley Datahtml-lines:716-742Code · publicThe code used to develop and evaluate the model in this study is publicly available to support transparency and reproducibility of the research. The implementation, along with relevant scripts and documentation, can be accessed through the following GitHub repository: https://github.com/nishitbohra/RoViT-KAN-Interpretable-Vision-Transformer-for-Rose-Disease-Severity-Estimation.Open asset ↗GitHubhtml-lines:716-742Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Early detection of plant diseases is essential for maintaining crop health and ensuring sustainable agricultural productivity. Guava fruit and leaf diseases, if not identified at an early stage, can lead to significant yield losses. Recent advances in deep learning offer promising solutions; however, challenges remain in achieving both high accuracy and model interpretability for practical agricultural deployment. Methods This study proposes an explainable deep learning-based framework for the classification of guava fruit and leaf diseases. A real-world dataset consisting of 527 annotated images across five classes-Disease Free, Phytophthora, Red Rust, Scab, and Styler and Root Rot-was utilized. Six hybrid model architectures were developed by integrating transfer learning backbones (VGG16, MobileNetV2, InceptionV3, and ResNet50) with custom convolutional neural network (CNN) classifiers. Model performance was evaluated using accuracy, precision, recall, F1-score, and class-wise metrics. To enhance transparency, Gradient-weighted Class Activation Mapping (Grad-CAM) was employed to visualize disease-relevant regions. Results Among all evaluated models, the proposed VGG16 + MobileNetV2 hybrid architecture achieved the best performance, attaining an accuracy of 96%, an F1-score of 0.96, and strong generalization across all disease classes. Comparative analyses using confusion matrices, ROC-AUC curves, precision-recall curves, and radar plots confirmed the superior and consistent performance of the proposed model over other hybrid configurations. Discussion The results demonstrate that combining deep feature extractors with lightweight architectures enhances both classification accuracy and computational efficiency. The integration of Grad-CAM provides meaningful visual explanations, increasing trust and interpretability in AI-assisted disease diagnosis. This framework shows strong potential for deployment in real-time smart farming systems and mobile-based diagnostic applications, particularly in resource-constrained agricultural environments.
Why it matches plant phenotyping methodsグアバの葉・果実画像から病害状態を推定する深層学習手法が研究の中心であり、複数モデルの比較評価とGrad-CAMによる説明可能性検証も行っているため、植物フェノタイピング方法論として採用する。
abstractThis study proposes an explainable deep learning-based framework for the classification of guava fruit and leaf diseases.
Reproduction assets foundThe paper's plant image dataset (527 annotated guava fruit/leaf disease images) is a public Kaggle deposit explicitly cited by the authors with a URL, making it a paper-specific, publicly actionable asset. No author analysis code or trained model checkpoints are stated as publicly available; the data availability only指Dataset · publicKaggle ). Available online at: https://www.kaggle.com/datasets/noamaanabdulazeem/guava-dataset (Accessed January 10, 2024 ).Open asset ↗Kagglelines:550-617Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Monitoring the growth dynamics in field-grown cabbage is critically important for ensuring stable vegetable production and advancing precision agricultural management. However, conventional two-dimensional (2D) image-based monitoring approaches are limited to planar projection information and lack representations of spatial structural characteristics, rendering them inadequate for supporting high-precision, full-cycle phenotypic monitoring of cabbage under open-field conditions. In this study, a high-precision three-dimensional (3D) point cloud dataset covering the period from the seedling stage to maturity was constructed using depth cameras in conjunction with multi-view spatial registration techniques. Building on this dataset, an adaptive point cloud segmentation network designed for the whole-cycle growth monitoring was proposed, incorporating a Head Refinement Module (HRM), a Leaf Instance Segmentation Module (LISM), and Cross Module Interaction (CMI) to address leaf adhesion and head boundary delineation. Experimental results demonstrated that the proposed method consistently outperformed state-of-the-art models in both semantic and instance segmentation tasks. For semantic segmentation, the mean Intersection over Union (mIoU) reached 0.767, with a point classification accuracy of 94.8%. The model comprises 54.25 million parameters and achieves an average response time of 0.76 s. For instance segmentation, the Average Precision (AP) improved by 2.3% for cabbage heads and 3.8% for leaves, while the Average Recall (AR) increased by 6.9%. Growth parameters, including plant height and canopy spread, extracted from the segmentation results showed strong agreement with ground-truth measurements, with correlation of coefficients (R 2 ) exceeding 0.9 for plant height, canopy length, and canopy width. Leveraging these multidimensional phenotypic descriptors, the temporal dynamics of cabbage growth throughout the entire growth cycle were systematically characterized. Overall, this study enables dynamic monitoring of cabbage phenotypes across the full growth cycle, providing a novel technical pathway for extending 3D phenotyping from controlled environments to open-field applications and offering important support for precise crop monitoring and the development of digital twin agriculture.
Why it matches plant phenotyping methods3D点群データセット、セグメンテーションネットワーク、形質抽出を開発・検証し、圃場キャベツの草高や冠幅を定量化する植物フェノタイピング手法が研究の中心である。
abstracta high-precision three-dimensional (3D) point cloud dataset covering the period from the seedling stage to maturity was constructed using depth cameras in conjunction with multi-view spatial registration techniques.
Reproduction assets foundThe paper's authors publicly release their improved OneFormer3D point cloud segmentation code on GitHub; the cabbage 3D point cloud dataset is only available upon request.Code · publicThe code is available at https://github.com/PandaDalin/improve_oneformer3d. The data of this study are available from the corresponding author upon request.Open asset ↗PandaDalin/improve_oneformer3dhtml-lines:449-475Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Objectives Nutritional deficiency in coffee is a major problem that compromises plant health, crop yield, and bean quality, directly threatening the economies of coffee-dependent regions. Traditional detection methods are primarily manual, time-consuming, and relied upon expert availability. Methods This study introduces a novel Deep Learning (DL)-based dual-track architecture designed for the efficient classification of nutritional deficiencies in coffee leaf. The first track utilizes a MobileNetV3 backbone integrated with a Multi-Convolutional Shape-Aware Kernel (MCSK) block to capture spatially adaptive features from leaf textures and vein patterns. The second track employs a Hierarchical Shuffled Group Attention Network (HSGAN), utilizing Efficient Channel Attention (ECA) and Local Group Attention (LGA) modules to balance fine-grained local variations with broad spatial dependencies. Finally, a Multidimensional Collaborative Attention (MCA) mechanism is applied to the fused features to enhance cross-channel interactions and feature extraction. Results The proposed model was evaluated using the CoLeaf dataset, where it achieved an accuracy score of 96.04%. This performance demonstrates an improvement over existing research and current state-of-the-art models, highlighting the architecture's ability to identify complex nutrient-related patterns in coffee leaves. Conclusion The performance of the proposed DL approach offer a solution for the automated monitoring of coffee plants. By providing a reliable alternative to manual inspection, this method presents the potential to help coffee production and support the agricultural regions worldwide.
Why it matches plant phenotyping methodsコーヒー葉の栄養欠乏という植物状態を画像から分類する深層学習手法を開発・評価しており、表現型取得・推定が研究の中心であるため。
abstractThis study introduces a novel Deep Learning (DL)-based dual-track architecture designed for the efficient classification of nutritional deficiencies in coffee leaf.
Reproduction assets foundThe paper's primary phenotyping asset is the CoLeaf coffee leaf nutrient-deficiency image dataset, which the authors state is publicly available via a Mendeley Data URL matching an allowed URL. No author analysis code or trained model checkpoints are disclosed.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://data.mendeley.com/datasets/brfgw46wzb/1 .Open asset ↗brfgw46wzb/1lines:733-764Code / dataset availability confirmedCrossref · checked 14 Sept 2026
The decline in environmental quality caused by industrial pollution and climate change has weakened the natural resistance of rice plants (Oryza sativa), increasing their susceptibility to various diseases. Conventional disease identification methods that rely on manual observation are often limited by subjectivity and human visual constraints. This study proposes a deep learning–based system for automatic rice leaf disease classification using the You Only Look Once version 8 (YOLOv8) architecture. The model was trained using a publicly available rice leaf image dataset consisting of 6,889 images categorized into eight classes: Bacterial Leaf Blight, Brown Spot, Leaf Blast, Leaf Scald, Sheath Blight, Narrow Brown Leaf Spot, Rice Hispa, and Healthy Rice Leaf. The research methodology includes image pre-processing, data augmentation, dataset splitting, and training using the YOLOv8n-cls model for 50 epochs. Experimental results demonstrate high classification performance with an accuracy of 99.5%, precision of 99%, recall of 98%, and an F1-score of 0.99. The trained model was then deployed into a web-based application that allows users to upload rice leaf images and obtain real-time disease classification results. The proposed system provides a practical tool to support early detection of rice plant diseases and assist farmers in improving crop management in modern agriculture.
Why it matches plant phenotyping methodsイネ葉画像から病害状態を推定するYOLOv8画像解析手法の開発と性能評価が中心であり、植物病害フェノタイピングに該当する。
abstractThis study proposes a deep learning–based system for automatic rice leaf disease classification using the You Only Look Once version 8 (YOLOv8) architecture.
Reproduction assets foundThe paper's rice leaf disease image dataset (6,889 images, eight classes) used for YOLOv8n-cls training is a publicly available Kaggle dataset cited by the authors with an explicit URL. No author code, trained model, or other paper-specific assets are reported.Dataset · publicThe primary dataset was obtained from a
publicly available dataset on Kaggle [16], which provides a
comprehensive collection of rice leaf disease images for
machine learning research.Open asset ↗Kagglepdf-raw-page:3 lines:1-102Code / dataset availability confirmedOpenAlex · Crossref · checked 5 Sept 2026
Abstract 3D models are used in plant phenotyping for non-destructive quantification and analysis of morphological characteristics. Analyzing plant structure allows breeders to select for desirable traits, associated with e.g. drought tolerance or increased productivity. In sugar beet, morphological parameters depict an essential element of the variety approval for distinguishing between genotypes. However, only a limited number of measured or scored parameters are considered at a single time point. In contrast, 4D data adds a temporal component and can depict the dynamic development of 3D parameters. To explore the potential of spatio-temporal 4D phenotyping for automated crop genotype differentiation, a greenhouse experiment was conducted by us covering twelve sugar beet genotypes. High-resolution 3D models were generated twice a week over the course of two months and both common and novel 3D morphological parameters were extracted. The importance of these parameters was assessed by us, and the dataset was analyzed using unsupervised pointwise clustering and time series clustering. Varying importance of parameters depending on the time point and significantly higher importance of plant parameters compared to leaf parameters are demonstrated by our results. Moreover, increased and more stable genotype differentiation is archived using time series clustering compared to pointwise clustering. Furthermore, taproot formation of sugar beet was found to have a crucial impact on morphological development. Substantial variations in the dynamic development of 3D morphological parameters underline the importance of 4D data for plant genotype differentiation. Thus, a novel foundation for genotype differentiation in plant phenotyping is provided by our findings.
Why it matches plant phenotyping methods3Dモデルから植物形態形質を抽出し、時系列クラスタリングで遺伝型識別を評価する4Dフェノタイピング手法が研究の中心である。
titleSpatio-temporal 4D phenotyping for automated morphological genotype differentiation of sugar beet
Reproduction assets foundThe paper publicly deposits its generated sugar beet point cloud dataset under CC BY 4.0 at a Dataverse DOI, directly reproducing the paper's phenotyping measurements. Supplementary Python codes and extracted parameter values are stated to be included with the article, but no authors' public URL for the code is presentDataset · publicThe generated point cloud dataset is available at https://doi.org/10.60507/FK2/IS8YBZ under CC BY 4.0 license.Open asset ↗10.60507/FK2/IS8YBZlines:277-363Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published23 Apr 2026International Journal of Drug Delivery TechnologyCited by 0 · OpenAlex ↗
The currently growing effects that plant diseases have on global agriculture require the creation of highly intelligent and accurate detection systems. Convolutional Neural Networks (CNNs), as deep learning, have proved effective in the detection of plant diseases based on images. The CNN performance, however, is very sensitive to hyperparameter tuning, which usually requires manual, sub-optimal tuning. The study suggests a new method to detect plant diseases with an optimized CNN architecture optimized by the Bacterial Colony Optimization (BCO). The BCO algorithm replicates the adaptive foraging behavior of bacterial colonies to automatically determine optimal CNN hyperparameters, such as the number of filters, kernel size, pooling methods, learning rate, and dropout probability of the CNN. The experiment conducted on the PlantVillage dataset demonstrated that the proposed BCO-CNN achieved an accuracy of 96.68% with a false alarm rate (FAR) of 4.05% outperforming other heuristic-fitted models such as particle swarm optimization (PSO) - CNN, CNN with support vector machine (SVM), and CNN, VGG16 in accuracy, precision, recall, and F1-score. The given work offers an automated, scalable approach to the accurate, early detection of the diseases of the plant, contributing to better yields of crops and sustainable agriculture.
Why it matches plant phenotyping methods植物葉画像から病害状態を推定するCNNのハイパーパラメータ最適化手法を開発・評価しており、植物フェノタイピング手法が研究の中心である。
abstractThe study suggests a new method to detect plant diseases with an optimized CNN architecture optimized by the Bacterial Colony Optimization (BCO).
Reproduction assets foundThe paper's plant-phenotyping experiments (CNN/BCO leaf disease detection) were run on the public Kaggle Plant Disease (PlantVillage) dataset, explicitly cited with its public URL. No author code or models are reported as available.Dataset · publictectural and hyperparameter changes. A
comparison of the results also indicates the success of data
augmentation, which indicates whether the model has a better
generalization when the data is more diverse. The MATLAB
2022b was used to implement the proposed method. The
tomato dataset was collected from Plant Disease datasets
(“https://www.kaggle.com/datasets/emmarex/plantdisease").
The average values of 30 independent runs with varying
random seeds are used to report the obtained results. This will
minimize the effects of the chance and provide a fair
representation of the strength of the model. Also, 80 % of the
data is employed in the training and the rest 20 % in the test
purposes. TableOpen asset ↗Kaggle · emmarex/plantdiseasepdf-raw-page:6 lines:1-115Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Accurate classification of corn leaf diseases is critical for timely detection and control of pests and diseases. By accurately recognizing different types of leaf diseases, farmers and agricultural experts can quickly take targeted control measures to reduce crop losses and safeguard corn yield and quality. Since these corn leaf disease images usually contain complex backgrounds, similar lesion features, and limited labeling data, it causes traditional convolutional neural networks (CNNs) to easily confuse the lesion region with the background, making it difficult to distinguish between different disease types. To address these limitations, we propose G-ResNet, a hybrid CNN-Vision Mamba network that enhances disease-relevant feature learning through a hierarchical feature attention module and a scale feature attention module. It was demonstrated experimentally that G-ResNet can better classify maize leaf disease images. The code is available at https://github.com/gustafmy/g_resnet.git.
Why it matches plant phenotyping methodsトウモロコシ葉の病害状態を画像から分類するCNN・Vision Mamba手法の開発と実験評価が研究の中心であり、植物病害フェノタイピングに該当する。
titleCorn or maize leaf disease classification based on CNN and vision mamba model.
Reproduction assets foundThe paper's analysis code (G-ResNet) is publicly available on GitHub with explicit availability statements, and the plant image dataset used for the disease classification experiments is a public Kaggle dataset explicitly cited in the Experiment section. The underlying data availability statement also mentions request,Code · publicSource code for the algorithms described in this paper is available at https://github.com/gustafmy/g_resnet.git.Open asset ↗gustafmy/g_resnethtml-lines:300-328Code / dataset availability confirmedEurope PMC · Crossref · checked 15 Sept 2026
Abstract Current phytopathological diagnostic systems rely on manual inspections or laboratory analyses, which delay early detection and limit in-field responsiveness. Phytopathogenic fungal diseases pose a persistent threat to food security, directly affecting the productivity of essential crops such as potato ( Solanum tuberosum ) and tomato ( Solanum lycopersicum ) [1]–[5]. Among these diseases, Phytophthora infestans , the causal agent of late blight, is characterized by its high virulence and rapid spread, capable of generating significant losses in short periods when detection occurs too late [3], [4]. To address this issue, a computer vision and deep learning–based system for multistage detection of fungal infections in potato and tomato crops is proposed. The system comprises a convolutional neural network optimized for edge processing and a mobile robotic platform equipped with a manipulator arm for localized treatment application. The developed model was deployed on a Raspberry Pi 4 connected to a 12-MP Raspberry Pi Camera Module 3 NoIR, responsible for acquiring RGB images in the field. The proposed network was compared with reference architectures—ResNet-50, VGG16, MobileNetV2, and Inception-v3—within a four-stage detection pipeline: crop identification, health-state classification, infection diagnosis, and foliar severity estimation. A dataset of 18,200 images obtained from publicly accessible online sources, under diverse lighting and background conditions, was used, partitioned into 70% for training, 20% for validation, and 10% for testing. Preliminary results show an average accuracy in the range of 0.90–0.92, with inference latencies below 60 ms per image, ensuring smooth performance on the Raspberry Pi 4 without requiring cloud connectivity. Additionally, the network demonstrated higher sensitivity to visual variations compared to the baseline models.
Why it matches plant phenotyping methods植物病害の健康状態・感染・葉面重症度を画像から推定するコンピュータビジョン手法を開発・比較検証しており、植物表現型取得が中心である。
abstracta computer vision and deep learning–based system for multistage detection of fungal infections in potato and tomato crops is proposed
Reproduction assets foundThe paper's CNN training data are two publicly available third-party plant-disease image datasets (Mendeley Data and Kaggle Plant Village) with explicit URLs in the Data Availability Statement. The authors' field-test images and experimental records are only available on request, and no analysis code or trained model/сDataset · public10
Network, DOI: 10.17632/tywbtsjrjv.1, available at
https://data.mendeley.com/datasets/tywbtsjrjv/1, and the
Kaggle Plant Village dataset, available at
https://www.kaggle.com/datasets/emmarex/plantdisease.The field-test images and experimental records generated during
the current study during the real-world evaluation of the
embedded-vision system are available from the corresponding
author on reasonable request.
IX. REFERENCOpen asset ↗10.17632/tywbtsjrjv.1pdf-raw-page:11 lines:1-98Dataset · public10
Network, DOI: 10.17632/tywbtsjrjv.1, available at
https://data.mendeley.com/datasets/tywbtsjrjv/1, and the
Kaggle Plant Village dataset, available at
https://www.kaggle.com/datasets/emmarex/plantdisease.The field-test images and experimental records generated during
the current study during the real-world evaluation of the
embedded-vision system are available from the corresponding
author on reasonable request.
IX. REFERENCES
[1] P. W. Crous, A. Y. Rossman, M. C. Aime, W. C. Allen, T.
Burgess, J. Z. Groenewald y L. A. CastlebuOpen asset ↗Kagglepdf-raw-page:11 lines:1-98Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Accurate identification of maize diseases is crucial for safeguarding global food security. Traditional image-based methods often struggle with lighting variations, occlusions, and noise, limiting their robustness and generalisation. Multimodal approaches that integrate visual and textual information have shown promise. However, these methods frequently require manually curated textual descriptions for each image, increasing data collection costs and limiting scalability and practical implementation. To address these limitations, we proposed a maize image-text framework with Cross-Modal Category Alignment (mIT-CMCA). This approach enforces category-level alignment between image and text modalities, enabling more accurate and interpretable cross-modal mapping. First, we construct cross-modal representations by aligning image and text modalities at the category level within a shared embedding space. Second, inspired by contrastive learning, we introduce a Cross-Modal Category Alignment (CMCA) loss based on category-level textual descriptions, reducing annotation complexity. Finally, we present an Efficient Channel-Spatial Hybrid Attention (CSHA) module that preserves inter-class boundaries while incurring minimal computational overhead, thereby enhancing feature discriminability under complex conditions. Experimental results on the maize subset of the PlantVillage dataset (MPVD) show that mIT-CMCA achieves 99.48% accuracy, 99.28% precision, 99.54% recall, and 99.41% F1-score. These results represent improvements of 0.24%, 0.13%, 0.17%, and 0.15% over the strongest vision-only baseline, MaxViT_tiny. On the self-built Maize Leaf-Field dataset (MLFD), the model achieves 93.67% accuracy, 93.76% precision, 93.67% recall, and 93.71% F1-score. It uses only 8.27 million parameters, which is 71.6% fewer than MaxViT_tiny. Its model size is 32.13 MB, which is 72.3% smaller. The proposed method also outperforms comparative models in robustness experiments under artificially added perturbations. These results demonstrate that mIT-CMCA achieves a favorable balance between accuracy and efficiency, making it suitable for practical agricultural deployment.
Why it matches plant phenotyping methodsトウモロコシ葉画像から病害状態を推定する画像・マルチモーダル手法の開発と性能評価が研究の中心であり、植物病害フェノタイピングに該当する。
abstractwe proposed a maize image-text framework with Cross-Modal Category Alignment (mIT-CMCA).
Reproduction assets foundThe paper's maize disease identification analysis code and trained models are explicitly stated as publicly available in a GitHub repository. The phenotype image datasets (MPVD subset and self-built MLFD) are not publicly available and require contacting the corresponding author.Code · publicCode availability
The code and models are available in the GitHub repository at https://github.com/TANGFEILONG626/mIT-CMCA..Open asset ↗TANGFEILONG626/mIT-CMCAhtml-lines:673-695Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Rapid population growth and climate change have intensified the need for sustainable agricultural productivity. Plant leaf diseases significantly impact the crop yield, quality, and food safety, necessitating accurate and automated detection methods. Methods This study proposes a deep learning (DL)-based framework for automated detection and classification of tomato and soybean leaf diseases. The proposed framework is trained and evaluated over a large-scale datasets comprising 16,012 tomato leaf images and 6,410 soybean leaf images. Multiple convolutional neural network (CNN) models, including DenseNet121, MobileNetV2, and InceptionV3, are employed for classification. Object detection is performed using YOLOv12. To enhance interpretability, Gradient-Weighted Class Activation Mapping (Grad-CAM) is integrated. Furthermore, a novel Hybrid Attention-Based Stacking Ensemble Model is developed using ResNet152V2, VGG19, and EfficientNetB0, combined with Convolution Block Attention Module (CBAM) and spatial attention mechanisms. Results The CNN models achieved classification accuracies of 97% for DenseNet121, 98% for MobileNetV2, and 99.94% for InceptionV3. YOLOv12 attained a mean average precision (mAP) of 99.5%. The proposed hybrid ensemble model achieved an accuracy of 99.18%, demonstrating improved feature learning through combined channel and spatial attention. Grad-CAM visualizations confirmed that the model effectively identifies the disease-relevant regions. Discussion The results indicate that the proposed framework has attained a high accuracy, robustness, and interpretability for plant disease detection. The integration of attention mechanisms and explainable AI enhances model reliability and transparency. This framework shows a strong potential for the real-time agricultural monitoring, although further validation across diverse crops and real-world field conditions is required.
Why it matches plant phenotyping methods植物葉の病害状態を画像から自動検出・分類する深層学習フレームワークの開発であり、病害表現型の取得・推定が研究の中心。
abstractThis study proposes a deep learning (DL)-based framework for automated detection and classification of tomato and soybean leaf diseases.
Reproduction assets foundThe paper's leaf-disease classification/detection experiments are built on two public Kaggle image datasets cited by the authors as the study's data sources: a soybean leaf dataset (Patil 2024) and a tomato leaf disease dataset (Rex 2019). No authors' analysis code, trained model checkpoints, or paper-specific phenotypDataset · publicPatil A. ( 2024 ). Soyabean-Latest Dataset (
Kaggle ). Available online at: https://www.kaggle.com/datasets/adityapatil1205/soyabean-latestOpen asset ↗Kagglelines:1523-1646Dataset · publicRex E. ( 2019 ). Plant Disease Dataset (Tomato Leaf Diseases) (
Kaggle ). Available online at: https://www.kaggle.com/datasets/emmarex/plantdisease197Open asset ↗Kagglelines:1523-1646Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Wood apple (Feronia limonia L.) is an underutilized perennial fruit tree with substantial ecological, nutritional, and economic potential, yet its phenotypic diversity and trait organization remain poorly characterized. Here, we applied a mixed-scale multivariate framework to resolve phenotypic structure in 62 wood apple genotypes using 31 ordinal and categorical vegetative, leaf, floral, fruit, and seed descriptors. Trait interrelationships were examined through the complementary use of Spearman’s rank correlation and Cramér’s V association analyses, capturing both directional rank-based dependencies and scale-independent categorical linkages. Hierarchical clustering based on Gower distance separated the genotypes into three distinct phenotypic clusters, with inter-cluster dissimilarities (0.92–1.18) consistently exceeding intra-cluster variation (0.42–0.55), indicating well-supported phenotypic stratification based on cluster validation. Multiple Correspondence Analysis (MCA) explained 23.30% of total inertia across the first two dimensions, with tree growth habit, branch angle, tree shape, and fruit color emerging as the principal drivers of phenotypic differentiation. Vegetative and leaf traits formed a tightly integrated module, whereas fruit-related traits displayed weaker monotonic but persistent categorical associations, reflecting partial phenotypic independence. The strong concordance among association analyses, clustering, and MCA indicates structured patterns of coordinated and partially independent trait associations in wood apple. Overall, this study demonstrates the effectiveness of mixed-scale multivariate approaches for resolving complex trait architecture in underutilized perennial fruit crops and provides a quantitative phenotypic framework to support germplasm conservation, parent selection, and ideotype-oriented improvement of wood apple.
Why it matches plant phenotyping methods混合尺度の多変量解析を用いて植物遺伝資源の表現型構造を定量化する手法が研究の中心であり、単なる生物学的実験の routine 測定ではない。
abstractHere, we applied a mixed-scale multivariate framework to resolve phenotypic structure in 62 wood apple genotypes using 31 ordinal and categorical vegetative, leaf, floral, fruit, and seed descriptors.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Supplement · publicAll data generated or analyzed during this study are available in the article and the accompanying Supplementary Table S1.Open asset ↗lines:137-161Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
BACKGROUND: Leaf inclination angle (LIA) is a key trait affecting crop canopy structure and photosynthetic efficiency, but its accurate measurement is challenging due to complex leaf geometry, especially in narrow, curved rice leaves. As the flag leaf serves as the primary photosynthetic organ in rice, the precise spatial parsing of its architecture is crucial for optimizing canopy light interception and yield potential. With the rapid development of high-throughput phenotyping technologies, an increasing number of studies have focused on the fine-grained characterization of 3D crop architecture. However, accurate methodologies for extracting the flag leaf inclination angle (FLIA) in rice, as well as systematic investigations into its spatiotemporal variation patterns, remain largely unexplored. RESULTS: In this study, we systematically evaluated multiple plane-fitting strategies based on SfM-MVS point clouds, finding that voxel-based piecewise analysis outperformed traditional global approaches. To further improve accuracy, skeleton extraction methods were innovatively extended to LIA estimation. A proposed multi-method ensemble, based on the median of eight skeleton extraction combinations, yielded high robustness (R2 = 0.923, RMSE = 2.072°) against photographic ground truth. By applying the proposed framework to both field- and pot-grown rice, we observed no significant FLIA differences between varieties or nitrogen treatments under field-grown conditions, likely due to phenotypic plasticity regulated by population effects. However, pot-grown plants, experiencing reduced interplant competition, exhibited significant varietal differences in FLIA. Across growth environments, varieties, and nitrogen treatments, FLIA at maturity was significantly lower than at anthesis and grain filling stages due to leaf senescence. CONCLUSIONS: This study establishes a robust and accurate measurement framework for LIA based on 3D point clouds, improving estimation performance through piecewise analysis, voxelization, and ensemble strategies. The proposed approach is demonstrated to be an effective tool for the precise quantification of rice leaf phenotypes.
Why it matches plant phenotyping methodsSfM-MVS点群からイネ葉の傾斜角を抽出する手法を開発・検証し、圃場および鉢植えで適用しているため、植物フェノタイピング手法が研究の中心である。
abstractA proposed multi-method ensemble, based on the median of eight skeleton extraction combinations, yielded high robustness (R2 = 0.923, RMSE = 2.072°) against photographic ground truth.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicThe python program, complete dataset, including the original two-dimensional images and corresponding piecewise measurement trajectories, is publicly available at https://github.com/Interstingsun/LIA (accessed on 6 February, 2026).Open asset ↗Interstingsun/LIAlines:77-83Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Background Accurate segmentation of foliar diseases under field conditions is essential for large-scale phenotyping, as breeding programs rely on reliable severity estimates to identify genotypes with improved resistance. However, most deep learning approaches have been developed as pathogen-specific models, which limits scalability in field-grown barley where multiple diseases naturally co-occur and exhibit substantial visual similarity. Results We evaluated whether a multiclass segmentation model can simultaneously detect and distinguish two fungal diseases of barley, Puccinia hordei and Ramularia collo-cygni, and compared its performance with two disease-specific binary models. Using 336 high-resolution leaf scans collected in the field with naturally occurring co-infections, the multiclass model achieved higher Dice scores for brown rust (0.59 vs 0.40; +47.5% relative improvement) and ramularia (0.60 vs 0.53; +13.2% relative improvement). It also captured a greater proportion of individual lesions across both classes. At the genotype level, the model-predicted disease area percentages were highly consistent with those from ground truth annotations ([Formula: see text]). Conclusions A unified multiclass framework can more effectively segment visually similar foliar diseases than separate binary models, while simplifying the computational workflow. This provides a scalable basis for automated resistance assessment within breeding pipelines. Code and data are publicly available at https://github.com/grimmlab/BarleyDiseaseSegmentation, with Mendeley Data dataset DOI 10.17632/4ny92p2r8f.1.
Why it matches plant phenotyping methods圃場画像から葉面病害面積をセグメンテーションし、遺伝子型レベルの病害重症度を推定する手法を開発・比較・検証しており、植物フェノタイピングが中心です。
abstractAccurate segmentation of foliar diseases under field conditions is essential for large-scale phenotyping
Reproduction assets foundThe paper's annotated barley leaf disease segmentation dataset (Mendeley Data DOI 10.17632/4ny92p2r8f.1) and the authors' analysis/segmentation code (GitHub grimmlab/BarleyDiseaseSegmentation) are explicitly declared publicly available, directly reproducing this paper's phenotyping measurements and computational modelsDataset · publicThe annotated dataset and the code implementing our machine learning–based model are publicly available on Mendeley Data (https://doi.org/10.17632/4ny92p2r8f.1) and GitHub (https://github.com/grimmlab/BarleyDiseaseSegmentation).Open asset ↗Mendeley Data · 10.17632/4ny92p2r8f.1lines:133-140Code · publicCode and data are publicly available at https://github.com/grimmlab/BarleyDiseaseSegmentation, with Mendeley Data dataset DOI 10.17632/4ny92p2r8f.1.Open asset ↗GitHub · grimmlab/BarleyDiseaseSegmentationlines:1-70Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Plant diseases are a major problem for farmers around the world, reducing crop yields. The absence of expertise makes plant disease detection difficult and complicated. Plant disease detection is made easier by deep learning algorithms; however, they are computationally demanding and need huge training datasets. This research work proposes a novel Conv-7 DCNN model with modified ParNet attention layer to classify plant leaves into distinct categories with improved accuracy. Because of its architecture, the proposed network can identify leaf diseases with more accuracy by extracting the wider range of features from the images. The proposed Conv-7 DCNN model classifies the leaf diseases of three plants such as tomato, potato, and pepper-bell into fifteen categories. The CNN model is trained using publicly accessible Kaggle dataset, utilising image augmentation techniques. It is evident from the simulation results that the proposed model outperforms several pre-trained, and other trending deep learning models. Proposed model achieves 99.18% classification accuracy with an average precision of 99.17% and area under the curve (AUC) of 1, making this model highly effective in leaf diseases detection. Additionally, Conv-7 DCNN achieved high FPS of 112.49, low inference time of 18.34 s, and low GFLOPS of 13.98, making it suitable for real-time applications in smart agriculture systems.
Why it matches plant phenotyping methods植物葉画像から病害状態を推定する深層学習モデルを開発・比較しており、病害表現型の取得・分類手法が研究の中心である。
abstractThis research work proposes a novel Conv-7 DCNN model with modified ParNet attention layer to classify plant leaves into distinct categories with improved accuracy.
Reproduction assets foundThe paper's plant-phenotyping measurements (leaf disease classification of tomato, potato, and pepper-bell) are based on a publicly available Kaggle dataset explicitly named in the Data Availability statement. No author analysis code, trained model checkpoints, or other paper-specific assets are disclosed.Dataset · publicThe dataset is available online at https://www.kaggle.com/datasets/emmarex/plantdisease.Open asset ↗Kaggle · emmarex/plantdiseasehtml-lines:824-854Code / dataset availability confirmedCrossref · Europe PMC · OpenAlex · checked 5 Sept 2026
Automated quantification of plant-level development from multi-plant greenhouse scenes requires separating individual plants from shared scene-level reconstructions and quantifying organ-level development, a challenge that single-plant acquisition workflows do not directly address. This study presents an end-to-end phenotyping pipeline built on 3D Gaussian Splatting (3DGS) and a post-reconstruction extraction framework, LCR-GS, designed to isolate plant instances from full greenhouse scenes without scene-specific model retraining. LCR-GS integrates zero-shot 2D cues with multi-view lifting, geometric clustering, and chromatic refinement to convert large scene-level reconstructions (~2M Gaussians) into compact per-plant subsets (~16K Gaussians). Experiments on greenhouse-grown muskmelon at the early vegetative stage demonstrate high plant-extraction precision (0.933) and strong organ-level instance segmentation (mean AP50 = 0.924). Plant height and leaf count are validated against manual measurements (height R² = 0.98, RMSE = 1.88 cm; leaf count R² = 0.86), whereas additional morphological traits, including leaf area, leaf area index, mean internode length, and stem node count, are reported as pipeline-derived descriptors for within-cohort comparison. By decoupling semantic inference from reconstruction, the pipeline reduces scene-scale data by over 99% and provides a practical route to derive compact per-plant 3D representations from multi-plant greenhouse imagery for downstream organ-level analysis.
Why it matches plant phenotyping methods3DGS画像から個体・器官を抽出し、植物形質を定量化するフェノタイピング手法の開発と検証が中心である。
abstractThis study presents an end-to-end phenotyping pipeline built on 3D Gaussian Splatting (3DGS) and a post-reconstruction extraction framework, LCR-GS, designed to isolate plant instances from full greenhouse scenes without scene-specific model retraining.
Reproduction assets foundThe paper's Data Availability Statement points to a public GitHub repository containing the muskmelon 3DGS phenotyping dataset (scenes, Gaussian-level plant/background annotations, and point-level organ labels) used in this study.Dataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://github.com/bblabNTU/3dgs-muskmelon-phenotyping-dataset.Open asset ↗bblabNTU/3dgs-muskmelon-phenotyping-datasethtml-lines:485-547Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Accurate detection of agricultural diseases and pests is essential for crop protection and food security worldwide. Real-world field applications face challenges including small objects, complex backgrounds, high class similarity, and limited computational resources. This work proposes a Random Rolling Transformer (RRT), which introduces random circular shifts along channel and sequence dimensions into multi-head self-attention to enrich feature interactions without increasing parameters or computation. Integrated into YOLOv12, the proposed RRT-YOLO is evaluated on the IP102 pest dataset and a tomato leaf disease dataset. Results show that RRT-YOLO improves mAP@50 by 2.5% and mAP@50–95 by 3.9% on IP102, and by 8.8% and 4.2% on the tomato disease dataset, while maintaining identical model size and complexity. This attention perturbation strategy offers an effective and efficient solution for lightweight agricultural vision detection and can be extended to other visual computing tasks. The code and detailed descriptions can be accessed via the following repository: \href{https://github.com/glorioustory/Random-Rolling-Attention-Augmentation.git}{https://github.com/glorioustory/Random-Rolling-Attention-Augmentation.git}.
Why it matches plant phenotyping methods植物葉の病害状態を画像から検出する計算手法を開発・評価しており、植物病害の表現型取得が技術的中心である。害虫検出も含むが、トマト葉病害データセットで手法性能を検証している。
abstractThis work proposes a Random Rolling Transformer (RRT), which introduces random circular shifts along channel and sequence dimensions into multi-head self-attention to enrich feature interactions without increasing parameters or computation.
Reproduction assets foundThe authors explicitly state their code and detailed descriptions are publicly available in a GitHub repository containing the RRT-YOLO implementation used for the paper's disease/pest detection experiments. The IP102 and tomato leaf disease datasets are third-party public resources, not paper-specific assets.Code · publicThe code and detailed descriptions can be accessed via the following repository:
https://github.com/glorioustory/Random-Rolling-Attention-Augmentation.git.Open asset ↗https://github.com/glorioustory/Random-Rolling-Attention-Augmentation.gitpdf-page:3 lines:1-51Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
LeafStem / branchPhysiological trait estimationLeaf traitsWater status / transpiration
Leaf and hydraulic traits are key determinants of growth rates, and hence potentially exhibit significant associations with wood density (WD) and its intraspecific variation (ITV). However, the extent to which functional traits could improve WD prediction accuracy, and how ITV in WD correlates with functional traits remain incompletely understood. We investigated WD and its ITV across 10,218 plant species, mapped the global distribution of WD, and analyzed the association of ITV in WD with niche breadth and functional traits. Plant species with an acquisitive resource-use strategy, characterized by higher specific leaf area (SLA), leaf nitrogen concentration (LN), and leaf maximum stomatal conductance (g max ), exhibited lower WD. Associations of WD with hydraulic traits indicated species with greater hydraulic safety exhibited higher WD. Moreover, the integration of leaf traits (i.e., SLA and LN) and hydraulic traits with environmental factors substantially enhanced WD prediction accuracy in a random forest model, raising the explained variance from 55% to 95%. Furthermore, resource-acquisitive species demonstrated higher ITV for WD. ITV was positively related to relative niche breadth concerning both climatic factors and soil properties. Overall, functional traits significantly improve WD prediction accuracy, and plant species with an acquisitive resource-use strategy exhibit lower WD but greater intraspecific variation.
Why it matches plant phenotyping methods木材密度という植物形質の予測モデルを構築し、機能形質・環境因子の統合による予測精度を検証しており、形質推定手法が中心的です。
abstractthe integration of leaf traits (i.e., SLA and LN) and hydraulic traits with environmental factors substantially enhanced WD prediction accuracy in a random forest model, raising the explained variance from 55% to 95%.
Reproduction assets foundThe paper's Data Availability Statement points to a public Zenodo deposit containing the authors' global wood density distribution data, which directly reproduces this paper's measurements. The TRY Plant Trait Database is a generic third-party database, not a paper-specific asset, and no author analysis code is stated.Dataset · publicData for the global distribution of wood density is available on Zenodo Repository https://sandbox.zenodo.org/records/425279.Open asset ↗Zenodo · 425279html-lines:405-429Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Climate change, particularly increasing frequency and intensity of spring frost events, poses a serious threat to viticulture by reducing yield and product quality. This study proposes an image processing and machine learning-based framework for early, rapid, and accurate segmentation of frost damage in vineyards using YOLOv11s enhanced with Atrous Spatial Pyramid Pooling (ASPP). A unique dataset called FGVL dataset from Sultana seedless grape vineyards in Manisa, Türkiye, following a severe frost event in April 2025. FGVL includes 418 frost-damaged grapes, 510 frost-damaged leaves, 395 healthy grapes, and 698 healthy leaves, all manually annotated by experts under natural field conditions. By integrating ASPP into YOLOv11s, proposed model improved multi-scale contextual feature extraction and achieved mAP@50 of 0.7686, demonstrating stronger performance in instance segmentation of small, overlapping, and visually similar grapevine organs. In addition, Dynamic Confidence Thresholding (DCT) strategy was introduced to improve prediction reliability in dense and visually complex vineyard scenes. Despite challenges such as background clutter, object overlap, and small target structures, model maintained stable performance with low computational demand, requiring only 6.45 GB of GPU memory. Proposed framework offers an accurate, efficient, and practically deployable early recognition system for frost damage assessment in viticulture.
Why it matches plant phenotyping methodsブドウの器官における霜害状態を画像からセグメンテーションする手法を開発・評価しており、植物の病害・障害状態の取得が研究の中心である。
abstractThis study proposes an image processing and machine learning-based framework for early, rapid, and accurate segmentation of frost damage in vineyards using YOLOv11s enhanced with Atrous Spatial Pyramid Pooling (ASPP).
Reproduction assets foundThe paper's Data availability statement explicitly shares the FGVL frost-damage dataset and source code in the corresponding author's public GitHub repository, matching an allowed URL.Code · publicSource code and dataset are publicly shared in GitHub repository of corresponding author. GitHub repo: https://github.com/kaanarikk/Grape-Instance-Segmentation-For-ViticultureOpen asset ↗https://github.com/kaanarikk/Grape-Instance-Segmentation-For-Viticulturelines:230-236Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Purpose/significance Sugarcane is a vital global crop, critical for sugar and energy production. The accurate and timely identification of its leaf diseases is paramount for sustaining the health and stability of the sugarcane industry. While deep learning models offer promising solutions, their deployment on mobile or edge devices is often hindered by substantial model size and high computational demands. Conversely, existing lightweight models frequently compromise on feature extraction capabilities and recognition accuracy. To bridge this gap, this study develops an architecturally improved lightweight model designed to achieve both high accuracy and computational efficiency. Methods We propose the ReMA-MobileViT model, which significantly enhances feature representation by incorporating a newly designed Residual Multi-head Attention (ReMA) module. This module ingeniously leverages a multi-head attention mechanism to capture richer contextual information from diverse subspaces, while its residual connection structure effectively mitigates network degradation and facilitates robust gradient flow. The proposed model underwent rigorous training and evaluation on a comprehensive Mendeley Data repository for classification tasks. Results Experimental evaluations demonstrate that the ReMA-MobileViT model achieves an outstanding classification accuracy of 99.02% on the sugarcane leaf disease dataset, substantially surpassing existing state-of-the-art methods. An ablation study confirms the module's efficacy, showing that the ReMA-MobileViT model, integrated with the ReMA module, improved accuracy, recall, and F1-Score by 1.58, 1.76, and 1.58 percentage points, respectively, over the baseline MobileViT. Comparative analyses further illustrate ReMA-MobileViT's superior overall performance; it exceeds classic lightweight MobileNetV2 by 15.77 percentage points and the mainstream Vision Transformer by 2.96 percentage points in accuracy. Critically, ReMA-MobileViT achieves this with significantly fewer model parameters and reduced computational complexity compared to Vision Transformer, establishing a superior balance between accuracy and efficiency. Conclusion The proposed ReMA-MobileViT model offers an effective and lightweight solution for improving sugarcane leaf disease recognition accuracy, particularly in challenging complex backgrounds. Its ability to balance high accuracy with computational efficiency presents a promising technical avenue and a deployable solution for high-precision crop disease diagnosis systems on resource-constrained mobile or edge platforms.
Why it matches plant phenotyping methodsサトウキビ葉の病害状態を画像から認識する軽量深層学習モデルを開発し、精度・計算量・アブレーションを評価しており、植物表現型取得・判定手法が中心である。
abstractWe propose the ReMA-MobileViT model, which significantly enhances feature representation by incorporating a newly designed Residual Multi-head Attention (ReMA) module.
Reproduction assets foundThe paper's sugarcane leaf disease image dataset (2022 Sugarcane Leaf Disease Dataset, Thite et al.) is publicly available on Mendeley Data and directly constitutes the image inputs used for the paper's disease recognition experiments. No author analysis code or trained model checkpoints are reported.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://data.mendeley.com/datasets/9424skmnrk/1 .Open asset ↗9424skmnrklines:757-778Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Introduction Accurate plant disease identification is of great importance for ensuring agricultural productivity and food security. However, complex illumination variations, leaf occlusion, and diverse disease spot scales throughout plant growth stages significantly increase the difficulty of real-time detection, leading to limited accuracy and robustness in existing approaches. Methods To address these challenges, we propose an improved RT-DETRv2-based plant disease detection model, termed AF-RT-DETR. A Bidirectional Cross Gate (BCG) module is introduced in the feature extraction stage to reduce channel redundancy and enhance discriminative feature representation through multi-level feature interactions. The original RepVGG structure is replaced with a Dynamic Channel Shift (DCS) module, effectively enlarging the receptive field and strengthening contextual feature fusion without additional computational overhead. Additionally, an improved Scale-aware Multi-level Loss (SML) emphasizes low-quality feature maps to improve detector robustness. Results The model achieves mAP50 and mAP50:95 of 93.6% and 67.2% on the Plant-Disease dataset, surpassing the baseline by 5.1% and 4.5%. Furthermore, the model was evaluated on multiple crops and growth stages under diverse field conditions, demonstrating robust performance and adaptability. Discussion These results indicate that AF-RT-DETR effectively enables real-time plant disease detection in complex field environments.
Why it matches plant phenotyping methods植物の病徴を画像から検出するモデルの開発と、複数作物・生育段階・圃場条件での性能評価が中心であり、植物病害状態の表現型計測手法に該当する。
Reproduction assets foundThe paper evaluates AF-RT-DETR on three public Roboflow plant-disease image datasets, each cited with an explicit public URL. No author analysis code or trained model release is mentioned. The Ultralytics YOLOv8 repository is a generic third-party library, not a paper-specific asset.Dataset · publicRoboflow Detecting rice crop diseases object detection dataset . Available online at: https://universe.roboflow.com/crop-diseases-l2qhk/detecting-rice-crop-diseases/dataset/21Open asset ↗lines:816-932Dataset · publicRoboflow Disease detection object detection dataset . Available online at: https://universe.roboflow.com/projects-h0apg/disease-detection-0slunOpen asset ↗lines:816-932Dataset · publicRoboflow Plant Disease v2 512×512 . Available online at: https://universe.roboflow.com/sangeeth-mathew-john-nl43i/plant-disease-czcfe/dataset/2Open asset ↗lines:933-1045Code / dataset availability confirmedOpenAlex · arXiv · checked 15 Sept 2026
Multispectral / hyperspectralLeafCalibration / preprocessingSegmentationVisualization / data management
Hyperspectral imaging (HSI) allows researchers to study plant traits non-destructively. By capturing hundreds of narrow spectral bands per pixel, it reveals details about plant biochemistry and stress that standard cameras miss. However, processing this data is often challenging. Many labs still rely on loosely organized collections of lab-specific MATLAB or Python scripts, which makes workflows difficult to share and results difficult to reproduce. MVOS_HSI is an open-source Python library that provides an end-to-end workflow for processing leaf-level HSI data. The software handles everything from calibrating raw ENVI files to detecting and clipping individual leaves based on multiple vegetation indices (NDVI, CIRedEdge and GCI). It also includes tools for data augmentation to create training-time variations for machine learning and utilities to visualize spectral profiles. MVOS_HSI can be used as an importable Python library or run directly from the command line. The code and documentation are available on GitHub. By consolidating these common tasks into a single package, MVOS_HSI helps researchers produce consistent and reproducible results in plant phenotyping
Why it matches plant phenotyping methods葉レベルHSIの校正・葉検出・切り出しを含む再現可能な植物表現型解析用ソフトウェアであり、手法が中心。
abstractMVOS_HSI is an open-source Python library that provides an end-to-end workflow for processing leaf-level HSI data.
Reproduction assets foundThis is a software paper describing MVOS_HSI, the authors' open-source Python library for hyperspectral plant-phenotyping preprocessing (calibration, leaf segmentation/clipping, augmentation, spectral plotting). The authors' code is explicitly and publicly available on GitHub at the allowed URL, making it a paper-phenyCode · publicyping.
K eywords Hyperspectral imaging ⋅ \cdot
Plant phenotyping ⋅ \cdot
Data preprocessing ⋅ \cdot
Vegetation indices ⋅ \cdot
Data augmentation ⋅ \cdot
Python
Table 1: Code Metadata for MVOS_HSI
Nr.
Code metadata description
Metadata
C1
Current code version
v0.2.1
C2
Permanent link to code/repository used for this code version
https://github.com/MVOSlab-sdstate/mvos_hsi
C3
Permanent link to Reproducible Capsule
N/A
C4
Legal Code License
MIT License
C5
Code versioning system used
git
C6
Software code languages, tools, and services used
Python 3.x; NumPy, SciPy, Matplotlib
C7
Compilation requirements, operating environments & dependencies
Standard scientific Python environment on Windows, LinOpen asset ↗MVOSlab-sdstate/mvos_hsi · mvos_hsilines:1-122Code / dataset availability confirmedCrossref · checked 5 Sept 2026
Abstract. Large-scale mapping of plant biophysical and biochemical traits is essential for ecological and environmental applications. Given their finer spectral resolution and unprecedented data availability, hyperspectral data, in concert with machine and particularly deep learning models, have emerged as a promising, non-destructive tool for accurately retrieving these traits. However, when deploying these methods on a large scale, reliably quantifying the associated uncertainty remains a critical challenge, especially when models encounter out-of-domain (OOD) data, i.e., samples that differ substantially from those of the training data, such as unseen geographical regions, species, biomes, data acquisition modalities, or scene components (e.g., clouds and water bodies). Traditional uncertainty quantification methods for deep learning models, including deep ensembles (deterministic and probabilistic) and Monte Carlo dropout, rely on the variance of predictions but often fail to capture uncertainty in OOD scenarios, leading to overly optimistic and possibly misleading uncertainty estimates. To address this limitation, we propose a distance-based uncertainty estimation method (Dis_UN) that quantifies prediction uncertainty by measuring the dissimilarity in the predictor space (spectral inputs) and embedding space (features learned by the deep model) between the training and test data. Dis_UN leverages residuals as a proxy for uncertainty and employs dissimilarity indices in data manifolds to estimate worst-case errors via 95-quantile regression. We evaluate Dis_UN using a pretrained deep learning model to predict multiple plant traits from hyperspectral images, analyzing its performance across OOD data, such as pixels containing spectral variations from urban surfaces, bare ground, water, clouds, or open surface waters. In this study, we target six leaf and canopy traits: leaf mass per area, chlorophylls, carotenoids, nitrogen content, equivalent water thickness, and leaf area index. Compared to scaled variance-based methods, Dis_UN provides (1) a superior estimation of uncertainty in OOD scenarios, achieving 36 % higher contrast (KS distances: 0.648 vs. 0.475) between non-vegetation pixels, particularly under mixed-pixel conditions at medium resolution (30 m); (2) uncertainty quantification without requiring normality or symmetry assumptions, accommodating asymmetric error patterns; (3) enhanced interpretability of uncertainty sources, as uncertainty is directly linked to sample dissimilarity from the training data; and (4) computational efficiency at inference (2.6–7.7× faster), requiring only a single forward pass compared to multiple passes for ensemble-based methods. Challenges remain for traits that are affected by spectral saturation. These findings highlight the advantages of distance-aware uncertainty quantification methods and underscore the necessity of diverse training datasets to minimize sampling biases and enhance model robustness. The proposed framework improves the reliability of uncertainty estimation in vegetation monitoring and offers a promising approach for broader applications.
Why it matches plant phenotyping methods植物形質をハイパースペクトル画像から推定する深層学習について、OOD条件での不確実性推定手法Dis_UNを開発・評価しており、表現型取得・推定手法が中心である。
abstractwe propose a distance-based uncertainty estimation method (Dis_UN) that quantifies prediction uncertainty
Reproduction assets foundThe paper's authors publicly released their uncertainty-analysis code (two GitHub repositories) and the study data (Hugging Face dataset) with explicit availability statements and URLs. The EnMAP and NEON hyperspectral scenes are third-party public data sources, not paper-specific deposits, and the supplement is not anCode · publicThe code for this study is available at: https://github.com/echerif18/Multi_trait_Uncertainty/ (last access: 8 March 2026).Open asset ↗echerif18/Multi_trait_Uncertaintylines:449-456Dataset · publicThe data used in this study are available on Hugging Face: https://doi.org/10.57967/hf/7838 (Cherif et al., 2026).Open asset ↗Hugging Face · 10.57967/hf/7838lines:457-483Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
This dataset consists of a collection of high-resolution RGB images of grapevine leaves, designed to support research in plant pathology, precision viticulture, and computer vision. The images were collected in situ from experimental and commercial vineyards in the north of Portugal, covering different vineyard conditions and management practices. The dataset includes healthy leaves from three grapevine Portuguese cultivars Loureiro, Viosinho and Malvasia Fina, photographed under natural lighting conditions without artificial adjustments. It is organized into five categories: healthy leaves and leaves showing symptoms of downy mildew ( Plasmopara viticola ), powdery mildew ( Erysiphe necator ), Esca complex and Erineum Mite ( Colomerus vitis ). Images are provided in JPEG format with a resolution of 3000 × 3000 pixels and 1024 × 1024 pixels and arranged in folders by health status and disease type. This dataset can be used for machine learning and deep learning applications in disease detection/classification, cultivar identification, and can support other precision agriculture applications, as well as being used for agricultural robotics and educational purposes. An evaluation on three deep learning architectures demonstrated the suitability of the dataset into separating the five classes.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像化した再利用可能なデータセットで、植物の健康状態・病害状態の画像ベース推定を支えることが中心です。深層学習による5クラス分類評価も記載されています。
abstractThis dataset consists of a collection of high-resolution RGB images of grapevine leaves, designed to support research in plant pathology, precision viticulture, and computer vision.
Reproduction assets foundThe paper is a Data in Brief article describing a public Zenodo repository of RGB grapevine leaf images (healthy plus downy mildew, powdery mildew, Esca complex, erineum mite) collected for plant disease/phenotyping research, with explicit data accessibility details. No author analysis code or trained model checkpointsDataset · publicData accessibility
Repository name: Zenodo
Data identification number: https://doi.org/10.5281/zenodo.17343473
Direct URL to data: https://zenodo.org/records/17343473Open asset ↗Zenodo · 10.5281/zenodo.17343473html-lines:93-144Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Proper detection and treatment of plant leaf diseases are essential factors for achieving good crop yields and ensuring food security. Convolutional Neural Networks (CNNs) have shown significant potential for classifying diseases from leaf images. Instead, most current work focuses on image-level prediction and ignores the relationship between infected leaves. This limitation somewhat constrains their use in modelling disease spread. Also, it makes them less efficient in typical field situations where disease is transmitted from plant to plant by physical contact. Moreover, existing CNN architectures do not access inter-lobar contextual information, an essential factor for early detection and control. To tackle this, we propose AdjLeafGNN, an innovative hybrid deep learning and graph neural network model that performs multi-class leaf disease classification and probabilistic prediction of adjacent-leaf disease spread in a single pass. The method uses the enhanced CNN model (LDDNet), with Atrous Spatial Pyramid Pooling (ASPP) and a Channel-Spatial Attention Module (CSAM), to achieve a more precise representation across multiple scales. These embeddings are then used to construct a similarity graph, enabling a GNN to infer likely disease transmission paths among leaves. We evaluate the PlantVillage dataset on the proposed model, and the results show that it outperforms state-of-the-art CNN-based methods, achieving 98.88% classification accuracy and 98.71% F1 Score. Additionally, we were able to predict disease spread with a high AUC-ROC of 0.942 and an MCC of 0.884 using our framework. These results confirm that AdjLeafGNN can accurately model both local and relational patterns. The approach we propose is scalable and interpretable, facilitating real-time monitoring and control of diseases in precision agriculture.
Why it matches plant phenotyping methods葉画像から植物病害を分類し、隣接葉間の病害拡大を推定する深層学習・GNN手法を提案・評価しており、植物の病害状態の取得・推定が研究の中心である。
abstractwe propose AdjLeafGNN, an innovative hybrid deep learning and graph neural network model that performs multi-class leaf disease classification and probabilistic prediction of adjacent-leaf disease spread in a single pass.
Reproduction assets foundThe paper uses the public Kaggle PlantVillage leaf-image dataset as its phenotyping input and releases the complete AdjLeafGNN implementation (model, preprocessing, training, evaluation) on GitHub with a Zenodo-archived DOI.Dataset · publicthe dataset was obtained from the publicly available Kaggle distribution of the PlantVillage dataset: https://www.kaggle.com/datasets/mohitsingh1804/plantvillageTheOpen asset ↗html-lines:584-607Code · publicThe complete source code of the proposed AdjLeafGNN framework, including model implementation, training scripts, and evaluation pipeline, is publicly available. GitHub repository: https://github.com/surekhareddy123/AdjLeafGNN. A permanent archived version of the repository has been deposited in Zenodo and assigned the following DOI: 10.5281/zenodo.18862439.Open asset ↗https://github.com/surekhareddy123/AdjLeafGNN · 10.5281/zenodo.18862439html-lines:584-607Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Deep learning has improved automated plant disease detection by increasing recognition accuracy and robustness compared with traditional vision-based methods. Self-supervised learning (SSL) further reduces dependence on manual labels, but its transferability across heterogeneous agricultural datasets remains insufficiently characterized. Here, we evaluate a contrastive SSL pretraining and fine-tuning pipeline, termed PlantCLR, for plant disease classification under cross-dataset transfer with target-domain fine-tuning. PlantCLR combines SimCLR-style contrastive pretraining with a lightweight convolutional classifier to balance representation quality and deployment efficiency. Experiments on PlantVillage and Cassava Leaf Disease show strong performance, achieving 99.10% accuracy and 99.04% F1-score on PlantVillage, and 96.83% accuracy and 96.70% F1-score on Cassava. Feature embedding visualization using t-SNE and explanation maps using Grad-CAM indicate improved class separability and attention to disease-relevant regions. These results suggest that contrastive SSL can improve representation transfer while maintaining computational efficiency, supporting scalable plant disease diagnostics in practical agricultural settings. Code is available at GitHub .
Why it matches plant phenotyping methods植物病害を画像から分類するPlantCLR手法を開発し、異なるデータセット間で性能評価・検証しているため、植物の病害状態を推定するフェノタイピング手法が中心である。
abstractHere, we evaluate a contrastive SSL pretraining and fine-tuning pipeline, termed PlantCLR, for plant disease classification under cross-dataset transfer with target-domain fine-tuning.
Reproduction assets foundThe paper's plant disease detection experiments use two publicly available image datasets with explicit Kaggle URLs in the Data availability statement. The authors also state code is available at GitHub, but no concrete URL is provided, so no code asset is included.Dataset · publicThe Cassava Leaf Disease Classification dataset is available at https://www.kaggle.com/c/cassava-leaf-disease-classificationOpen asset ↗Kaggle · cassava-leaf-disease-classificationlines:232-268Dataset · publicthe PlantVillage dataset is available at https://www.kaggle.com/datasets/emmarex/plantdiseaseOpen asset ↗Kaggle · emmarex/plantdiseaselines:232-268Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Crop diseases significantly threaten global food security by directly affecting the crop yield and quality. The traditional diagnostic methods are labour intensive and human error prone. However, the existing deep learning solutions suffer with poor generalization due to the sharp loss landscapes. The proposed work addresses this limitation and optimizes the Convolutional Neural Network (CNN) using the Sharpness-Aware Minimization (SAM). This method minimizes both the training loss and loss landscape sharpness and enables the model to converge to a flatter-minima with improved generalization. The proposed work is evaluated on 60,000 corn leaf image samples for four classes with 15,000 balanced samples per class after augmentation. The optimized CNN model has achieved 99.66% test accuracy at 0.33% classification error rate and outperforms the conventional optimizers like Adam (98.44% accuracy) and the Stochastic Gradient Descent (SGD). The state-of-the-art analysis presents a 99% average precision rate along with 99.66% F1-score and 0.0013% mean squared error (MSE). The quantized model achieves an inference latency of 22.7 ms/image (≈44 FPS) on a Raspberry Pi 4 and reduces model overfitting and enhances feature discriminability. These results underscore the potential of SAM-based optimization in precision agriculture by driving a scalable automation of disease management. This work bridges the gap between theoretical advances in deep learning optimization and practical deployment in resource-constrained farming environments.
Why it matches plant phenotyping methodsトウモロコシ葉画像から病害を分類するCNNの最適化と性能評価が中心であり、植物の病害状態を画像から推定するフェノタイピング手法に該当する。
abstractThe proposed work addresses this limitation and optimizes the Convolutional Neural Network (CNN) using the Sharpness-Aware Minimization (SAM).
Reproduction assets foundThe paper's Data availability statement lists the public corn leaf image datasets used for its disease-classification experiments (Kaggle corn/maize leaf disease dataset, New Bangladeshi crop disease dataset, New Plant Diseases Dataset, and the MahindiNet maize leaf disease dataset on Science Data Bank). No author codeDataset · publicg.; Gireesh Kumar: Formal Analysis, Visualization, Writing – Review & Editing, Resources.
Funding
Open access funding provided by Manipal University Jaipur. Open access funding provided by Manipal University Jaipur, Jaipur. No external funding was received for this research.
Data availability
Corn or Maize Leaf Disease Dataset, https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset [ 33 ] New Bangladeshi crop disease dataset. https://www.kaggle.com/datasets/nafishamoin/new-bangladeshi-crop-disease [ 38 ] New Plant Diseases Dataset, https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset [ 46 ] Mohammed Abo-Zahhad et al. (2023). MahindiNet: Maize Leaf DOpen asset ↗Kagglelines:2568-2620Dataset · publicby Manipal University Jaipur. Open access funding provided by Manipal University Jaipur, Jaipur. No external funding was received for this research.
Data availability
Corn or Maize Leaf Disease Dataset, https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset [ 33 ] New Bangladeshi crop disease dataset. https://www.kaggle.com/datasets/nafishamoin/new-bangladeshi-crop-disease [ 38 ] New Plant Diseases Dataset, https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset [ 46 ] Mohammed Abo-Zahhad et al. (2023). MahindiNet: Maize Leaf Disease Dataset[DS/OL]. V1. Science Data Bank. https://cstr.cn/31253.11.sciencedb.12556 . CSTR:31,253.11.sciencedb.12556 [ 55 ] Open asset ↗Kagglelines:2568-2620Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published31 Mar 2026Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable ApplicationsCited by 0 · OpenAlex ↗
The aspect of crop monitoring takes into consideration the timely detection of crop stages, leaf disorders, and deficiencies to enhance crop yield and decrease losses in agriculture. However, most of the current methods are limited to either disease detection or nutrient evaluation and do not examine the conditions of crops at various stages of growth, even though several AI -IoT-based solutions have been suggested to be applied to crop health monitoring. In addition, the estimation of the severity of the diseases is neglected, and this restricts decision-making in favor of the farmers. To address these constraints, the paper presents a Parametrized Elliptical Cauchy Fuzzy Inference System (PECFIS) combined with a Deep Glorot Bessel Elliott Softplus Convolutional Neural Network (DGBESCNN), proposed as an AI-based solution for crop monitoring and IoT support. The IoT devices in the form of drones are used to get real-time field images, and they are preprocessed in terms of noise reduction, contrast enhancement by LHM-CLAHE, conversion to HSV color space, and feature discrimination by vegetation indexing, as well as C3MEK-Means. PECFIS is used to determine eight key stages of rice growth and the severity of leaf diseases, whereas DGBESCNN provides proper classification of leaf diseases and nutrient deficiencies at each growth stage. The evaluation of the proposed framework was conducted using publicly available datasets on rice leaf disease and nutrient deficiency. The results of the experiments show that the system achieves high classification performance, with an accuracy of 98.82, a precision of 98.65, a recall of 98.73, an F1-score of 98.59, and low error rates (MSE = 0.0135, RMSE = 0.116). The findings show that the developed AI-IoT system is superior to available approaches and can serve as a dependable, real-time, and scalable solution in precision agriculture and intelligent crop monitoring.
Why it matches plant phenotyping methodsドローン画像からイネの生育段階と葉病害の重症度を推定・分類するAI-IoT手法が研究の中心であり、植物状態の取得・抽出方法を技術的に評価している。
abstractThe IoT devices in the form of drones are used to get real-time field images
Reproduction assets foundThe paper evaluates its PECFIS-DGBESCNN crop monitoring framework on two publicly available Kaggle datasets (Nutrient Deficiency Symptoms in Rice, 1,156 images; Rice Leaf Diseases, 120 images), with explicit dataset links provided by the authors. No author code, models, or other paper-specific assets are shared.Dataset · publicn of the low-cost ground-based IoT and weather sensors and enhanced robustness in the
current unfavorable environmental conditions.
Future Enhancement
In the future, enhanced techniques will be developed to classify the numerous types of nutrient
deficiencies in rice crops for improved productivity in agriculture.
Dataset link: https://www.kaggle.com/datasets/guy007/nutrientdeficiencysymptomsinrice
https://www.kaggle.com/datasets/vbookshelf/rice-leaf-diseases/data
References
[1] Aggarwal, M., Khullar, V., Goyal, N., Alammari, A., Albahar, M. A., & Singh, A. (2023).
Lightweight federated learning for rice leaf disease classification using non independent and
identically distributed images. SuOpen asset ↗Kaggle · guy007/nutrientdeficiencysymptomsinricepdf-raw-page:21 lines:1-50Dataset · publicstness in the
current unfavorable environmental conditions.
Future Enhancement
In the future, enhanced techniques will be developed to classify the numerous types of nutrient
deficiencies in rice crops for improved productivity in agriculture.
Dataset link: https://www.kaggle.com/datasets/guy007/nutrientdeficiencysymptomsinrice
https://www.kaggle.com/datasets/vbookshelf/rice-leaf-diseases/data
References
[1] Aggarwal, M., Khullar, V., Goyal, N., Alammari, A., Albahar, M. A., & Singh, A. (2023).
Lightweight federated learning for rice leaf disease classification using non independent and
identically distributed images. Sustainability, 15(16), 12149.
https://doi.org/10.3390/su151612149
[2] AlfOpen asset ↗Kaggle · vbookshelf/rice-leaf-diseasespdf-raw-page:21 lines:1-50Code / dataset availability confirmedEurope PMC · bioRxiv · Crossref · checked 5 Sept 2026
Premise: Herbarium specimens are increasingly used to extract morphological traits for ecological and evolutionary studies, yet the effects of tissue desiccation on trait measurements remain poorly understood. Here, we tested whether higher tissue water content leads to greater measurement changes after herborization (H1) and whether fresh trait values can be reliably predicted from herbarium measurements (H2). Methods: We evaluated the reliability of herbarium-based measurements by comparing fresh and dried traits of leaves, flowers, fleshy fruits, and seeds across 262 individuals representing 133 Neotropical Myrtaceae species. Phylogenetic least square models and machine-learning regressions were used to test H1 and H2. Results: Leaves and flowers generally shrank after herborization, fruits size metrics tended to increase, and seeds were largely unaffected. Water content was significantly associated with the magnitude of herborization effects in flowers and some leaf and seed traits. Fresh trait values were accurately predicted from herbarium measurements. Prediction errors were lowest for leaf traits, followed by fruits, flowers, and seeds. Discussion: These results partially support H1 and support H2, indicating that herbarium specimens can be reliably used for trait analyses when organ-specific responses are considered, providing a practical framework to account for potential desiccation bias in functional trait research.
Why it matches plant phenotyping methodsハーバリウム標本による植物形態形質測定の信頼性評価と、生鮮形質の予測手法が研究の中心であり、植物フェノタイピング手法の検証に該当する。
abstractWe evaluated the reliability of herbarium-based measurements by comparing fresh and dried traits of leaves, flowers, fleshy fruits, and seeds across 262 individuals representing 133 Neotropical Myrtaceae species.
Reproduction assets foundThe authors explicitly state that the code used for the PGLS and machine-learning analyses is publicly available in a GitHub repository; raw phenotype data is promised only upon acceptance, so the code asset qualifies while the dataset is not yet actionable.Code · publicSupporting Information and the code used to perform the analyses are available at
https://github.com/ykilsztajn/fresh_dry_myrtaceae. All raw data will be made available in the
same repository upon acceptance for publication.Open asset ↗ykilsztajn/fresh_dry_myrtaceaepdf-page:9 lines:1-48Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Abstract. Accurate and timely identi cation of plant diseases is essential for improving crop productivity and ensuring sustainable agricultural practices. This paper presents a comprehensive image dataset of fruit and leaf diseases covering six economically important horticultural crops: Apple, Banana, Citrus, Guava, Mango, and Papaya. The dataset comprises high-quality RGB images representing both healthy and diseased samples, with disease symptoms including spots, lesions, discoloration, blight, rot, and fungal and bacterial infections captured under diverse real-world conditions. Variations in illumination, background complexity, viewing angles, growth stages, and symptom severity are intentionally included to enhance the robustness and generalizability of learning models developed using this data. The dataset is structured in a class-wise manner and preprocessed to support direct integration with deep learning frameworks. It is extensively used to train, validate, and evaluate deep learning based plant disease classi cation models, enabling automatic feature learning from raw images without manual intervention. Experimental usage demonstrates that the dataset is well suited for convolutional neural networks and attentionbased architectures, facilitating e ective discrimination between multiple disease categories across di erent crops and plant organs. By providing a uni ed multi-crop, multi-disease benchmark, this dataset aims to accelerate research in automated crop disease diagnosis, precision agriculture, and intelligent decision-support systems for sustainable farming.
Why it matches plant phenotyping methods植物の葉・果実の病徴画像を収録したデータセット／ベンチマークであり、病害状態の画像ベース推定を中心的に扱うため。
abstractThis paper presents a comprehensive image dataset of fruit and leaf diseases covering six economically important horticultural crops
Reproduction assets foundThe paper's core asset is the ABCGMP fruit and leaf disease image dataset, publicly deposited on Mendeley Data, with author analysis code also stated to be available on GitHub. Both are paper-specific, public, and actionable.Dataset · publicData is available on Mendeley:1Open asset ↗pdf-page:33 lines:1-56Code · publicCode availability: Code is available on GitHub 2Open asset ↗GitHubpdf-page:33 lines:1-56Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Sugarcane diseases significantly reduce crop yield and quality, posing persistent challenges to the agricultural sector. This study presents a novel ensemble framework that integrates Vision Transformer and Swin Transformer architectures for accurate sugarcane leaf disease detection. By combining global self-attention with localized window-based attention mechanisms, the proposed model effectively captures multi-scale visual features associated with diverse disease symptoms. Experimental evaluation on a large, labeled sugarcane leaf dataset achieved a validation accuracy of 98.16% and a test accuracy of 97.06%, outperforming several convolutional neural network baselines. Additionally, a large language model (LLM) interface is employed as a post-prediction decision-support module, generating disease-specific descriptions and management suggestions based solely on the predicted disease class. This integrated framework indicates the potential effectiveness of transformer-based ensemble models combined with intelligent advisory support for practical decision-making in precision agriculture.
Why it matches plant phenotyping methodsサトウキビ葉の病徴を画像から分類するTransformerベースの手法開発と性能評価が中心であり、植物病害状態のフェノタイピングに該当する。
abstractThis study presents a novel ensemble framework that integrates Vision Transformer and Swin Transformer architectures for accurate sugarcane leaf disease detection.
Reproduction assets foundThe paper's sugarcane leaf disease image dataset (19,926 images, six classes) is explicitly stated to be publicly available on Kaggle; no author code or model checkpoints are shared.Dataset · publicThe Sugarcane Plant Diseases Dataset used in this study is publicly available on Kaggle at: https://www.kaggle.com/datasets/akilesh253/sugarcane-plant-diseases-dataset . The dataset is released for academic research and benchmarking purposes.Open asset ↗Kagglelines:112-131Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Field / plotMultimodalLeafSegmentationDisease symptoms / severity
Camellia oleifera is an economically vital woody oil crop. Its productivity and oil quality are severely compromised by various diseases. Implementing pixel-level lesion segmentation within complex field environments is crucial for advancing precision plant protection. Despite recent progress, existing segmentation methods struggle with three primary challenges: semantic ambiguity arising from evolving pathological stages, blurred boundaries due to overlapping lesions, and the high omission rate of micro-lesions. To address these issues, this paper presents TB-DLossNet (Text-Conditioned Boundary-Aware Network with Dynamic Loss Reweighting), a novel segmentation framework based on semantic-visual multi-modal fusion. Leveraging VMamba as the visual backbone, the proposed model innovatively integrates BERT-encoded structured text as an auxiliary modality to resolve visual ambiguities through cross-modal semantic guidance. Furthermore, a boundary enhancement branch is incorporated alongside a multi-scale deep supervision strategy to mitigate boundary displacement and ensure the topological continuity of lesion structures. To tackle the detection of small-scale targets, we designed a dynamic weight loss function conditioned on lesion area, significantly bolstering the model's sensitivity to minute pathological features. Additionally, to alleviate the scarcity of high-quality data, we curated a comprehensive multi-modal dataset encompassing seven typical diseases of Camellia oleifera . Experimental results demonstrate that TB-DLossNet achieves a Mean Intersection over Union (mIoU) of 87.02%, outperforming the state-of-the-art unimodal VMamba and multimodal Lvit by 4.9% and 2.59%, respectively. Qualitative evaluations confirm that our model exhibits lower false-negative rates and superior boundary-fitting precision in heterogeneous field scenarios. Finally, generalization tests on an apple disease dataset further validate the robustness and transferability of the proposed framework.
Why it matches plant phenotyping methods植物病害の病斑を画素レベルで抽出する新規セグメンテーション手法を開発し、データセット整備と性能比較・汎化検証も行っているため、病害状態の画像ベース表現型計測が中心である。
abstractImplementing pixel-level lesion segmentation within complex field environments is crucial for advancing precision plant protection.
Reproduction assets foundThe authors state their code and experimental dataset (the multimodal Camellia oleifera disease segmentation dataset) are publicly available on GitHub, matching an allowed URL.Code · publicOur code and experimental dataset are available at https://github.com/zzzsq239/TB-1.Open asset ↗zzzsq239/TB-1html-lines:820-841Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Accurate plant health monitoring relies on hyperspectral imagery to extract vegetation spectral signatures and compute vegetation indices (VIs), which are critical for phenotyping and crop condition assessment. However, the requirement for high spectral resolution significantly increases the cost and complexity of data acquisition. In this study, we proposed a novel machine learning-based framework for predicting VIs from down-sampled hyperspectral reflectance data. The aim was to reduce the dependency on high-resolution spectral imagery without compromising prediction accuracy. The framework integrated correlation-based feature selection with four regression models to identify and utilize the most informative spectral bands from coarsely sampled data. The system was trained and validated using a data set consisting of 555 spectral signatures collected from olive leaves at five stages of dehydration, with spectral resolutions ranging from 1 to 100 nm. A total of 25 vegetation indices, commonly used in the estimation of water stress, chlorophyll, and nitrogen, were predicted on various sampling scales. Experimental results show that even with 100 nm spectral resolution, the proposed framework achieves high prediction accuracy, with coefficients of determination reaching 0.99 for RVSI, VOPT, and SPADI indices. These findings demonstrate that accurate vegetation index estimation is achievable with significantly fewer spectral bands, offering a cost-effective solution for large-scale plant health monitoring. This framework lays the groundwork for the development of low-cost, data-efficient remote sensing systems for precision agriculture, especially in crops such as olives, where health dynamics are sensitive to water and nutrient status.
Why it matches plant phenotyping methodsオリーブ葉のハイパースペクトルデータから植物状態に関わる植生指数を推定する、低コストな機械学習・スペクトル測定フレームワークの開発と検証が中心である。
abstractwe proposed a novel machine learning-based framework for predicting VIs from down-sampled hyperspectral reflectance data
Reproduction assets foundThe paper's Data Availability statement points to a Figshare deposit (DOI 10.6084/m9.figshare.26950660.v2), which per the statement hosts the study's data — the 555 olive-leaf hyperspectral signatures and vegetation index measurements underlying the phenotyping analysis. This is a paper-specific, publicly accessible,直接Dataset · publicnm.
(PDF)
S2 File
Inclusivity in global research questionnaire.
(PDF)
Acknowledgments
The authors thank the Advanced Center of Electric and Electronic Engineering - AC3E ANID. The authors acknowledge the support provided by Universidad Técnica Federico Santa María and the Direction of Post-Grade programs DDP.
Data Availability
https://doi.org/10.6084/m9.figshare.26950660.v2 .
Funding Statement
This work was funded by the ANID FB240002 basal center AC3E, and ANID national doctorate scholarship, folio N°21231129. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References
1. Ruiz-Carrasco B, Fernández-Lobato L, López-Open asset ↗figshare · 10.6084/m9.figshare.26950660.v2lines:266-293Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Abstract The precise classification of plant diseases is crucial for ensuring food security for all people and boosting agricultural productivity. Although there has been significant progress in this field using deep learning approaches, cross-dataset training hasn’t drawn as much attention from researchers as intra-dataset training has. Moreover, very few models have successfully blended intra-dataset and cross-dataset training approaches. This paper proposes a novel attention-based Convolutional Neural Network (CNN) to overcome these limitations. The model improves feature extraction and classification accuracy across multiple datasets by using attention mechanisms. It was tested on five datasets (Digipathos, Northern Leaf Blight (NLB), PlantVillage, PlantDoc, and the CD&S dataset) that covered leaf diseases of both corn and potatoes. During intra-dataset training, the model achieved the highest classification accuracy of 99.38% when trained on images of potato leaves from the PlantVillage dataset. During cross-dataset training, the model exhibited the highest average classification accuracy of 82.93% for corn leaf diseases when trained on images from the CD&S dataset with their backgrounds removed. When compared to the techniques taken into consideration in this study under comparable experimental conditions, the results demonstrate improved performance. This study shows how the model may be flexible for both intra- and cross-datasets, offering a flexible way to categorize diseases that affect plants. Because of its ability to generalize across different datasets, it may be helpful in real-world agricultural applications with a wide variety of image quality and situations. This encourages the advancement of precision farming techniques and disease control.
Why it matches plant phenotyping methods植物葉の画像から病害状態を分類するCNN手法の開発・データセット間検証が中心であり、植物病害の表現型推定に該当する。
abstractThis paper proposes a novel attention-based Convolutional Neural Network (CNN) to overcome these limitations.
Reproduction assets foundThe paper's plant disease classification experiments rely on five publicly available leaf-image datasets, each cited with an explicit public access URL in the reference list: PlantVillage (GitHub), PlantDoc (GitHub), Digipathos (Embrapa), NLB (SciDB), and CD&S (OSF). No author analysis code or trained model checkpoint,Dataset · publicHughes, D., & Salathé, M. (2015). An open access repository of images on plant health to enable the development of mobile disease diagnostics. arXiv preprint arXiv:1511.08060. Dataset accessed via GitHub: https://github.com/spMohanty/PlantVillage-DatasetOpen asset ↗GitHub · spMohanty/PlantVillage-Datasethtml-lines:1013-1082Dataset · publicDataset available at: https://github.com/pratikkayal/PlantDoc-DatasetOpen asset ↗GitHub · pratikkayal/PlantDoc-Datasethtml-lines:979-1012Dataset · publicCD&S dataset: Handheld imagery dataset acquired under field conditions for corn disease identification and severity estimation. arXiv preprint arXiv:2110.12084. Dataset available at: https://osf.io/s6ru5/files/osfstorageOpen asset ↗OSF · s6ru5html-lines:1013-1082Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
The growing global demand for food production, coupled with the increasing threat of plant diseases, necessitates advanced and automated solutions for crop health monitoring. Among various crops, pome fruits such as apples and pears are widely cultivated yet highly susceptible to multiple diseases that can significantly reduce yield and quality. Existing approaches for disease detection and severity classification are often limited by their dependency on manual inspection and their inability to handle complex real-world imagery, especially when multiple diseases coexist on a single leaf. To address these limitations, this research introduces a novel dual-model deep learning framework for multi-disease severity detection and classification in pome fruit leaves. A fine-tuned MobileNetV2 backbone is employed to extract high-level discriminative features from a specialized pome leaf dataset annotated with multiple disease types and severity levels. The proposed system integrates a lightweight Lite-U-Net for semantic segmentation to isolate diseased regions and an enhanced Lite-YOLACT for instance segmentation using a linear combination of prototype masks and mask coefficients. Moreover, a new multi-disease severity scale is proposed to quantify the impact of multiple coexisting infections on a single leaf, an aspect not addressed in previous studies. To enhance interpretability, an improved Grad-CAM technique generates visual heatmaps highlighting the most influential regions in the model's decision-making process, providing transparency and validation for agricultural experts. Experimental evaluations demonstrate that the proposed framework achieves 95% accuracy in disease severity estimation, effectively identifying and grading multiple infections simultaneously. This study represents a significant step forward in precision agriculture, offering an efficient, interpretable, and scalable deep learning solution for real-world crop health monitoring and management. The source code and trained models are publicly available at: https://github.com/mqasim0787/Multi-Disease-Severity .
Why it matches plant phenotyping methods果樹葉の病斑領域を画像から分割し、複数病害の重症度を定量推定する深層学習フレームワークが研究の中心であり、植物状態の画像ベース表現型計測に該当する。
abstractthis research introduces a novel dual-model deep learning framework for multi-disease severity detection and classification in pome fruit leaves.
Reproduction assets foundThe paper's authors publicly release source code and trained models on GitHub, and the study analyzes two public Kaggle plant-image datasets (DiaMOS Plant and PlantVillage) used directly for the multi-disease severity phenotyping experiments.Code · publicThe source code and trained models are publicly available at: https://github.com/mqasim0787/Multi-Disease-Severity .Open asset ↗https://github.com/mqasim0787/Multi-Disease-Severity · mqasim0787/Multi-Disease-Severitylines:1-23Dataset · publicThe datasets analyzed during the current study are available publicly in the Kaggle repository, DiaMOS dataset (1) and PlantVillage Dataset (2) 0.1. [https://www.kaggle.com/datasets/alexandraneagu101/diamos-plant-dataset]Open asset ↗https://www.kaggle.com/datasets/alexandraneagu101/diamos-plant-dataset · diamos-plant-datasetlines:964-977Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Precise detection of crop leaf damage is essential for real-time plant health monitoring and yield estimation. However, conventional deep learning models often exhibit poor generalization when deployed across varying species and complex, unstructured field environments. To address these limitations, we propose a new modeling paradigm that shifts from traditional task-specific training to foundation model adaptation. Specifically, we introduce a novel architecture integrating the DinoV3 foundation model with a Unet framework to achieve robust leaf lesion segmentation. By incorporating a Spatial Prior Module (SPM) and a Projection Module, our approach effectively bridges the gap between general-purpose pre-training and domain-specific requirements. Experimental results on coffee and black gram datasets demonstrate that this paradigm consistently outperforms standard networks, including Unet, Unet++, and SwinUnet. On the coffee leaf dataset, the proposed model achieves an Intersection over Union (IoU) of 78.31% and a Pixel Accuracy of 88.00%, surpassing the baseline Unet by over 10.5% in IoU. Remarkably, the architecture reduces inference time by approximately 93.6% (from 63.41s to 4.07s), proving that high-parameter foundation models can be adapted for extreme computational efficiency in agricultural scenarios. To further validate scalability, we conduct additional experiments on a larger dataset, AMG HS . The proposed paradigm achieves the best overall detection performance while maintaining superior computational efficiency, confirming its robustness under increased data scale. Interpretability analysis reveals that the foundation model backbone effectively captures high-level semantic features of lesions, providing a clear explanation for its superior performance and cross-domain reliability. This research establishes a scalable, high-performance paradigm for intelligent crop protection, demonstrating that coupling customized encoders with foundation models is a superior strategy for cross-domain agricultural tasks.
Why it matches plant phenotyping methods植物葉の病変を画像からセグメンテーションし、葉の損傷状態を定量化する手法の開発・検証が研究の中心であるため。
abstractwe introduce a novel architecture integrating the DinoV3 foundation model with a Unet framework to achieve robust leaf lesion segmentation.
Reproduction assets foundThe paper analyzes two publicly available plant image datasets hosted on Mendeley Data: a coffee leaf rust/leaf miner dataset and a black gram leaf disease dataset, both explicitly linked in the Data Availability Statement. No author analysis code or trained model checkpoints are disclosed.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://data.mendeley.com/datasets/vfxf4trtcg/5; https://data.mendeley.com/datasets/45djgf3p96/1.Open asset ↗html-lines:473-493Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://data.mendeley.com/datasets/vfxf4trtcg/5; https://data.mendeley.com/datasets/45djgf3p96/1.Open asset ↗html-lines:473-493Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Abstract Tomato leaf diseases significantly impact agricultural productivity, necessitating accurate and efficient diagnostic methods. Deep learning has emerged as a robust approach for plant disease detection, but challenges such as inefficient feature extraction, classification, model complexity and limited computational resources hinder its widespread adoption. This study introduces a PhytoNet (Mish-Optimized SqueezeNet) Framework to enhance tomato leaf disease prediction. The SqueezeNet architecture, known for its lightweight design, is optimized with the Mish activation function to improve feature extraction and classification capabilities while maintaining computational efficiency. The methodology involves training the SqueezeNet model on 10 classes of mendaley dataset of tomato leaf images, encompassing multiple disease classes and healthy samples. Data preprocessing techniques, including image augmentation and normalization, are employed to ensure model robustness. The integration of the Mish activation function in critical layers enhances non-linearity, aiding in better gradient flow and improved performance during training. Model evaluation is conducted using metrics such as accuracy, precision, recall and F1-score. Experimental results demonstrate that the PhytoNet outperforms traditional SqueezeNet and other lightweight architectures in terms of classification accuracy, achieving over 0.9957 accuracy on the dataset. Additionally, the model maintains low computational overhead, making it suitable for deployment on resource-constrained devices. Hence, the proposed framework effectively balances accuracy and efficiency, addressing critical limitations in existing plant disease detection models. This work underscores the potential of lightweight and activation-optimized deep learning frameworks for real-time agricultural applications, paving the way for scalable and sustainable solutions in precision farming.
Why it matches plant phenotyping methodsトマト葉画像から病害状態を推定する深層学習モデルを開発・評価しており、植物病害フェノタイピング手法が中心的な貢献である。
abstractThis study introduces a PhytoNet (Mish-Optimized SqueezeNet) Framework to enhance tomato leaf disease prediction.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe dataset used in this study,Open asset ↗pdf-page:38 lines:1-41Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Early detection of maize leaf diseases is essential to prevent yield losses. Existing vision-based models face challenges in real-world environments due to data imbalance, lighting variations, and interpretability. This study presents MaizeFormerX, a lightweight Vision Transformer designed for cross-domain, explainable maize disease detection on resource-limited settings. MaizeFormerX employs multi-scale patch embeddings and a Cross-Scale Attention Fusion (CSAF) module to capture both detailed lesion textures and larger disease patterns. The CSAF output is processed through a transformer encoder stack using multi-head self-attention to model long-range dependencies. Robust preprocessing and dataset-specific augmentations were applied to improve feature extraction and address class imbalances in the Dataverse, Tanzania, and Plagues Maiz datasets. For interpretability, Grad-CAM was used for pixel-level saliency mapping in an efficient web application. When benchmarked against MobileViT, EfficientFormer, TinyViT, and Swin Transformer, MaizeFormerX achieved 97.8% accuracy on Dataverse, 97.5% on Tanzania, and 96.9% on Plagues Maiz, outperforming Swin Transformer V2 by 2–3%. Cross-domain testing yielded 88.9% accuracy when trained on Dataverse and tested on Tanzania, surpassing baseline performance by 3–6%. Class-wise analysis revealed F1 scores over 98% for Healthy and MLB classes with 6× augmentation, and over 97% for MSV. Ablation studies highlighted the significance of the cross-scale attention module for high MCC during domain shifts. This study introduces a precise, explainable, and efficient image-based method for classifying maize diseases, which could aid in more targeted crop management, reduce unnecessary agrochemical use, and promote sustainable maize production in future decision-support environments.
Why it matches plant phenotyping methodsトウモロコシ葉の病徴を画像から分類する手法の開発・ベンチマーク・交差ドメイン検証が中心であり、植物病害状態の画像ベース表現型計測に該当する。
abstractThis study presents MaizeFormerX, a lightweight Vision Transformer designed for cross-domain, explainable maize disease detection on resource-limited settings.
Reproduction assets foundThe paper's Data Availability statement explicitly lists three public maize leaf image datasets used for its phenotyping/disease-classification experiments (Dataverse, Tanzania/Mendeley, Plagues Maiz/figshare) and an authors' GitHub repository containing all code, preprocessing pipelines, and experimental configs. All四Dataset · publicThe datasets used in this study are publicly available and sourced from Dataverse (https://doi.org/10.7910/DVN/LPGHKK)Open asset ↗Dataverse · 10.7910/DVN/LPGHKKhtml-lines:2304-2339Code · publicAll code, preprocessing pipelines, and experimental configurations used in this work are available at: https://github.com/rezaul-h/MaizeFormerX/.Open asset ↗github · rezaul-h/MaizeFormerXhtml-lines:2304-2339Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Many Artificial Intelligence and Machine Learning technologies have been applied to detect rice diseases. These approaches are either unable to identify the diseases or have a slow recognition speed. Therefore, an improved Faster-RCNN (Faster-RCNN-Pro) model is proposed to overcome these issues. First, SENet attention modules are embedded in the backbone of Faster-RCNN to enhance confidence of objects that are difficult to recognize by enhancing key image information and suppressing background information. Second, structure of the feature extraction network and RPN are improved by using multi-feature scale fusion to increase the utilization of micro-target features. Third, the quantization error introduced in the process of pooling the region of interest is then eliminated by ROI Align. Finally, a balanced L1 loss function is designed to effectively reduce the imbalance between samples with a large gradient that are difficult to learn, and samples with a small gradient that are easy to learn. The experiment results show that the improved model has a better detection accuracy and robustness in recognizing the fine features of rice leaf diseases. Therefore, the application of this model to the intelligent identification of rice leaf disease can significantly improve the accuracy and reduce the misjudgment rate.
Why it matches plant phenotyping methodsイネ葉の病害状態を画像から識別する改良Faster-RCNNを開発・評価しており、植物病害表現型の取得手法が研究の中心である。
abstractan improved Faster-RCNN (Faster-RCNN-Pro) model is proposed to overcome these issues.
Reproduction assets foundThe paper's rice leaf disease detection experiments were performed on a public Kaggle image dataset (rice blast, brown spot, hispa, healthy leaves), which the authors explicitly state is publicly available with a URL matching the allowed list. No author analysis code or trained model checkpoints are disclosed.Dataset · publical Internet of Things, aiming to recognize large-scale rice leaf diseases. Moreover, it is beneficial for the modernization of the agricultural industry.
Acknowledgments
The authors would like to thank the anonymous reviewers.
Data Availability
All relevant data for this study are publicly available from the Kaggle repository ( https://www.kaggle.com/minhhuy2810/rice-diseases-image-dataset ).
Funding Statement
This work is supported by the National Natural Science Foundation of China (62402308).
References
1. Mondal S, Ghosh S, Mukherjee A. Application of biochar and vermicompost against the rice root-knot nematode (Meloidogyne graminicola): an eco-friendly approach in nematode management. JOpen asset ↗Kaggle · minhhuy2810/rice-diseases-image-datasetlines:220-237Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Oranges, mandarins, bitter oranges, and lemons are examples of citrus fruits that make delicious meals and are highly nutritious. Citrus fruits suffer from a variety of infections that affect their yield. The Department of Agriculture wants to increase the production of oranges and lemons. On the other hand, several plant diseases and their advanced stages have impacted production. The quality of fruit influences market value and its financial effect. Therefore, accurate detection of ailments and their severity is crucial for improving the output and market value of oranges and lemons. To automatically evaluate and predict diseases in citrus leaves and fruits, this paper has proposed a modified convolutional neural network (ICNN) model. Python is used to create the ICNN model, and testing is performed using benchmark datasets from various repositories. The research presented here shows that ICNN performs better than traditional deep learning and machine learning models, such as the Convolutional Neural Network (CNN) and K-Nearest Neighbours (KNN). This illustrates how machine learning models require supplementary approaches to extract parameters from data that arrives in non-automated ways. Additionally, to improve the accuracy of their classification or prediction, deep learning models require pre-trained models. As a result, ICNN, an enhanced deep learning model that can automatically predict disease with higher accuracy than other models, represents an advancement over standard CNNs. Compared with KNN and CNN, ICNN achieves 99.69% accuracy.
Why it matches plant phenotyping methods柑橘の葉・果実の病害と重症度を画像から自動推定するCNN手法を開発し、ベンチマークデータセットで比較評価しており、植物フェノタイピング手法が中心である。
abstractTo automatically evaluate and predict diseases in citrus leaves and fruits, this paper has proposed a modified convolutional neural network (ICNN) model.
Reproduction assets foundThe paper's Data Availability statement explicitly lists three public Kaggle URLs as the datasets used and analysed in the study (citrus/plant leaf disease image datasets). These are paper-specific, publicly accessible image assets directly supporting the phenotyping/disease-classification analysis. The Mendeley URL (3Dataset · publicg agricultural specialists to properly understand and accept the model’s predictions.
Author contributions
Arunapriya.R – Problem Statements, Implementation and Testing Dr.S.P.Valli – Results, Conclusion, and Summary.
Data availability
The datasets used and/or analysed during the current study available and mentioned in below [ https://www.kaggle.com/code/ghazanfarali96/leaf-disease-classification-using-cnn-lstm-rnn ]. (https:/ www.kaggle.com/code/ghazanfarali96/leaf-disease-classification-using-cnn-lstm-rnn ). [ https://www.kaggle.com/code/moazeldsokyx/plant-leaf-diseases-detection-using-cnn ]. (https:/ www.kaggle.com/code/moazeldsokyx/plant-leaf-diseases-detection-using-cnn ). [ https://wwOpen asset ↗kagglelines:372-388Dataset · publicts, Conclusion, and Summary.
Data availability
The datasets used and/or analysed during the current study available and mentioned in below [ https://www.kaggle.com/code/ghazanfarali96/leaf-disease-classification-using-cnn-lstm-rnn ]. (https:/ www.kaggle.com/code/ghazanfarali96/leaf-disease-classification-using-cnn-lstm-rnn ). [ https://www.kaggle.com/code/moazeldsokyx/plant-leaf-diseases-detection-using-cnn ]. (https:/ www.kaggle.com/code/moazeldsokyx/plant-leaf-diseases-detection-using-cnn ). [ https://www.kaggle.com/code/ritzing/plant-disease-detection-using-keras-cnn-model ]. (https:/ www.kaggle.com/code/ritzing/plant-disease-detection-using-keras-cnn-model ).
Declarations
Competing interOpen asset ↗kagglelines:372-388Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
To address the challenge of balancing model lightweight and detection accuracy in maize leaf disease detection, as well as the limitations of edge device deployment resources, we propose an enhanced target detection model, YOLOv11n-DualPC-Lite.Firstly, the C2fDualPConv module was designed, integrating PartialConv to replace some C3k2 modules in the backbone and neck networks. This approach enhances feature representation while reducing the number of parameters. Secondly, the Slim-Neck architecture is introduced in the neck network. To improve accuracy without increasing the number of parameters, the VoVGSCSPC_SimAm module enables the new Slim-Neck structure to reduce parameters while strengthening feature representation. Finally, an EfficientHead detection head is introduced that uses an inverted bottleneck MBConv module to improve performance. This significantly reduces computational load while efficiently extracting features. This study constructed a maize leaf disease dataset integrating a publicly available Kaggle dataset and a field-collected dataset from Anhui Science and Technology University's experimental plots. The dataset includes four categories: Blight, Common_Rust, Gray_Leaf_Spot, and Health. Through techniques such as rotation and gamma correction, the dataset was expanded from 3,876 to 5,165 images for model training and performance validation. Test results show this improved model performs better than other popular lightweight models overall, with a mAP50 score of 90.9%. Meanwhile, the model has only 2.13 million parameters; its computational complexity is reduced to 4.55 G, and the model size is 4.41 MB. Compared with the original YOLOv11n, its mAP50 is 1.9% higher, while the number of parameters is down by 17.8%, computational complexity is cut by 29.3%, and file size is reduced by 15.7%. When run on a Raspberry Pi 5, the model's detection speed reaches 2.3 FPS, an increase of 27.8%. This model achieves a good balance between detection accuracy and lightweight performance for maize leaf diseases, providing an efficient and practical method for real-time crop disease monitoring.
Why it matches plant phenotyping methodsトウモロコシ葉の病徴を画像から検出・分類する軽量モデルを開発し、データセット構築、性能比較、エッジデバイス検証まで行っており、植物病害状態の画像ベース表現型取得が中心である。
abstractwe propose an enhanced target detection model, YOLOv11n-DualPC-Lite
Reproduction assets foundThe paper's maize leaf disease detection study uses a public Kaggle maize leaf disease image dataset (Dataset 1) combined with a field-collected dataset. The Kaggle dataset is a public, paper-specific image asset directly used for the model's training and validation. No author analysis code, trained model checkpoints,或Dataset · publicre, the model was successfully run on a Raspberry Pi 5 edge device, realizing stable, real-time detection and providing a workable technical method for field disease monitoring.
2
Materials and methods
2.1
Dataset introduction
The dataset constructed in this study comprises two datasets: Dataset 1 from the Kaggle data website ( https://www.kaggle.com/datasets/hendriyunuswijaya/maize-leaf-disease ) and Dataset 2 collected from the experimental field at Anhui Science and Technology University in Chuzhou City, Anhui Province. Dataset 1 contains a total of 4,188 images, including 1,162 images in the Health category. All images depict only specific regions of healthy maize leaves without complex Open asset ↗Kaggle · hendriyunuswijaya/maize-leaf-diseaselines:46-63Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Introduction Crop diseases have to be diagnosed early to save crop yields and food safety. Commonly used conventional practices are time-consuming and likely to involve human error. Automatic plant disease detection is a highly efficient technology that has been brought into existence by the development of deep learning and computer vision technology and can effectively detect the signs of diseases in plant species. Sustainable agriculture and early intervention depend on accurate and interpretable detection of plant diseases. Methods This study introduces a hybrid model based on deep learning techniques that effectively identifies and categorizes leaf diseases. The proposed model, LeafFusionNet, incorporates Convolutional Neural Network (CNN) and Vision Transformer (ViT) with an efficient attention module, LeafTAM (Leaf Texture Attention Module), to effectively capture both global and local information. This architecture is enhanced by the addition of a Gabor filter layer before the CNN-ViT fusion, hence augmenting the model’s ability to extract physiologically relevant texture features. Results and discussion The model was trained and validated on the Plant Village dataset. This model attained an accuracy of 99.33%, 99% precision, recall, and F1 score, demonstrating strong generalization on new data when compared with the state-of-the-art models. This proposed hybrid model can be utilized to develop a strong agricultural diagnostics system. These results point to the possibility of designing a powerful, interpretable system by utilizing transformer-based vision modules, Gabor filters, and explainability systems such as Grad-CAM.
Why it matches plant phenotyping methods植物葉の病徴を画像から検出・分類する深層学習手法を開発し、PlantVillageデータセットで訓練・検証しており、植物病害状態の表現型取得が研究の中心です。
abstractThis study introduces a hybrid model based on deep learning techniques that effectively identifies and categorizes leaf diseases.
Reproduction assets foundThe paper's only analysis input is the public PlantVillage dataset, explicitly linked in the data availability statement; no author code or models are shared. The CABI Global Burden of Crop Loss reference is only a background statistic, not a phenotyping asset.Dataset · publicPublicly available datasets were analysed in this study. This data can be found here: https://www.kaggle.com/datasets/mohitsingh1804/plantvillage.Open asset ↗Kaggle · mohitsingh1804/plantvillagehtml-lines:554-590Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Reliable identification of maize leaf diseases is critical for mitigating crop losses, particularly in regions where farmers have limited access to experts. Although vision transformers (ViTs) have recently demonstrated strong performance in image recognition, their weak inductive bias and limited modeling of local texture patterns make them non-ideal for fine-grained maize leaf disease classification. To address these limitations, we propose ConvDeiT-Tiny, a lightweight hybrid ViT that improves DeiT-Ti by placing depthwise convolutions in parallel with multi-head self-attention modules in the first three transformer blocks. The local and global features captured by the convolution and attention modules are concatenated along the embedding dimension and fused using a multilayer perceptron. This results in richer token representations without significantly increasing model size. Across three datasets, ConvDeiT-Tiny (6.9 M parameters) consistently outperformed DeiT-Ti, DeiT-Ti-Distilled, and DeiT-S (21.7 M parameters) when trained from scratch. With transfer learning, ConvDeiT-Tiny achieved an accuracy of 99.15%, 99.35%, and 98.60% on the CD&S, primary, and Kaggle datasets, respectively, surpassing many previous studies with far fewer parameters. For explainability, we present gradient-weighted transformer attribution visualizations showing the disease lesions driving model predictions. These results indicate that injecting local inductive bias in early transformer blocks is beneficial for accurate maize leaf disease classification.
Why it matches plant phenotyping methodsトウモロコシ葉の病徴画像を対象に、病害分類のための新規Vision Transformerモデルを開発・比較しており、植物病害状態の画像ベース表現型推定が中心である。
abstractwe propose ConvDeiT-Tiny, a lightweight hybrid ViT that improves DeiT-Ti by placing depthwise convolutions in parallel with multi-head self-attention modules in the first three transformer blocks.
Reproduction assets foundThe authors publicly release their analysis code and dataset splits (including their field-collected primary maize leaf image dataset) via a GitHub repository, and the paper's classification experiments use the public Kaggle Corn or Maize Leaf Disease Dataset (COMLDD). Both are paper-specific, public, and actionable.Code · publicThe program code and dataset splits for the three datasets used in this study, including our primary data, can be found at https://github.com/DamarisWaema/ConvDeiT-Tiny (accessed on 18 March 2026).Open asset ↗DamarisWaema/ConvDeiT-Tinylines:121-289Dataset · publicGhose S.
Corn or Maize Leaf Disease Dataset
Available online: https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset (accessed on 10 July 2025)Open asset ↗lines:474-623Code / dataset availability confirmedEurope PMC · Crossref · OpenAlex · checked 5 Sept 2026
High-density planting is an effective strategy to increase maize yield but imposes greater demands on plant architectural adaptability. To elucidate the structural response mechanisms of maize under varying planting densities, we developed a high-throughput 3D phenotyping system tailored to complex field conditions. High-precision point clouds of field-sampled plants were obtained via multi-view 3D reconstruction. Using a deep learning network, stem and leaf organs were semantically segmented (95.6% accuracy), while leaves were individually separated via clustering (94.8% accuracy). From these data, 31 plant architectural traits and 14 ear-leaf traits were extracted, establishing a hierarchical trait characterization system. Results showed that increased planting density significantly influenced plant architecture reshaping and structural coordination, leading to more compact plant forms and ear height position centralization. Ear leaves exhibited heightened sensitivity to density variation, particularly in leaf area, vertical distribution, and leaf inclination angle, suggesting an early-response role. Principal component analysis and clustering further revealed patterns of structural differentiation and key traits driving these changes under density treatments. The integrated workflow-comprising data acquisition, modeling, segmentation, clustering, trait extraction, and analysis-offers a robust approach for structural phenotyping and intelligent breeding selection in maize and other tall crops. This pipeline provides valuable technical support and data resources for optimizing dense planting strategies and advancing digital agriculture.
Why it matches plant phenotyping methods高スループット3D表現型システムを開発し、点群再構成・器官分割・クラスタリングから多数の植物構造形質を抽出することが中心であるため。
abstractwe developed a high-throughput 3D phenotyping system tailored to complex field conditions
Reproduction assets foundThe paper's authors provide a public GitHub repository for the study's source code (segmentation/trait-extraction pipeline). The phenotype point-cloud dataset itself is only available on request from the corresponding author, so it is not a public asset.Code · publicThe code of this study will be made publicly available upon publication. The source code is available at https://github.com/CSC-csc426/3D-Point-Cloud-Driven-Organ-Semantic-Segmentation-to-Assess-Maize-Structural-Responses .Open asset ↗CSC-csc426/3D-Point-Cloud-Driven-Organ-Semantic-Segmentation-to-Assess-Maize-Structural-Responseslines:330-415Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Recognizing lineages is a central challenge in plant systematics, making it essential to explore multiple analytical tools. In this context, this study investigates how frond shape can assist in discriminating against lineages within the Scaly clade of Microgramma (Polypodiaceae), and tests whether the integration of multiple lines of evidence enables a more consistent recognition of lineages than exclusively macromorphological approaches. We analyzed 271 specimens representing eight species, using Elliptical Fourier Analysis (EFA) to quantify frond shape, followed by multivariate statistical tests (PCA, MANOVA, LDA). Evolutionary relationships between spectral and morphometric data were assessed through phylogenetic generalized least squares (PGLS) regressions and phylogenetic partial least squares (Phylo-PLS) analyses. Dimorphic species exhibited higher discrimination capacity (average accuracy of 80–83%). Fertile and combined fronds yielded the highest accuracy values. Morphologically similar species, such as M. reptans and M. tobagensis, showed significant overlap, whereas M. percussa achieved the best performance (average accuracy of 80%). Morphometric-spectral integration showed a strong correlation (R² = 0.72; P = 0.003), and both the combined datasets (spectra and outline) and the individual datasets of spectral and shape features revealed a high phylogenetic signal (λ = 1–0.84), indicating partial coevolution between frond shape, chemical composition, and the evolutionary history of the group. Outline morphometry combined with infrared spectroscopy within a phylogenetic framework improves lineage discrimination, although overlap zones persist, reflecting complex evolutionary processes. Our study highlights the potential of integrative systematics to elucidate species boundaries in groups with high morphological disparity, as well as the need for broad sampling and multi-evidence approaches in future systematic reviews.
Why it matches plant phenotyping methodsフロンド形状をElliptical Fourier Analysisで定量化し、赤外分光との統合を用いて系統識別性能を評価しており、植物器官形質の取得・解析手法が研究の中心です。
abstractusing Elliptical Fourier Analysis (EFA) to quantify frond shape, followed by multivariate statistical tests (PCA, MANOVA, LDA).
Reproduction assets foundThe authors state that raw data, processed data, and R analysis code for the frond outline morphometrics are publicly available on GitHub (Microgramma-Outline), and the FT-NIR spectral data repository (Microgramma-FTNIR) is referenced in the methods. Both are paper-specific, public, and actionable.Code · publicSciELO Preprints - Este documento é um preprint e sua situação atual está disponível em: https://doi.org/10.1590/SciELOPreprints.15500
573 The raw data, processed data, and R analysis code are publicly available on GitHub:
574 https://github.com/labevofern/Microgramma-Outline.git.
575
576 REFERENCES
577 Ackerly D.D. (2004) Adaptation, Niche Conservatism, and Convergence: Comparative
578 Studies of Leaf Evolution in the California Chaparral. The American Naturalist, 163, 654–
579 671.
580 Adams D.C., Collyer M.L. (2018) Multivariate Phylogenetic Comparative Methods:
581 Evaluations, Comparisons, and RecoOpen asset ↗labevofern/Microgramma-Outline · Microgramma-Outlinepdf-layout-page:25 lines:1-48Dataset · publicbiting the highest
157 perpendicular distance from the line connecting the first and last bands in the R² × ranking
158 plot (Fig. S2). Following the methods described in Mendonça et al. (2026), spectral data were
159 acquired using a PerkinElmer Frontier™ near-infrared Fourier transform spectrometer (FT-
160 NIR) available at (https://github.com/labevofern/Microgramma-FTNIR).
161 Phylogenetic comparative analyses
162 To provide a phylogenetic framework for comparative morphometric and spectral analyses,
163 we used the pruned version of the Microgramma chloroplast phylogenetic inference from
164 Mendonça et al. (2026). This tree was based on the Bayesian phylogenetic tree published by
165 AOpen asset ↗labevofern/Microgramma-FTNIR · Microgramma-FTNIRpdf-layout-page:8 lines:1-55Code / dataset availability confirmedEurope PMC · OpenAlex · checked 14 Sept 2026
Sun-induced fluorescence (SIF) has emerged as a promising tool for tracking photosynthetic dynamics, yet its application in monitoring biotic stress remains underexplored in field conditions. In this study, we investigated the effects of Cercospora leaf spot (CLS), a destructive foliar disease of sugar beet (Beta vulgaris L.), for which traditional monitoring methods often fail to capture subtle disease effects or distinguish between structural and physiological stress responses. CLS infection was induced through artificial inoculation and manually scored. Canopy-level reflectance indices were acquired along with red and far-red passive SIF signals and active PSII efficiency traits using FloX and LIFT sensors mounted on an automated high-throughput phenotyping platform. The results demonstrate that SIF effectively detects CLS in sugar beet, with responses comparable with structural and disease- specific indices. Despite visible symptoms, PSII efficiency (Fq'/Fm') remained stable across treatments, indicating limited impairment of leaf photosynthetic efficiency at early stages. However, the canopy-level electron transport rate varied significantly and showed a strong relationship with red and far-red SIF, suggesting that CLS primarily affects canopy light absorption and utilization. After structural normalization, SIF yield remained largely unchanged, confirming that observed SIF reductions were mainly driven by canopy structural alterations. Overall the study demonstrates the effectiveness of SIF for large-scale disease monitoring and integration into high-throughput phenotyping, while also revealing structural and physiological factors influencing the SIF signal under disease stress.
Why it matches plant phenotyping methodsSIFおよびPSIIセンサーを搭載したハイスループット表現型解析プラットフォームで、サトウダイコンの病害状態と構造・生理応答を評価する手法の実質的な適用・検証が中心である。
abstractCanopy-level reflectance indices were acquired along with red and far-red passive SIF signals and active PSII efficiency traits using FloX and LIFT sensors mounted on an automated high-throughput phenotyping platform.
Reproduction assets foundThe paper's phenotyping dataset (SIF, reflectance indices, LIFT PSII traits, disease scores from the CLS sugar beet field trial) is deposited in the open access Jülich DATA repository under DOI 10.26165/JUELICH-DATA/FOQOFI. No separate author analysis code repository with explicit availability language is stated; R/lmeDataset · publicThe dataset has been deposited in the open access Jülich DATA reposi ease using UAV-supported image data and deep learning. Sugar Industry
tory: https://doi.org/10.26165/JUELICH-DATA/FOQOFI. 147, 79–86.
Ispizua Yamati FR, Bömer J, Noack N, Linkugel T, Paulus S, Mahlein
A-K. 2025. Configuration of a multisensor platform for advanced plant phe
References notyping and disease detection: case study on cercospora leaf spot in sugar
Ač A, Malenovský Z, Olejníč ková J, Gallé A, Rascher U, Mohammed beet. Smart AgricultOpen asset ↗Jülich DATA · 10.26165/JUELICH-DATA/FOQOFIpdf-layout-page:14 lines:52-72Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
Leaf morphology in tea plants (Camellia sinensis L.) profoundly influences tea quality and agronomic value, yet its genetic basis remains elusive due to labor-intensive phenotyping, foliage architecture, and ecological sensitivity of traits. Moreover, traditional methods forfeit quantitative color gradients and population-level morphological complexity. To address this challenge, we developed coleaf, an open-source image recognition-based software that demonstrated 97.6% accuracy over conventional ImageJ measurements, while offering higher efficiency and color hues quantification. We then estimated 7 key morphological traits focusing on leaves from a collection of ~ 4,200 mature leaves and ~ 5,000 bud-leaf samples across 167 genetically diverse tea accessions by coleaf. While classical understanding suggests leaf shape differentiation between two varieties in genus sinensis assamica (CSA) and sinensis (CSS), our phenotypic clustering revealed incomplete congruence with phylogenetic relationships, suggesting the presence of additional genetic or environmental modulators beyond population divergence. Furthermore, we integrated phenotypic data with whole-genome resequencing for multi-model genome-wide association studies (GWAS). Candidate genes associated with leaf architecture were involved in plant development (e.g., CsFAS2), cell division and elongation (e.g., CsFIP1), and cellular morphogenesis (e.g., CsRLK), whereas those associated with leaf color, regulated pigment accumulation (e.g., ABC transporters, CsMYB113). In conclusion, this study establishes a standardized computational framework validating automated image recognition for plant leaf phenomics. The end-to-end framework from high-throughput phenotyping to gene discovery provides critical genetic targets for tea breeding, demonstrating transformative potential in accelerating the genetic improvement of tea plants.
Why it matches plant phenotyping methods茶葉形態の画像認識ソフトウェアを開発・検証し、高スループットな形質抽出フレームワークとして適用しており、植物フェノタイピング手法が研究の中心である。
abstractwe developed coleaf, an open-source image recognition-based software that demonstrated 97.6% accuracy over conventional ImageJ measurements
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicAll codes and tools used in this study are described in Methods, coleaf is available on github (https://github.com/mengmeng-jiang/coleaf).Open asset ↗mengmeng-jiang/coleafhtml-lines:390-460Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Summary Genetically encoded biosensors are one of the essential tools in biological research. They enable visualization of molecules of interest from the subcellular level to entire organism level in vivo and can be used to monitor presence of small molecules, gene expression, protein activity, and protein degradation. However, multiplexing fluorescent biosensors in plants is notoriously difficult due to signal bleed-through and strong autofluorescence from chlorophyll. In this study, we investigated the potential of multiplexing biosensors based on the selection of reporter fluorescent proteins. We characterized the emission spectra, fluorescence lifetimes, and relative brightness of diverse fluorescent proteins in plant leaves. We show that selected proteins exhibit comparable brightness, supporting their use in co-expression experiments and reliable quantification of individual signals. To separate three overlapping signals, we applied two different linear unmixing approaches and compared them to results obtained without unmixing. We identified channel separation unmixing approach as the most suitable for biosensors. Additionally, we show how unmixing with the selected approach can be applied to separate autofluorescence and five fluorescent proteins. We further validated this approach in virus-infected cells by following organelle dynamics in vivo . Finally, we demonstrate the feasibility of high-throughput segmentation and quantification with a custom MATLAB workflow for nuclei, chloroplasts, and cytoplasm signal analysis. Overall, our work demonstrates that biosensors can be multiplexed, even when their emission spectra overlap. Significance statement Multiplexing genetically encoded biosensors in plants has been limited by overlapping fluorescent signals and strong autofluorescence. This study presents an optimized framework for linear unmixing and provides a MATLAB-based organelle segmentation tool, allowing precise quantification of multiple fluorescent reporters in vivo and advancing real-time visualization of complex cellular processes in plants.
Why it matches plant phenotyping methods植物組織における蛍光シグナルの分離、検出、セグメンテーション、定量化手法を開発・比較・検証しており、植物の細胞・細胞小器官状態を取得する方法が中心である。
abstractTo separate three overlapping signals, we applied two different linear unmixing approaches and compared them to results obtained without unmixing.
Reproduction assets foundThe paper deposits raw confocal image data on Zenodo (10.5281/zenodo.19691651) and a MATLAB nuclei segmentation/quantification script on GitHub. Only the GitHub repository URL appears in the allowed URL list, so the code asset is reported; the Zenodo image deposit is noted but cannot be listed without a matching URL.Code · publici (ORCID: 0000-0002-6235-2816)
14
15 DATA AVAILABILITY
16 Raw image data supported with metadata were deposited to Zenodo:
17 10.5281/zenodo.19691651and can be opened with LAS X available at https://www.leica-
18 microsystems.com/products/microscope-software/p/leica-las-x-ls/downloads/. MATLAB script
19 was deposited to GitHub: https://github.com/NIB-SI/Nuclei-segmentation.
20 FUNDING
21 This research was funded by the Slovenian Research and Innovation Agency (research core
22 funding No. P4-0165, P4-0463, projects J4-1777, J4-60073, J4-70169 and ARIS program for
23 young researchers).
24 CONFLICT OF INTEREST
25 The authors declare no conflicts of interest. This article does not contain any Open asset ↗NIB-SI/Nuclei-segmentationpdf-layout-page:1 lines:1-34Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Abstract Disease detection and categorization in rice leaf play a crucial role in mitigating crop damage and supporting sustainable agriculture. Traditional approaches, which often rely on manual inspection, are limited by labor intensity, variability and error susceptibility. This paper introduces a Hybrid Convolutional Vision Transformer (CVT) model with Spatial Attention (SA) to enhance the detection accuracy and classification reliability in rice leaves. The proposed CVT framework integrates a Convolutional Neural Network (CNN), which is the backbone for initial feature extraction with a Vision Transformer (ViT) in advanced feature representation. Convolutional Neural Network captures the essential textures and shapes, while the Vision Transformer applies attention across image patches, effectively learning the complex spatial dependencies necessary for identifying disease-specific characteristics within diverse field environments. Further, SA module refines the model by assigning greater weight to diseased regions, reducing interference from non-leafbackground areas. Experimental results on rice leafdataset demonstrate that the hybrid CVT with SA model achieves over 98.12% feature extraction accuracy, 98.56% classification accuracy in dataset 1 and 98.26% feature extraction accuracy, 98.67% classification accuracy in dataset 2 across multiple rice leaf categories, outperforming baseline CNN and ViT models. Spatial Attention heat maps highlight the most important locations during decision-making process, making the model more interpretable. This hybrid CVT model offers a scalable solution for rice leaf disease detection and categorization, with potential applications in precise agriculture systems, including drone-based or mobile implementations for field monitoring. The presented model exhibits maximum performancethan the othertraditional methods.
Why it matches plant phenotyping methodsイネ葉の病害状態を画像から推定する深層学習モデルの開発と比較評価が論文の中心であり、植物病害フェノタイピング手法に該当する。
abstractThis paper introduces a Hybrid Convolutional Vision Transformer (CVT) model with Spatial Attention (SA) to enhance the detection accuracy and classification reliability in rice leaves.
Reproduction assets foundThe paper uses two public Kaggle rice leaf disease image datasets as its phenotyping inputs, with explicit URLs. The authors' code and generated data are only available on request, so no code/model asset qualifies.Dataset · publicconducted interviews with Department
of Agriculture—particularly those from the Regional Crop Protection Center. Images of several
rice plant diseases are collected using the means available, which included digital cameras and
smart phones. After gathering, all the imagesare pre-processed and included in the dataset.
Dataset 1: https://www.kaggle.com/datasets/nashehannafii/datasetleafblast
Dataset 2: https://www.kaggle.com/datasets/vbookshelf/rice-leaf-diseases
4.2 Data preprocessingOpen asset ↗kaggle · nashehannafii/datasetleafblastpdf-raw-page:13 lines:1-17Dataset · publicm the Regional Crop Protection Center. Images of several
rice plant diseases are collected using the means available, which included digital cameras and
smart phones. After gathering, all the imagesare pre-processed and included in the dataset.
Dataset 1: https://www.kaggle.com/datasets/nashehannafii/datasetleafblast
Dataset 2: https://www.kaggle.com/datasets/vbookshelf/rice-leaf-diseases
4.2 Data preprocessingOpen asset ↗kaggle · vbookshelf/rice-leaf-diseasespdf-raw-page:13 lines:1-17Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Background: A major challenge to agricultural productivity in the tea industry is disease. that affects the quantity and quality of tea leaves produced. The extensive development of computational methods for treating diseases has been widely used due to fast and accurate detection. Methods: The proposed method uses sequential Convolutional Neural Network (CNN) computations with many hidden layers to classify diseased and healthy tea leaves into multiple groups. By enhancing feature identification, this structure increases the criteria for accurate disease detection. The data having 5 diseased and one healthy category is obtained from the Kaggle database. After preprocessing the data, it is split into 80:20 ratios for training and testing steps. CNN is constructed using the Keras Sequential API in Jupiter notebook using Anaconda environment. Result: The total accuracy of the ML neural network training for classification was 98.52%. After 50 epochs of training, the model performed well, achieving high accuracy on training and validation datasets. The examination of the confusion matrix showed that several tea leaf diseases may be identified with high accuracy and few misclassifications. In general, the model demonstrated remarkable precision in differentiating between unhealthy and undamaged tea leaves.
Why it matches plant phenotyping methods茶葉画像から病害状態をCNNで分類する手法が研究の中心であり、植物の病徴・健全性という状態を直接推定しているため、植物フェノタイピング手法として含める。
abstractThe proposed method uses sequential Convolutional Neural Network (CNN) computations with many hidden layers to classify diseased and healthy tea leaves into multiple groups.
Reproduction assets foundThe paper's tea leaf disease image dataset is publicly available on Kaggle, with an explicit dataset link in the references. No author code or trained model is publicly deposited; other data are available only upon request.Dataset · publical tealeaf disease recognition
using a convolutional neural network model. Symmetry.
11(3): 343. https://doi.org/10.3390/sym11030343.Cho, O.H., Na, I.S. and Koh, J.G. (2024). Exploring advanced machine
learning techniques for swift legume disease detection.
Legume Research. 47(7): 1221-1227. doi: 10.18805/LRF-789.
Dataset Link: https://www.kaggle.com/datasets/shashwatwork/identifying-disease-in-tea-leafs?select=tea+sickness+
dataset. (Accessed on 06/05/2024).
Datta, S. and Gupta, N. (2023). A novel approach for the detection
of tea leaf disease using deep neural network. Procedia
Computer Science. 218: 2273-2286. https://doi.org/10.1016/j.procs.2023.01.203.Deka, N. and Goswami, K. (2020). EcOpen asset ↗Kaggle · shashwatwork/identifying-disease-in-tea-leafspdf-raw-page:8 lines:1-75Code / dataset availability confirmedEurope PMC · bioRxiv · Crossref · checked 15 Sept 2026
Quantitative studies of plant growth and environmental responses increasingly rely on time-series imaging, yet automated segmentation remains challenging due to continuous growth, large non-rigid morphological change, and frequent self-occlusion. Traditional image-processing pipelines and task-specific deep learning models often require extensive annotated datasets and retraining, limiting portability across species, developmental stages, and imaging conditions. Here we present SAP (Segment Any Plant), a plant-focused framework that leverages the pretrained Segment Anything Model 2 (SAM2) to enable few-shot, training-free segmentation of plant time-series imagery. SAP integrates interactive prompting, automated temporal mask propagation, and centerline extraction within a web-based interface, allowing users to move from raw images to quantitative descriptors of organ shape and dynamics without programming expertise. Across multiple systems, including Arabidopsis thaliana rosette development, root growth, sunflower gravitropism, and confocal root microscopy, SAP achieves high segmentation accuracy (mean IoU 0.89–0.93) and sub-pixel centerline precision from single-frame prompting. By reducing the need for task-specific retraining, SAP provides a transferable framework for reproducible time-series phenotyping across diverse experimental contexts.
Why it matches plant phenotyping methods植物の時系列画像から器官形状・動態を抽出するセグメンテーション手法とWeb基盤を開発し、複数系で精度検証しているため、植物フェノタイピング手法が中心である。
abstractHere we present SAP (Segment Any Plant), a plant-focused framework that leverages the pretrained Segment Anything Model 2 (SAM2) to enable few-shot, training-free segmentation of plant time-series imagery.
Reproduction assets foundThe paper's authors publicly release both the SAP analysis code (GitHub repository) and the datasets generated/analyzed in the study (Zenodo), including raw images, ground-truth and SAP-generated segmentation masks, centerline validation data, and supplementary videos. Both are paper-specific, public, and directly cit.Dataset · publicCode Availability. The code is available at
https://github.com/merozlab/plant-segmentation-app.Data Availability. The datasets generated and an-
alyzed during this study are available on Zenodo at
https://doi.org/10.5281/zenodo.18732705. This includes
raw images and segmentation masks for the sunflower
gravitropism and Arabidopsis root growth experiments,
SAP-generated masks for the Lee et al. (9) and Strauss
et al. (13) datasets, centerline validation data, and supple-
mentary videos.
Funding. Y.M. acknowledges support from the Israel Sci-
ence Foundation ResOpen asset ↗zenodo · 10.5281/zenodo.18732705pdf-raw-page:9 lines:1-74Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Rice production is integral to the agricultural sector of India; over 65% of the populations are dependent on rice as their major staple. The cultivation of rice sustains this important agricultural sector; yet, there are many challenges encountered by rice producers, one of which is several types of disease that negatively impact yield and quality. Due to the fact that rice leaf smut, brown spot and bacterial leaf blights are among the most important types of diseases that can significantly reduce the yield and quality of rice, it is important to be diligent when identifying these diseases using accurate and speedy methods on an annual basis for successful and sustainable production of rice crops. As technology advances there continue to be emerging technologies such as Deep Learning (DL) as applied in agriculture to identify diseases and therefore reshape the agricultural paradigm so as to address agricultural disease challenges more readily. This research proposes a previously undemonstrated approach for identifying Rice Leaf Disease using EfficientNetV2; a Diffusion Bounded Attention method for disease detection. The quality of the input imagery has been greatly increased using a Preceding Noise Reduction (PNR) using the Guided Filopic Diffusion (GFD) technique, retaining important characteristics of Rice Leaves (Leaf Texture) which are critical for disease classification within agricultural imaging. To evaluate the performance of our model we utilized the Dice Similarity Coefficient (DSC). This coefficient measures how much the predicted image areas representing disease overlap with the actual affected areas of the image. Therefore, DSC is a reliable way to evaluate model segmentation capability. The Rice Leaf Diseases Dataset we used to identify and classify Rice Leaf Diseases was very comprehensive. Our model achieved an accuracy rate of 98.92% and also attained the best recall, precision and F1 score.
Why it matches plant phenotyping methodsイネ葉の病害症状を画像から検出・分類・セグメンテーションする手法が研究の中心であり、植物の病害状態を直接推定している。
abstractThis research proposes a previously undemonstrated approach for identifying Rice Leaf Disease using EfficientNetV2; a Diffusion Bounded Attention method for disease detection.
Reproduction assets foundThe paper's sole data asset is the public Kaggle Rice Leaf Diseases Dataset used for all experiments; no author code or models are deposited.Dataset · publicl analysis and data collection. N.K has done the initial drafting and statistical analysis. P.R. did the investigation. All the authors of the article have read and approved the final article.
Funding
Open access funding provided by Vellore Institute of Technology.
Data availability
The rice leaf disease data are assessed using https://www.kaggle.com/datasets/vbookshelf/rice-leaf-diseases
Declarations
Competing interests
The authors declare no competing interests.
References
1.
Upadhyay N Gupta N
Detecting fungi-affected multi-crop disease on heterogeneous region dataset using modified ResNeXt approach
Environ. Monit. Assess. 2024 196 7 610
10.1007/s10661-024-12790-0
38862723
Upadhyay, N. & Open asset ↗Kaggle · vbookshelf/rice-leaf-diseaseslines:553-627Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Service crops are grown to provide ecosystem services in viticulture, but their adoption remains limited due to their competition with grapevine for soil resources. To identify trade-offs between services, the effect of service crops management strategies on grapevine performances still need further research. This dataset presents data from two experiments conducted to study the effect of service crops management on soil resources and grapevine performances. The inter-row vegetation was sampled in two Mediterranean vineyards using quadrats for biomass estimation. In addition, an unmanned aerial vehicle (UAV) was regularly flown over the vineyards for a period spanning more than four years in total over the two vineyards. The dataset presented here includes both raw data acquired during fieldwork and processed data derived from this raw inputs. The raw data consists of image series captured by two UAVs during each flight campaign, including RGB and multispectral imagery. Images were acquired between 2021-06-10 and 2022-07-29 for the first vineyard, and between 2023-06-08 and 2025-03-12 for the second vineyard. Based on these raw data, the processed data comprises spatial vectors, raster layers, and dense point clouds generated from UAV images using a Structure from Motion (SfM) photogrammetry workflow, at a 5 cm spatial resolution. The raster layers and dense point clouds provide specific information on vineyard characteristics for each UAV flight date, including elevation, vegetation indices, visible and near-infrared reflectance, and canopy height. In addition, the processed data include measurements of vegetation dry biomass, as well as separate measurements of dry biomass and leaf area measured for selected service crops species. This dataset can be reused for the calibration and/or evaluation of classification algorithms aimed at discriminating vines from the inter-row vegetation, or as part of a larger dataset to explore relationships between remotely-sensed vegetation indices and field-measured vegetation biomass or surface.
Why it matches plant phenotyping methodsUAV画像とSfM処理により、植生指数・樹冠高・バイオマス等の植物形質を取得した再利用可能なデータセットで、分類アルゴリズムの校正・評価用途も明示されており、植物フェノタイピング手法・データ基盤が中心です。
abstractThe dataset presented here includes both raw data acquired during fieldwork and processed data derived from this raw inputs.
Reproduction assets foundThe paper is a Data in Brief article describing a public dataset on Research Data Gouv (doi: 10.57745/MXM55R) containing UAV RGB/multispectral imagery, SfM-derived rasters and point clouds, and field-measured vegetation biomass/leaf-area data from two Mediterranean vineyards — directly the paper's phenotyping inputs. ADataset · publicollected in vineyards located in southern France near Montpellier (43°32.5243′N, 3°50.8240′E). Data are stored on Research Data Gouv, a remote storage solution curated by the French Department of Research.
Data accessibility
Repository name: Research Data Gouv
Data identification number: doi: 10.57745/MXM55R
Direct URL to data: https://doi.org/10.57745/MXM55R
Related research article
None
1.
Value of the Data
•
The fine scale imaging of vineyards (i.e., 5 cm resolution) allows for classification of the vegetation in the vineyard inter-rows, and subsequent exploration of its respective dynamics.
•Open asset ↗Research Data Gouv · 10.57745/MXM55Rlines:1-47Code / dataset availability confirmedCrossref · checked 5 Sept 2026
Abstract Photosynthetic organisms have evolved multiple non-photochemical quenching (NPQ) processes, providing photoprotection by safely dissipating excess excitation energy. These processes involve various molecular players functioning on overlapping timescales from seconds to days, making it challenging to isolate and quantify their individual kinetics. In this study, we perform whole-leaf chlorophyll fluorescence lifetime and xanthophyll concentration measurements on wild-type and various newly characterized NPQ mutants of Nicotiana benthamiana , a vascular land plant. Based on these measurements, we construct a fluorescence lifetime-based quantitative kinetic model that disentangles individual photoprotection components and, when integrated additively, accurately predicts wild-type and mutant NPQ behaviors under various light-dark regimes. Additionally, the model quantifies the per-molecule quenching effectiveness of various xanthophylls and the contributions of six quenching components (qE V , qE A, qE Z, qE L, qZ, and qI) across different genotypes. It also suggests improved overall quenching efficiency at specific VDE:ZEP:PsbS overexpression stoichiometries, aligning with previous studies and supporting translational efforts to optimize photoprotection and enhance crop yields under dynamic light environments.
Why it matches plant phenotyping methods葉の蛍光寿命測定を基盤に、NPQ成分を分離・定量するモデルを構築しており、植物の光防護状態を取得・抽出する方法が研究の中心です。
abstractBased on these measurements, we construct a fluorescence lifetime-based quantitative kinetic model that disentangles individual photoprotection components and, when integrated additively, accurately predicts wild-type and mutant NPQ behaviors under various light-dark regimes.
Reproduction assets foundThe paper's fluorescence lifetime/pigment phenotyping data and the NPQ model code are both publicly deposited on Zenodo (DOI 10.5281/zenodo.16755870), per explicit Data availability and Code availability statements.Dataset · publicThe data supporting the findings of this study are available within the article and at https://doi.org/10.5281/zenodo.16755870 .Open asset ↗Zenodo · 10.5281/zenodo.16755870lines:171-237Code · publicThe codes for NPQ models used in this study are available at https://doi.org/10.5281/zenodo.16755870 .Open asset ↗Zenodo · 10.5281/zenodo.16755870lines:171-237Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Rapid and accurate identification of DUS (Distinctness, Uniformity, and Stability) test traits in lettuce leaves is essential for advancing multi-omics-driven intelligent breeding. It also plays a critical role in germplasm protection and enhancing agricultural competitiveness. However, the phenotypic traits of lettuce leaves are highly diverse and complex due to both genotypic variation and environmental influences, posing significant challenges for precise DUS trait quantification. To address these challenges, we propose a high-precision phenotypic trait extraction pipeline and introduce an interpretable phenotypic fingerprinting framework for lettuce subgroup identification. First, a lightweight semantic segmentation network guided by group attention is developed to extract leaf components. Then, shape, color, and texture traits are comprehensively quantified. Following UPOV (International Union for the Protection of New Varieties of Plants) guidelines, we establish quantitative methods for seven DUS test traits: leaf shape, leaf tip shape, leaf margin shape, leaf vein shape, color hue, brightness, and anthocyanin coloration. Finally, PCA (Principal component analysis) was used to select 13 key traits, capturing over 95.82% of the total variance, for constructing "phenotypic ID" of lettuce varieties. Experiments conducted on 709 lettuce leaf image datasets showed that the accuracy of subgroup identification based on phenotypic fingerprints reached 98.59%. This study offers a scalable approach for automated DUS test trait evaluation and intelligent crop variety identification, providing a novel paradigm with strong potential for application in precision breeding and germplasm resource management.
Why it matches plant phenotyping methodsレタス葉画像からDUS形質を抽出・定量化する画像解析パイプラインを開発し、709画像で評価しており、植物フェノタイピング手法が研究の中心である。
abstractwe propose a high-precision phenotypic trait extraction pipeline and introduce an interpretable phenotypic fingerprinting framework for lettuce subgroup identification.
Reproduction assets foundThe article provides a public GitHub repository containing the authors' source code for the lettuce phenotypic fingerprint pipeline. The 709-image dataset and annotations are only available upon request, so they do not qualify as public assets.Code · publicThe data used to support the findings of this study are available upon request from the corresponding author, and the source code is accessible at https://github.com/qiuguangjie87/PP_Phenotypic_Fingerprint .Open asset ↗PP_Phenotypic_Fingerprintlines:263-278Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Time-series point clouds have emerged as an effective approach for precise, continuous crop monitoring and quantitative growth analysis. This study constructed a spatiotime-series point cloud dataset containing four species and eleven plant varieties, exploring crop organ instance segmentation, phenotypic parameter extraction, growth quantification, and canopy photosynthesis assessment. A skeleton-based framework for organ-level instance segmentation and time-series analysis is proposed, demonstrating robust performance across all four crops. To fully utilize the time-series data, a novel time-series leaf matching method was introduced, achieving a matching accuracy, defined as the proportion of correctly matched leaves, of over 0.823 for all species. By integrating the matching results with phenotypic parameter extraction, time-series phenotypic data were generated, and a phenotypic variation rate was defined as a suitable metric for quantifying crop growth. Furthermore, these results were integrated into a canopy photosynthesis model to derive key time-series photosynthetic metrics, including photosynthetic rate, absorbed light quantity, light energy utilization efficiency, and each crop organ's contribution to photosynthesis. These metrics provide insights into the crop's growth patterns and photosynthetic strategy. This study offers refined quantitative analysis of crop morphology and photosynthetic parameters through time-series point cloud segmentation, contributing valuable data for advancing plant biology research and enhancing the understanding of crop growth dynamics.
Why it matches plant phenotyping methods時系列点群から作物器官をセグメンテーションし、葉追跡、形態形質、成長量、光合成関連指標を抽出する手法が研究の中心であるため。
abstractA skeleton-based framework for organ-level instance segmentation and time-series analysis is proposed
Reproduction assets foundThe paper's Data availability statement explicitly provides authors' public URLs for a subset of the analysis code (GitHub) and the complete time-series 3D crop point cloud dataset (Baidu pan), both directly supporting this paper's phenotyping measurements and analysis.Code · publicA subset of the code and dataset used in this study is publicly available on our GitHub repository: https://github.com/JiarenZhou/LTPCDCCM .Open asset ↗JiarenZhou/LTPCDCCMlines:578-686Dataset · publicThe complete time-series 3D crop point cloud dataset can be downloaded from https://pan.baidu.com/s/1mNSDz4F0ZjOwmqzMuXozSQ?pwd=1234 .Open asset ↗lines:578-686Code / dataset availability confirmedOpenAlex · arXiv · checked 15 Sept 2026
Modeling plant growth dynamics plays a central role in modern agricultural research. However, learning robust predictors from multi-view plant imagery remains challenging due to strong viewpoint redundancy and viewpoint-dependent appearance changes. We propose a level-aware vision language framework that jointly predicts plant age and leaf count using a single multi-task model built on CLIP embeddings. Our method aggregates rotational views into angle-invariant representations and conditions visual features on lightweight text priors encoding viewpoint level for stable prediction under incomplete or unordered inputs. On the GroMo25 benchmark, our approach reduces mean age MAE from 7.74 to 3.91 and mean leaf-count MAE from 5.52 to 3.08 compared to the GroMo baseline, corresponding to improvements of 49.5% and 44.2%, respectively. The unified formulation simplifies the pipeline by replacing the conventional dual-model setup while improving robustness to missing views. The models and code is available at: https://github.com/SimonWarmers/CLIP-MVP
Why it matches plant phenotyping methods植物画像から葉数・植物齢を推定するマルチビュー表現学習手法を開発し、ベンチマークで性能評価しており、表現型取得・推定が研究の中心である。
abstractWe propose a level-aware vision language framework that jointly predicts plant age and leaf count using a single multi-task model built on CLIP embeddings.
Reproduction assets foundThe paper explicitly states that the model and code are publicly available at the authors' GitHub repository, which qualifies as a paper-specific public code asset.Code · publicm 7.74 to 3.91 and mean leaf-count MAE from 5.52 to 3.08 compared to the GroMo baseline, corresponding to improvements of 49.5% and 44.2%, respectively. The unified formulation simplifies the pipeline by replacing the conventional dual-model setup while improving robustness to missing views. The modela and code is available at: https://github.com/SimonWarmers/CLIP-MVP
Index Terms:
Plant phenotyping, Multi-view learning, Multi-task regression, Precision agriculture
† † address: † Computer Vision Lab, CAIDAS, IFI, University of Würzburg, Germany
‡ Technological University Dublin, Ireland
1 Introduction
Plant phenotyping from multiview imagery is crucial for precision agriculture, enabling non-Open asset ↗SimonWarmers/CLIP-MVPlines:1-53Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Reliable identification of maize leaf diseases is critical for mitigating crop losses, particularly in regions where farmers have limited access to experts. Although vision transformers (ViTs) have recently demonstrated strong performance in image recognition, their weak inductive bias and limited modelling of local texture patterns make them non-ideal for fine-grained maize leaf disease classification. To address these limitations, we propose ConvDeiT-Tiny, a lightweight hybrid ViT that improves DeiT-Ti by placing depthwise convolutions in parallel with multi-head self-attention modules in the first three transformer blocks. The local and global features captured by the convolution and attention modules are concatenated along the embedding dimension and fused using a multilayer perceptron. This results in richer token representations without significantly increasing model size. Across three datasets, ConvDeiT-Tiny (6.9M parameters) consistently outperformed DeiT-Ti, DeiT-Ti-Distilled, and DeiT-S (21.7M parameters) when trained from scratch. With transfer learning, ConvDeiT-Tiny achieved an accuracy of 99.15%, 99.35%, and 98.60% on the CD&S, primary, and Kaggle datasets, respectively, surpassing many previous studies with far fewer parameters. For explainability, we present gradient-weighted transformer attribution visualizations showing the disease lesions driving model predictions. These results indicate that injecting local inductive bias in early transformer blocks is beneficial for accurate maize leaf disease classification.
Why it matches plant phenotyping methodsトウモロコシ葉の病害状態を画像から分類する手法を新規に開発し、複数データセットで性能比較・検証しており、病害フェノタイピング手法が中心である。
abstractwe propose ConvDeiT-Tiny, a lightweight hybrid ViT that improves DeiT-Ti by placing depthwise convolutions in parallel with multi-head self-attention modules
Reproduction assets foundThe paper's Data Availability Statement points to a public GitHub repository containing the authors' program code and dataset splits (including their field-collected primary dataset). The paper also evaluates on the public Kaggle Corn or Maize Leaf Disease Dataset, a public plant-image dataset directly used for the论文'sCode · publicData Availability Statement: The program code and dataset splits for the three datasets used in this study,
including our primary data, can be found at https://github.com/DamarisWaema/ConvDeiT-Tiny.Open asset ↗DamarisWaema/ConvDeiT-Tinypdf-page:18 lines:1-60Dataset · public43. Ghose, S. Corn or Maize Leaf Disease Dataset. Available online:
https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset (Accessed on 10 July
2025).Open asset ↗pdf-page:21 lines:1-59Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 14 Sept 2026
Pea (Pisum sativum) production is challenged by drought stress. Traditional methods for assessing drought tolerance are limited, and high-throughput phenotyping (HTP) can facilitate the rapid and automated assessment of plant traits. Herein, 180 Pisum spp. accessions were evaluated using an indoor HTP platform under two irrigation treatments, control (70% field capacity) and drought stress (30% field capacity), for 50 days. A combination of digital phenotyping via imaging and manual measurements was used to analyse biomass-related, architectural, and physiological traits. Drought conditions resulted in significant reductions in biomass-related traits including fresh weight (47%), total leaf area (43%), and dry weight (41%). In contrast, PSII photochemical efficiency, leaf weight ratio, and solidity showed negative sensitivity index values (ranging from -7% to -1%), indicating comparatively lower sensitivity to drought and suggesting relative stability of these traits under water-limited conditions. The high heritability value for water use efficiency (0.87) suggests that this parameter may be useful for distinguishing pea's responses to suboptimal soil moisture levels. Principal component analysis (PCA) highlighted patterns of trait variation and associations among biomass-related traits, such as fresh weight, dry weight, and leaf area, which were sensitive to drought conditions. This suggests that the plants may use a combination of strategies to cope with water limitations. Furthermore, studying the significant variation in drought response among the diverse Pisum species and subspecies revealed distinct adaptation strategies. These findings support the development of crops that are resilient to the negative effects of climate change.
Why it matches plant phenotyping methods屋内HTPプラットフォームと画像ベースのデジタルフェノタイピングを用いて、多数アクセッションの形態・生理形質を取得・解析しており、フェノタイピング手法の実質的な適用が研究の中心です。
abstracthigh-throughput phenotyping (HTP) can facilitate the rapid and automated assessment of plant traits
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicThe analysis software for the RGB side‐view imaging has been developed in Python by the NPEC data team, the source is published on Github, accessible via this link: https://github.com/NPEC‐NL/greenhouse_m5 .Open asset ↗NPEC‐NL/greenhouse_m5lines:68-83Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 5 Sept 2026
Abstract Observer bias and inconsistencies in traditional plant phenotyping methods limit the accuracy and reproducibility of fine-grained plant analysis. To address these limitations, TomatoMAP is introduced as a comprehensive dataset for Solanum lycopersicum . The dataset contains 68,080 RGB images: 3,616 high-resolution macrophotographs (3648 × 5472) with semantic annotations, and 64,464 moderate-resolution images (1080 × 1440) captured from 12 plant poses at four camera elevations. Each image is accompanied by manually annotated bounding boxes for seven regions of interest (leaves, panicle, flower clusters, fruit clusters, axillary shoot, shoot, and whole-plant area) and by labels spanning 50 BBCH classes representing phenologically growth stages. A general cascading structure is proposed. For real-time applicability, models emphasizing the accuracy-efficiency trade-off (MobileNetv3, YOLOv11, and Mask R-CNN) are prioritized and benchmarked against multiple state-of-the-art models. Performance is assessed using accuracy, mAP, inference FPS, and normalized confusion matrices. In a study involving five domain experts, AI models trained on TomatoMAP achieves comparable accuracy levels. Reliability of automated fine-grained phenotyping is supported by Cohen’s Kappa statistics and inter-rater agreement heatmaps.
Why it matches plant phenotyping methodsトマトの多視点画像、器官領域・生育ステージ注釈を備えたデータセットを構築し、画像モデルの精度・効率・専門家一致度をベンチマークしており、植物フェノタイピング手法が中心である。
titleTomato Multi-Angle Multi-Pose Dataset for Fine-Grained Phenotyping.
Reproduction assets foundThe paper's authors publicly release their analysis code (dataset construction scripts for TomatoMAP-Cls/Det and model training/evaluation code) on GitHub. The TomatoMAP phenotype image dataset itself is deposited at e!DAL (10.5447/ipk/2025/14), but no matching URL is present in the allowed list, so only the code assetCode · publicThe scripts for constructing TomatoMAP-Cls and TomatoMAP-Det, as well as the code used for model evaluation, are available at: https://github.com/0YJ/TomatoMAP.Open asset ↗https://github.com/0YJ/TomatoMAPhtml-lines:423-479Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
We used domestication as an in vivo replicated experiment to investigate how divergent selection has shaped the evolution of multivariate phenotypic spaces. We measured 11-57 qualitative and quantitative traits in 13 species, either unique or shared between species, and established a framework for cross-species comparisons. Our results revealed significant convergence that translated into a cross-species domestication syndrome. Most species exhibited a reduction of the multivariate phenotypic space during domestication. We brought evidence that Near-Infrared spectra measured on leaves reflect phenotypic evolution unrelated to domestication, enabling its use as a control for sampling effects across species. Building on this, we developed a multivariate phenotypic divergence index (mPDI) to rank species by the extent of phenotypic divergence under domestication. We found a high disjunction of wild and domestic phenotypic spaces in all species. Neither the mPDI nor the relative size of wild vs domestic multivariate phenotypic spaces was influenced by the domestication timing or mating system. Lastly, we observed a progressive decoupling of trait correlations with increasing time since domestication. In addition to introducing a new index that can be applied for cross-species comparisons, our study uncovers recurring patterns shared among species, pointing to general principles underlying plant domestication.
Why it matches plant phenotyping methods多変量形質空間を比較する枠組みと新しいmPDI指標を開発しており、植物形質の統合・比較手法が明示的な貢献であるため。
abstractestablished a framework for cross-species comparisons
Reproduction assets foundThe paper's phenotypic data, NIR spectra, and trait ontology are deposited at doi 10.57745/QWEKVK, and the authors' R analysis scripts are publicly available on INRAE Forge. Both are paper-specific, public, and actionable.Dataset · publicPhenotypic data and NIR spectra are available on https://doi.org/10.57745/QWEKVK .Open asset ↗10.57745/QWEKVK · 10.57745/QWEKVKlines:283-349Code · publicR scripts are available on the INRAE Forge at https://forge.inrae.fr/gqe‐gevad/domisol_phenotypic_spaces .Open asset ↗forge.inrae.fr/gqe‐gevad/domisol_phenotypic_spaceslines:283-349Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Abstract Diseases of tomato leaves are significant threats to the global food security and agricultural production. The old method of diagnosis is not reliable and is time consuming, and there is a demand to have effective and accurate automated systems. The paper uses transfer learning using Inception-V3 and Inception-ResNet-V2 network to detect tomato leaf diseases using an open dataset. To encourage generalizability, data augmentation and preprocessing techniques were used, whereas Grad-CAM was used to encourage visual interpretability. Experimentally, it has been demonstrated that Inception-ResNet-V2 and Inception-V3 performed with 92.33 and 89.33 accuracy, respectively, which is higher than the other existing methods. These results demonstrate the possibility of deep learning to improve precision agriculture and prepare further development of real-time and field-deployable systems of disease detection.
Why it matches plant phenotyping methodsトマト葉の病害状態を画像から深層学習で自動推定する手法が研究の中心であり、植物病害フェノタイピングに該当する。
abstractThe paper uses transfer learning using Inception-V3 and Inception-ResNet-V2 network to detect tomato leaf diseases using an open dataset.
Reproduction assets foundThe paper's Data Availability Statement explicitly identifies the public Kaggle tomato leaf image dataset used to train and evaluate the deep learning models, making it a paper-specific, publicly actionable asset. No author analysis code or trained model checkpoints are disclosed.Dataset · publicThe dataset used in this study is publicly available on Kaggle. It can be accessed at:
https://www.kaggle.com/datasets/kaustubhb999/tomatoleaf
The dataset contains labeled images of healthy and diseased tomato leaves and was used for training
and evaluating the proposed deep learning model.Open asset ↗Kaggle · kaustubhb999/tomatoleafpdf-page:30 lines:1-9Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
To address the inefficiency and high cost of manual counting of tobacco leaves, this study proposes a UAV-based method for automatic leaf counting in field-grown tobacco using 3D point clouds and an improved PointNext network. Although UAV imagery has been applied to crop phenotyping, most existing UAV-based leaf-counting methods still rely on 2D images or hand-crafted features and rarely exploit 3D point clouds with dedicated leaf-level segmentation, which limits accuracy and robustness under leaf overlap, variable viewing angles, and complex field backgrounds. In this work, oblique UAV photogrammetry is used to reconstruct individual plants into 3D point clouds, and a segmentation network, SRW-PointNext, is developed by integrating an SCSA attention mechanism and a Residual-SegHead to enhance feature extraction and segmentation performance, while a re-weighted loss alleviates class imbalance. Leaf point clouds are then clustered using MeanShift to obtain leaf counts. Experiments on field-grown tobacco demonstrate that the proposed method achieves a point-cloud segmentation precision of 92.09%, a MIoU of 76.13%. Compared with the original PointNext baseline, SRW-PointNext increased MIoU and overall precision by 3.34% and 2.42% respectively. The final accuracy rate of leaf counting was 92.61%, effectively achieving accurate and stable leaf counting under actual field conditions, and providing technical support for digital management, yield estimation and seedling breeding in tobacco production.
Why it matches plant phenotyping methodsUAV三次元画像と改良セグメンテーション手法により圃場タバコの葉数を推定する方法を開発・検証しており、表現型取得が研究の中心である。
abstractthis study proposes a UAV-based method for automatic leaf counting in field-grown tobacco using 3D point clouds and an improved PointNext
Reproduction assets foundThe paper reports a UAV-based tobacco leaf counting method with an annotated 1000-plant point cloud dataset and SRW-PointNext code, both explicitly declared publicly available at author-provided Zenodo and GitHub URLs matching the allowed list.Dataset · publicData supporting the reported results can be found at: https://zenodo.org/records/15130271 .Open asset ↗zenodo · 15130271lines:531-564Code · publicThe code used in this study is available at: https://github.com/Nan20377/SRW-Pointnext.git .Open asset ↗github · Nan20377/SRW-Pointnextlines:531-564Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Maize is a vital global crop, but its productivity is often threatened by plant diseases, highlighting the need for precise and timely diagnostic methods. Traditional manual inspection is inefficient and prone to errors, motivating the development of automated solutions. Recent advances in computer vision and deep learning have enabled effective automated plant disease diagnosis. While Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) have shown promise in plant disease classification, CNNs struggle to capture global contextual information, and ViTs require large datasets and high computational resources. Inspired by mixture-of-experts (MoE) architectures, we propose a lightweight hybrid model that integrates CNN and ViT components, adaptively emphasizing local or global features based on input characteristics. Evaluated on a novel, real-world dataset of full maize plant images, our approach achieves 99.90% classification accuracy, significantly outperforming state-of-the-art baselines such as MobileViT, PiT, EdgeNeXt, and DeiT. These results demonstrate that lightweight hybrid architectures can deliver high-performance disease diagnosis suitable for practical agricultural deployment. The code is available at: https://www.github.com/sabermehdipour/MXiT .
Why it matches plant phenotyping methodsトウモロコシ全身画像から病害状態を推定する軽量CNN-ViT手法を開発・評価しており、植物表現型取得・判定が中心的です。
abstractwe propose a lightweight hybrid model that integrates CNN and ViT components
Reproduction assets foundThe paper's authors' MXiT analysis code is publicly available via a GitHub URL stated in the abstract, and the PlantVillage image dataset used for evaluation is publicly available. The Plant Scanner maize dataset is paper-specific but only available upon request, so it is listed as request_only.Code · publicThe code is available at: https://www.github.com/sabermehdipour/MXiT.Open asset ↗sabermehdipour/MXiThtml-lines:1-77Dataset · publicThe PlantVillage dataset is publicly available (https://github.com/spMohanty/PlantVillage-Dataset).Open asset ↗spMohanty/PlantVillage-Datasethtml-lines:707-785Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Grapevines ( Vitis vinifera L.) undergo structural and physiological changes throughout the growing season, progressing through distinct phenological stages that require regular monitoring. This dataset consists of high-resolution point cloud data acquired with a stationary terrestrial laser scanner (TLS) to document grapevine development from early leaf development to dormancy. Georeferenced point clouds were generated from 15 TLS scans along two vineyard rows at nine phenological stages. The dataset also includes multispectral and RGB photogrammetric point clouds and orthorectified raster products from an unmanned aerial vehicle survey conducted before harvest. Ground-truth measurements leaf area index, grape production, and pruning wood biomass were collected for each monitored grapevine. As a result, the dataset provides multi-temporal TLS observations that support grapevine structural analysis and development, phenological monitoring, and can be used for the development of AI-based models for precision viticulture.
Why it matches plant phenotyping methodsブドウの生育・構造・フェノロジーを対象とするTLS点群および関連画像データセットであり、植物フェノタイピング用の再利用可能なデータ基盤として中心的です。
titleTLS-grapevine2024: A terrestrial laser scanner point cloud dataset of grapevines at different phenological stages.
Reproduction assets foundThe paper is a Data in Brief article describing the TLS-grapevine2024 dataset itself, publicly deposited on Zenodo with DOI 10.5281/zenodo.16751663. This is a paper-specific, openly available asset containing the TLS point clouds, UAV imagery/rasters, and ground-truth agronomic measurements (LAI, grape production, prunDataset · publicditions: clear sky.
Data source location
Institution: University of Trás-os-Montes e Alto Douro
City/Town/Region: Arroios, Vila Real, Norte
Country: Portugal
Coordinates: 41°17′28.83″N 7°43′17.90″W,
Altitude: 435 m
Data accessibility
Repository name: Zenodo
Data identification number: 10.5281/zenodo.16751663
Direct URL to data: https://doi.org/10.5281/zenodo.16751663
Related research article
None
1.
Value of the Data
•
This dataset covers nine phenological stages of grapevine growth from April 2024 to January 2025, providing multi-temporal terrestrial laser scanner (TLS) observations for structural and phenological analysis.
•
It includes TLS point clouds collected at multiple stages and muOpen asset ↗Zenodo · 10.5281/zenodo.16751663lines:1-50Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Common beanLeafClassificationStress / disease detectionDisease symptoms / severity
Early identification of bean leaf diseases, particularly Angular Leaf Spot and Bean Rust, is vital for ensuring crop productivity and global food security, especially within smallholder farming systems where disease outbreaks can rapidly escalate and cause severe yield losses. Conventional disease identification through visual inspection is labor-intensive, subjective, and highly dependent on expert knowledge, making it impractical for large-scale agricultural monitoring. Although recent deep learning-based approaches have demonstrated impressive accuracy in plant disease classification, their inherent “black-box” nature significantly limits real-world adoption, as farmers and agronomists often lack the ability to understand, trust, or act upon unexplained predictions. To address these challenges, this study proposes an automated and explainable disease diagnostic framework based on a Vision Transformer (ViT-B/16) architecture optimized through transfer learning from ImageNet. Unlike traditional convolutional neural networks that primarily focus on localized features, the Vision Transformer processes images as a sequence of flattened patches and leverages self-attention mechanisms to capture long-range dependencies and global contextual patterns across the entire leaf surface. This global representation enables the model to detect subtle and spatially distributed disease symptoms that are often overlooked by CNN-based approaches. To further enhance transparency and interpretability, GradCAM + + is integrated into the framework as an explainable artificial intelligence (XAI) mechanism. This method generates class-specific heatmaps that visually highlight the exact pathological regions influencing the model’s predictions, thereby establishing a human-interpretable validation loop for farmers, agronomists, and domain experts. The proposed framework was evaluated on the publicly available I-Bean dataset, achieving a validation accuracy of 97.52% along with strong precision, recall, and F1-score performance. The generated GradCAM + + visualizations consistently demonstrate the model’s sensitivity to true diseased regions, reinforcing both the reliability and trustworthiness of its predictions. By combining high-capacity global feature learning with visual explainability, the proposed approach offers a scalable, transparent, and practical solution for real-world precision agriculture. This framework not only enhances diagnostic accuracy but also bridges the critical gap between model performance and user trust, enabling informed decision-making and timely disease management in modern farming environments.
Why it matches plant phenotyping methods画像から豆葉の病害症状を分類・可視化する手法が研究の中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として適格。
abstractThis method generates class-specific heatmaps that visually highlight the exact pathological regions influencing the model’s predictions
Reproduction assets foundThe paper uses the publicly available I-Bean bean leaf disease image dataset (Healthy, Angular Leaf Spot, Bean Rust) and points to it via a Data availability DOI (10.21227/4k7y-vs03), which is an allowed URL. No author analysis code or trained model checkpoint is explicitly deposited.Dataset · publicPSP and SNT: Problem Formulation and MethodologyAS and DS: Implementation and VisualizationMVV PK and KB: Original Draft and Supervision.
Funding
Open access funding provided by Symbiosis International (Deemed University). This research received no external funding.
Data availability
[https://dx.doi.org/10.21227/4k7y-vs03]
Declarations
Competing interests
The authors declare no competing interests.
The authors declare that they have no conflict of interest.
References
1.
Wang Y Wang Q Su Y Jing B Feng M
Detection of kidney bean leaf spot disease based on a hybrid deep learning model
Sci. Rep. 2025 15 1 11185
10.1038/s41598-025-93742-7
40169647
POpen asset ↗10.21227/4k7y-vs03lines:325-415Code / dataset availability confirmedOpenAlex · Europe PMC · bioRxiv · checked 5 Sept 2026
1 Abstract Accurate quantification of plant disease is essential for resistance breeding, variety testing, and precision agriculture, yet visual ratings are limited by subjectivity, low precision, and restricted throughput. Image-based phenotyping can address these limitations, but field applications face substantial challenges due to spatial heterogeneity, symptom-level diagnostic requirements, and the need for very high-resolution imagery with limited spatial coverage. This introduces a fundamental trade-off: high-resolution images provide precise local measurements of disease, but spot-level estimates can be highly variable within experimental units. We analyzed a large image data set of wheat foliar diseases to characterize the distribution, spatial dependence, and aggregation behavior of spot-level severity estimates in plots. We combined high-resolution macro-scale imaging with focus bracketing to increase the sampled leaf area. Our results highlight focus bracketing as a promising approach for simultaneous diagnosis and quantification of disease in field plots. Autocorrelation in severity estimates both within focal image stacks and across plot positions was comparable, with 10 focal stack images or 10 positions per plot contributing approximately 2.5 independent observations each. Modeling plot-level severity as a latent Beta-distributed variable enabled robust estimation of mean severity and associated uncertainty. This supports both hypothesis testing and efficient sampling across the full range of disease severity associated with genotypic diversity and seasonality of developing epidemics. The proposed imaging approach is non-invasive and, in principle, transferrable to autonomous ground-based phenotyping platforms, offering the potential to shift the dominant source of uncertainty in estimating disease severity from measurement-related limitations toward biologically and environmentally driven variability in disease expression.
Why it matches plant phenotyping methods高解像度画像とフォーカスブラケティングを用いて植物病害の重症度を定量化し、圃場プロット単位の推定精度と不確実性を評価する手法が研究の中心であるため。
abstractWe combined high-resolution macro-scale imaging with focus bracketing to increase the sampled leaf area.
Reproduction assets foundThe paper states that R code to reproduce the full analysis (Beta-distribution modeling, autocorrelation/AR(1) mixed models, effective sample size estimation for wheat disease severity phenotyping) is publicly available on the authors' GitHub repository. The repository name appears truncated in the supplied text ('plotCode · publicR-code to reproduce the full analysis is available at https://github.com/and-jonas/plot-spot-Open asset ↗and-jonas/plot-spot-pdf-page:9 lines:1-61Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Plant Leaf Diseases represent a significant risk to global agricultural production. Crops ranging from Peppers and Tomatoes to Potatoes are affected by these diseases. Traditional methods of identifying leaf diseases based primarily on visual inspection have historically been slow and relatively inaccurate. A deep learning-based solution for the automated identification of plant leaf diseases utilizes Convolutional Neural Networks (CNNs) as the primary methodology. To identify leaf images into 15 disease categories, both pre-trained models such as VGG16, EfficientNetB3, Inception V3 and a custom CNN model were used. Using pre-trained models to allow for the use of Transfer Learning helps to mitigate some of the issues associated with computational resource limitations and data limitations in providing faster convergence rates and higher accuracy when compared to training a model from scratch. The Plant Village image collection which contains over 20,000 images of different plant leaf diseases was utilized for training and testing purposes. Each model's performance was evaluated based on its accuracy, loss and generalization capabilities. Additionally, each model was fine-tuned through hyperparameter optimization. As a result, the model that achieved the highest validation accuracy rate of 95% was the EfficientNetB3 model while the second highest accuracy rate was achieved by the Inception V3 model at 92%. This methodology provides an excellent answer to addressing early disease detection, enabling farmers to take the necessary actions quickly to reduce their losses and maximize their harvest.
Why it matches plant phenotyping methods植物葉画像から病害状態を推定する深層学習手法を比較・評価しており、病害フェノタイピングと手法検証が研究の中心です。
abstractA deep learning-based solution for the automated identification of plant leaf diseases utilizes Convolutional Neural Networks (CNNs) as the primary methodology.
Reproduction assets foundThe paper's plant-phenotyping measurements (CNN classification of 15 leaf disease classes) were performed on the public PlantVillage-derived Kaggle 'Plant Disease Dataset' by E. Marrex, which the authors cite as their training/testing image source. No author analysis code, trained model checkpoints, or paper-specific衍生Dataset · publicD. E. Popescu, M. K. Chowdary, and J. Hemanth, "Deep learning-based leaf
disease detection in crops using images for agricultural applications," Agronomy, vol. 12, no. 10,
p. 2395, 2022. doi: 10.3390/agronomy12102395. Available:
https://doi.org/10.3390/agronomy12102395.11. E. Marrex, "Plant Disease Dataset," Kaggle, Available:
https://www.kaggle.com/datasets/emmarex/ plantdisease. [Accessed: 03- Apr-2025].Open asset ↗Kagglepdf-raw-page:12 lines:1-8Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Tea leaf diseases seriously affect its yield and quality, and consequently there is an urgent need for intelligent detection methods with high precision and edge deployment capabilities. To address low detection accuracy in complex backgrounds, overfitting due to limited data, and redundant parameters for existing methods, this paper proposes an improved lightweight detection model FCHE-YOLO based on the YOLO11, which aims to achieve rapid and accurate identification of tea leaf disease combining low altitude remote sensing with unmanned aerial vehicle (UAV). The model has made three key optimizations in the structure: Introduce the self-developed lightweight backbone module FC_C3K2, which significantly reduces computation and parameter count while enhancing the robustness of the model to complex scenarios; construct an efficient feature fusion structure HSFPN, optimizing multi-scale information integration and compressing model volume; design the detection head Efficient Head, integrating group convolution and lightweight attention mechanism to improve detection accuracy and suppress overfitting. The experimental results from the self built tea gardens show that the FCHE-YOLO improves the average accuracy (mAP) from 94.1% to 98.1% compared to the benchmark model YOLO11, with an improvement of 4.0 percentage points. Meanwhile, the inference speed of the model increases from 43.3 FPS to 47.5 FPS, with an increase of 9.0%, meeting the real-time detection requirements. More importantly, by network structure optimization, the model's computational complexity is significantly reduced: The floating-point operations per second (FLOPs) decreases from 6.4 G to 4.2 G, with a decrease of 34.3%, and the parameter count decreases from 2.59 M to 1.46 M, with the compression rate reaching 38.9%, which makes the model more suitable for deployment on resource-constrained UAV edge devices. The final test show that the FCHE-YOLO significantly reduces the missed-detection rate, owns better detection accuracy and deployment practicality, and is suitable for real-time monitoring scenarios of tea leaf diseases with UAVs.
Why it matches plant phenotyping methods茶葉の病害状態をUAV画像から検出する軽量深層学習手法を開発・評価しており、植物病害表現型の取得が中心的な技術貢献である。
abstractthis paper proposes an improved lightweight detection model FCHE-YOLO based on the YOLO11, which aims to achieve rapid and accurate identification of tea leaf disease combining low altitude remote sensing with unmanned aerial vehicle (UAV).
Reproduction assets foundThe paper's Data Availability Statement points to a public figshare repository containing the study's relevant data (UAV tea leaf disease imagery/dataset). No separate author code deposit is stated.Dataset · publicAll relevant data for this study are publicly available from the figshare repository (https://figshare.com/s/316807b23895bc3ba3ae).Open asset ↗figsharehtml-lines:693-736Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Abstract Plant diseases continue to pose a significant threat to worldwide food security, resulting in notable yield reductions and economic consequences. Automated disease diagnosis through machine learning has arisen as a potential solution; nevertheless, current methods frequently have difficulty in capturing both detailed local attributes and overarching contextual patterns found in plant leaf images. This study presents a thorough comparative examination of conventional and deep learning methods—such as Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), YOLO, Support Vector Machines (SVMs), and Random Forests—for the classification of multi-class plant diseases. To overcome the constraints of individual CNN and transformer models, a new hybrid framework that integrates EfficientNet-B7 for strong spatial feature extraction with a Vision Transformer (ViT-B16) for comprehensive contextual modeling is suggested. The system is assessed on an extensive dataset consisting of 21,534 images covering 38 classes of plant diseases and healthy specimens. Experimental findings show that the suggested hybrid model reaches an accuracy of 98.13%, surpassing standalone CNN baselines and other rival models, while consistently achieving high precision, recall, and F1-scores for all classes. The results emphasize the success of combining convolutional and transformer-based models for scalable and precise plant disease detection, aiding the creation of smart decision-support systems for precision farming.
Why it matches plant phenotyping methods植物葉画像から病害状態を分類する新規ハイブリッド画像解析手法を提案し、複数手法との比較評価と大規模データセットでの検証を行っており、フェノタイピング手法が中心である。
abstractAutomated disease diagnosis through machine learning has arisen as a potential solution
Reproduction assets foundThe paper's plant disease image dataset (New Plant Diseases Dataset on Kaggle) and the authors' complete hybrid CNN–ViT implementation (GitHub repository with Zenodo DOI) are both publicly and explicitly available.Dataset · publicThe data that support the findings of this study are openly available in the New Plant Diseases Dataset at Kaggle [https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset/data].Open asset ↗Kaggle · new-plant-diseases-datasethtml-lines:296-327Code · publicThe source code, including model architecture, training scripts, evaluation routines, and Google Colab notebooks for inference, is hosted on GitHub at: https://github.com/mohdzunaidahmed15-ui/hybrid-cnn-vit-plant-disease-diagnosis.Open asset ↗GitHub · mohdzunaidahmed15-ui/hybrid-cnn-vit-plant-disease-diagnosishtml-lines:296-327Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Leaves are central indicators of photosynthesis and plant growth status, and their precise monitoring is crucial for smart agriculture. Dense leaf detection, as a foundation for leaf morphology analysis, must address challenges such as occlusion and overlap, directly enabling key tasks including phenotypic trait extraction, disease identification, and yield estimation. Leaves are the most important plant organs, and monitoring leaves is a crucial aspect of crop surveillance. Dense leaf detection plays an important role as a fundamental technology for leaf monitoring. Existing dense leaf detection methods rely on traditional modular detectors and generic feature extraction, lacking designs tailored to real-world dense leaf scenarios. The methods for dense leaf detection generally use traditional modular detectors and general feature extraction techniques, without designing methods specifically for dense leaves in reality. In detail, in complex field scenarios, it still faces challenges like incomplete individual feature extraction due to high leaf overlap and difficult network convergence caused by excessive leaf density. To this end, we propose the Leaf-DETR framework, which effectively addresses these challenges through the Progressive Feature Fusion Pyramid Network (P-FPN) and the Crowded Query Refinement Strategy (CQR). First, we construct the largest dense leaf detection dataset to date, containing 1696 images and 85,375 annotation boxes. Second, P-FPN alleviates the feature confusion problem of overlapping leaves through the multi-stage fusion of features and the Adaptive Feature Aggregation module (AFA), enhancing the interaction between low-level details and high-level semantics. Third, the CQR strategy significantly reduces the matching cost of crowded candidate boxes and improves the network convergence efficiency by culling a crowded query method and introducing a one-to-many matching mechanism. Finally, experimental results show that Leaf-DETR improves mAP@50 by 1% and AR@300 by 1.4% over the baseline model on our self-constructed dataset, outperforming existing detection methods. Furthermore, the model exhibits extremely fast training convergence and demonstrates strong generalization capability on both field-collected monitoring images and other staple crops, fully highlighting its practical value in complex agricultural scenarios. Finally, experiments show that Leaf-DETR outperforms existing detection methods on the self-built dataset and demonstrates good performance generalization in monitoring collected images, as well as for other staple food crops, which verifies its practicality in complex agricultural scenarios. The code and detailed information are available at http://leafdetr.samlab.cn.
Why it matches plant phenotyping methods葉の密集検出モデルとデータセットを開発・評価し、葉形態などの表現型抽出を可能にする画像ベース手法が研究の中心であるため。
abstractDense leaf detection, as a foundation for leaf morphology analysis, must address challenges such as occlusion and overlap, directly enabling key tasks including phenotypic trait extraction, disease identification, and yield estimation.
Reproduction assets foundThe paper's data availability statement explicitly points to an authors' public site (http://leafdetr.samlab.cn) hosting the Leaf-DETR code and detailed information, qualifying as a paper-specific public code asset. The self-constructed KiwiFruitLeaf dataset (1696 images, 85,375 annotation boxes) is described but its公开Code · publicThe code and detailed information are available at http://leafdetr.samlab.cn . For testing purposes, detailed instructions for running the model can be found in the repository's README file.Open asset ↗lines:504-529Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
Abstract PlantCV is an open‐source Python project aimed at developing tools to address a range of image‐based, plant phenotyping questions. PlantCV has been used for more than 10 years to automate trait collection from image data, and the newest release, PlantCV version 4, continues to lower the barrier to entry for users without substantial coding experience through extensive example use‐case tutorials and simplified installation. In addition to usability, we document added functionality since the release of PlantCV v2, including support for more image types such as fluorescence, thermal, and hyperspectral data. Finally, we describe the development of a new subpackage focused on morphological trait measurements like leaf angle, and demonstrate its utility as compared to more manual methods of data collection.
Why it matches plant phenotyping methodsPlantCV v4は、画像から植物形質を自動抽出するオープンソースソフトウェアの開発・機能拡張・比較評価を主題としており、植物フェノタイピング手法が中心である。
abstractPlantCV is an open‐source Python project aimed at developing tools to address a range of image‐based, plant phenotyping questions.
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits the authors' analysis scripts on GitHub (danforthcenter/plantcv-4-paper), which directly reproduces this paper's phenotyping analyses.Code · publicerest.
DATA AVA I L A B I L I T Y S TAT E M E N T
Links to code, tutorials, documentation, and other resources
are available on the PlantCV homepage at https://plantcv.org. PlantCV source code is available on GitHub at https://
github.com/danforthcenter/plantcv. Scripts used for analyses
in this paper are available on GitHub at https://github.com/danforthcenter/plantcv-4-paper.O RC I D
HaleySchuhl https://orcid.org/0000-0002-8825-8297
KeelyE. Brown https://orcid.org/0000-0002-5371-5830
ParagK. Bhatt https://orcid.org/0000-0002-0396-6412
DominikSchneider https://orcid.org/0000-0002-5846-5033
Anna L. Casto https://orcid.org/0000-0002-9597-0514
Lucia Acosta-Gamboa https://orcid.org/0000-0001-77Open asset ↗danforthcenter/plantcv-4-paperpdf-raw-page:15 lines:1-97Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Published15 Feb 2026Journal of Artificial Intelligence and Engineering Applications (JAIEA)Cited by 0 · OpenAlex ↗
Monitoring plant health is an important factor in maintaining agricultural productivity. Manual identification of leaf diseases requires expert knowledge and is prone to errors due to visual similarities among disease symptoms. This study aims to develop a plant leaf disease classification system based on digital images using a Convolutional Neural Network (CNN) approach. The dataset consists of plant leaf images representing three disease classes: Corn–Common rust, Potato–Early blight, and Tomato–Bacterial spot. Prior to model training, the images undergo preprocessing steps including image resizing and pixel normalization. The performance of the CNN model is evaluated using a testing dataset that is not involved in the training process, employing accuracy, confusion matrix, precision, recall, and F1-score as evaluation metrics. Experimental results show that the proposed model achieves a test accuracy of 95.56%, with balanced performance across all disease classes. In addition to quantitative evaluation, the trained model is implemented in a Streamlit-based application, allowing users to upload plant leaf images and obtain disease classification results interactively. The findings indicate that the CNN-based approach is effective for plant leaf disease classification and has potential application as an early decision-support system for plant health monitoring.
Why it matches plant phenotyping methods植物葉の画像から病害状態を推定するCNN分類法を開発し、独立テストデータで性能評価しているため、植物フェノタイピング手法が中心である。
abstractThis study aims to develop a plant leaf disease classification system based on digital images using a Convolutional Neural Network (CNN) approach.
Reproduction assets foundThe paper's phenotyping input is a publicly available PlantVillage image dataset (900 leaf images across three disease classes) obtained from Kaggle, with an explicit authors' URL. No author analysis code or trained model is publicly deposited.Dataset · publicleaf disease images obtained from the PlantVillage Dataset, which is publicly available through the Kaggle platform [17]. The dataset is
organized using a folder-based class structure, where each folder represents a specific leaf disease category. In this study, three disease
classes are used—Corn–Common rust, Potato–Early blight, and Tomato–Bacterial spot—with 300 images per class, resulting in a total of
900 images.Open asset ↗Kagglepdf-page:3 lines:1-51Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published14 Feb 2026International Journal of Engineering Trends and TechnologyCited by 0 · OpenAlex ↗
The study is a design and development of a strong disease detection system of cucumber and grape leaves with noisy image data, focusing on the ability to withstand salt-and-pepper and Gaussian noises. The image datasets used in agriculture are usually affected by noise because of changes in light, sensor defects, and environmental conditions, which may lead to lower diagnostic accuracy. In order to address this, the proposed system incorporates high noise reduction methods whereby a median filter and a Gaussian filter are used to restore the image quality without compromising on the important leaf texture information. After processing, colour, texture, and shape are used to extract features, which are effective in extracting disease-specific visual representations. These fine features are then trained on various optimized machine learning models, such as Light Gradient Boosted Machine (LGBM), Quantum Support Vector Machine (QSVM), a Modified Random Forest (MRF) with adaptive weighted features, and a Multi-SVM classifier with a custom kernel to map nonlinear features. Through experimental analyses, the proposed ensemble framework is shown to be highly accurate, robust, and noise-tolerant as opposed to the traditional frameworks. The hybrid method is effective in recognizing the significant cucumber diseases and grapes, including powdery mildew, downy mildew, and anthracnose, which will be utilized in the noisy agricultural conditions in the real world. In general, this system offers a noise-resistant, reliable, and computationally efficient system to detect early signs of plant diseases, which can be used in sustainable crop monitoring and precision farming.
Why it matches plant phenotyping methods植物葉の画像から病徴を推定する画像処理・特徴抽出・機械学習システムの設計開発が中心であり、植物病害状態のフェノタイピング手法に該当する。
abstractThe study is a design and development of a strong disease detection system of cucumber and grape leaves with noisy image data
Reproduction assets foundThe paper uses two public Kaggle leaf-image datasets (cucumber and grape) as its phenotyping inputs; both are publicly accessible with URLs given in the references. No author code or model checkpoints are reported.Dataset · publicGrape Disease Dataset, which was collected on Kaggle
[17], is an extensive collection of images created for the
classification and analysis of different diseases in grape
leaves.Open asset ↗Kagglepdf-raw-page:7 lines:1-64Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Potato leaf diseases, if left undetected, threaten food security in agricultural economies and cause substantial crop losses. To address this critical challenge, we developed an AI-based system called the Enhanced Single Shot Multibox Detector(EF-SSD), a variant of SSD that integrates multiscale feature fusion and Squeeze-and-Excitation attention to improve fine-grained lesion detection. The enhanced model processes high-resolution images (512×512 pixels) and analyzes leaves at ten magnification levels, enabling it to identify even minor signs of infection. The inclusion of Squeeze-and-Excitation filters allows the system to focus more effectively on characteristic disease patterns, increasing detection precision. After scanning the leaves, the system applies advanced image processing techniques to localize disease regions and assess their severity. We evaluated EF-SSD using 2,500 labeled potato leaf images representing healthy plants and cases of early and late blight. The proposed model achieved a mean Average Precision (mAP) of 97% at 0.5 IoU, an F1-score of 95%, and an Intersection over Union (IoU) of 89%, outperforming advanced detectors such as YOLOv5, YOLOv8, RetinaNet, and Faster R-CNN across all metrics. It also delivers real-time inference at 47 FPS, confirming its suitability for on-field deployment. An ablation study further demonstrates the effectiveness of SE blocks and extended feature hierarchies in enhancing detection accuracy. These outcomes highlight EF-SSD’s potential as a reliable, efficient, and scalable tool for smart agriculture and early crop disease management.
Why it matches plant phenotyping methodsジャガイモ葉の病斑を画像から検出・局在化し、病害の重症度を評価する深層学習手法を開発・検証しており、植物表現型取得が中心である。
abstractwe developed an AI-based system called the Enhanced Single Shot Multibox Detector(EF-SSD)
Reproduction assets foundThe paper's Data Availability statement explicitly provides a public GitHub repository with the authors' analysis/training code and a public Google Drive link to the custom 2,500-image potato leaf disease dataset with Pascal VOC annotations used in this study.Code · publicThe code implemented in this study is openly available at the following GitHub repository: https://github.com/bhavanisravan/potato-leaf-diseases-code.Open asset ↗bhavanisravan/potato-leaf-diseases-codehtml-lines:414-439Dataset · publicThe dataset used for training and evaluation can be accessed at: https://drive.google.com/drive/folders/1Yin9zp0gQKwqJ0LD3GWdLq_V_idLO2bG?usp=sharing.Open asset ↗html-lines:414-439Code / dataset availability confirmedEurope PMC · OpenAlex · checked 5 Sept 2026
Estimating canopy structure - leaf inclination distribution (LIDFa), leaf area index (LAI), and fractional vegetation cover (FCover) - is vital for breeding, yet the added value of multi-angular UAV sensing over nadir-only baselines remains insufficiently quantified. This study developed a UAV-based multi-angular inversion framework that derived high-resolution bidirectional reflectance factors (BRF) from oblique photogrammetry and fitted a kernel-driven BRDF model to characterize reflectance anisotropy. Using transfer learning across cultivars and dates, we compared the retrieval performance of multi-angle versus nadir-only baselines for LIDFa, LAI, and FCover. BRDF model simulations agreed well with airborne BRF (optimal R 2 > 0.80, RRMSE R 2 = 0.59 vs. 0.38 for the best MA and NAD models, respectively) and LIDFa ( R 2 = 0.46 vs. 0.37). For FCover, both configurations achieved high accuracy ( R 2 ≥ 0.73), with MA models providing marginal gains ( R 2 = 0.75). Methodologically, CNN-based transfer learning proved most effective for LAI and FCover, while a Random Forest model using raw multi-angle spectra yielded the best results for LIDFa. Optimal viewing configurations were trait-dependent, generally favoring forward scattering directions with zenith angles between 15° and 45°. These results indicate that kernel-driven BRDF modeling effectively captures spectral anisotropy in dense wheat canopies, and that multi-angular observations provide a distinct advantage for retrieving structural parameters with complex scattering behaviors, such as LAI and LIDFa.
Why it matches plant phenotyping methods小麦育種材料のキャノピー構造形質を対象に、UAVマルチアングルセンシング、BRDFモデル、CNN/RFによる推定フレームワークを開発・比較しており、形質取得手法が研究の中心である。
abstractThis study developed a UAV-based multi-angular inversion framework that derived high-resolution bidirectional reflectance factors (BRF) from oblique photogrammetry and fitted a kernel-driven BRDF model to characterize reflectance anisotropy.
Reproduction assets foundThe paper's data availability statement explicitly deposits the complete source code for BRDF modeling and the transfer learning pipeline, plus a subset of preprocessed field data, in a public GitHub repository matching an allowed URL. Additional data are available only on request.Code · publicThe complete source code for BRDF modeling and the transfer learning pipeline, along with a subset of the preprocessed field data used in this study, are openly available in the GitHub repository at https://github.com/ZWM-RS/UAV-multi-angle-inversion-of-canopy-structure-parameters-in-wheat-breeding-materials.git . Any additional data supporting the findings of this study are available from the corresponding author upon reasonable request.Open asset ↗ZWM-RS/UAV-multi-angle-inversion-of-canopy-structure-parameters-in-wheat-breeding-materialslines:451-474Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Medicinal plants including Ocimum tenuiflorum L. (Tulsi), Azadirachta indica A. Juss. (Neem), and Kalanchoe pinnata (Lam.) Pers. (Patharkuchi) are essential sources of bioactive compounds, yet leaf diseases threaten their yield and phytochemical integrity. This study proposes LSeTNet, a lightweight hybrid CNN (Convolutional Neural Network) Transformer architecture with Squeeze-and-Excitation (SE) blocks, achieving 99.72% accuracy, 1.00 macro F1-score, and AUC = 1.00 across 12 disease classes (1,000 images/class post-augmentation) using only 9.38 M parameters and 2.50 GFLOPs. Five-fold cross-validation yielded 99.74% ± 0.14% accuracy, with rapid convergence and no overfitting. Explainable Artificial Intelligence (XAI) via Gradient-weighted Class Activation Mapping (Grad-CAM) (mean intensity: 0.1664-0.2702), Local Interpretable Model-agnostic Explanations (LIME), and t-distributed Stochastic Neighbor Embedding (t-SNE) (silhouette score: 0.87) confirmed biologically meaningful attention on pathological regions. External validation on the independent BD-MediLeaves dataset (8 classes, 8,000 samples) achieved 99.42% accuracy and 0.99 macro F1. With 6.98 ms/image inference latency and 35.81 MB memory, LSeTNet enables real-time, edge-based deployment. It significantly outperforms DenseNet169 (95.56%), ViT-B16 (95.61%), and LW-CNN+SE (95.39%) ([Formula: see text], paired t-tests), establishing a transparent, efficient, and generalizable benchmark for precision phytopathology and sustainable medicinal plant cultivation.
Why it matches plant phenotyping methods植物葉の病害状態を画像から分類するCNN・Transformer手法を開発し、交差検証、外部データセット、既存モデルとの比較で検証しており、植物フェノタイピング手法が中心である。
abstractThis study proposes LSeTNet, a lightweight hybrid CNN (Convolutional Neural Network) Transformer architecture with Squeeze-and-Excitation (SE) blocks
Reproduction assets foundThe paper publicly releases its primary medicinal leaf image dataset (MedicinalLeaf-12) on Mendeley Data, uses a public external validation dataset (BD-MediLeaves, also on Mendeley), and provides full training/evaluation code for LSeTNet on GitHub. All three are paper-specific, public, and actionable.Dataset · publicThe primary dataset used in this study is available in the Mendeley Data Repository: https://data.mendeley.com/datasets/ncg7kk3gwx/1 .Open asset ↗Mendeley Data · ncg7kk3gwx/1lines:712-750Code · publicThe full training and evaluation code for the proposed LSeTNet model is publicly available on GitHub at: https://github.com/mdtuhinkhan101/LSeTNet .Open asset ↗GitHub · mdtuhinkhan101/LSeTNetlines:712-750Code / dataset availability confirmedCrossref · checked 13 Sept 2026
Field / plotLiDAR / point cloudRGB / grayscaleLeafStem / branchWhole plant / canopy / plot / fieldClassificationSegmentation
Abstract Annotated datasets are essential for training and evaluating machine learning models in forest ecology. This dataset provides high-resolution, annotated LiDAR point clouds of 674 individual trees from 12 forest plots in the Shivalik Range of northern Haryana, India, representing 24 species. Data were acquired using Terrestrial Laser Scanning (TLS) and Airborne Laser Scanning (ALS), include field-measured attributes such as species identity and Diameter at Breast Height (DBH), and terrestrial and aerial RGB imagery. TLS point clouds were georeferenced and co-registered with centimetre-level accuracy, enabling precise integration with ALS data. The dataset includes segmented individual trees and wood–leaf classifications, suitable for applications such as tree morphology analysis, biomass estimation, and species classification. To support benchmarking, outputs from established classification algorithms (LeWoS, TLSeparation, CANUPO, and Random Forest) are included. As one of the first open-access LiDAR datasets from Indian tropical forests, it provides critical reference data for developing and validating forest structure models. It can also aid biomass mapping efforts in support of large-scale missions such as NASA-ISRO’s NISAR and ESA’s BIOMASS.
Why it matches plant phenotyping methods個体樹木のLiDAR点群・RGB画像と樹木セグメンテーションを含む公開データセットで、樹形解析や森林構造モデルの開発・検証、分類アルゴリズムのベンチマークを目的としており、植物形質取得が中心です。
abstractThis dataset provides high-resolution, annotated LiDAR point clouds of 674 individual trees from 12 forest plots in the Shivalik Range of northern Haryana, India, representing 24 species.
Reproduction assets foundThe paper's authors explicitly state that all code used for data processing, wood-leaf classification, feature extraction, and tree volume estimation is openly available on GitHub at https://github.com/moonis-ali/Dataset, which is an allowed URL. The paper's core LiDAR dataset is deposited on Zenodo (10.5281/zenodo.153Code · publicAll code used for data processing, wood-leaf classification, feature extraction, and tree volume estimation is openly available on GitHub at https://github.com/moonis-ali/Dataset .Open asset ↗https://github.com/moonis-ali/Datasetlines:479-553Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Ensuring the post-harvest quality and health of leafy vegetables is critical for minimizing economic loss, enhancing food security, and promoting sustainable agricultural practices. Spinach, a highly nutritious yet perishable crop, is particularly susceptible to rapid freshness degradation and foliar diseases. While computer vision and deep learning have shown promise for automated quality assessment, existing models often lack the robustness to handle the dual-task classification of both freshness and disease states across diverse local spinach varieties. To bridge this gap, this paper introduces a novel hybrid Convolutional Neural Network (CNN) architecture specifically designed for the multi-class detection of freshness and visual disease symptoms in local spinach leaves. The proposed model synergistically integrates a powerful feature extraction backbone with a tailored attention and fusion mechanism, enhancing its ability to capture discriminative spatial and textural features critical for fine-grained classification. It was trained and validated on a curated dataset comprising high-resolution images of three prominent local varieties (Malabar, Water, and Red spinach) in both fresh and non-fresh conditions. The proposed hybrid model achieved a classification accuracy of 98.36%, significantly outperforming benchmark state-of-the-art models including DenseNet121, ResNet50, and EfficientNetB0. Furthermore, explainable AI (XAI) techniques visually validated the model’s decision-making process, confirming its focus on biologically relevant leaf regions. The results demonstrate that the proposed hybrid framework offers a highly accurate, reliable, and interpretable tool for non-destructive, real-time quality monitoring. This work provides a significant contribution towards intelligent post-harvest management systems, capable of reducing waste and supporting the value chain for local spinach cultivation.
Why it matches plant phenotyping methods葉画像から鮮度および視覚的な病徴を分類するCNN手法の開発・検証が中心であり、植物の状態を直接推定している。
abstractthis paper introduces a novel hybrid Convolutional Neural Network (CNN) architecture specifically designed for the multi-class detection of freshness and visual disease symptoms in local spinach leaves.
Reproduction assets foundThe paper's Data Availability statement points to a public Mendeley Data deposit of the local spinach leaf image dataset used for the CNN freshness/disease classification, matching the paper's phenotyping inputs.Dataset · publicThe datasets analyzed during the current study are publicly available in the Mendeley Data repository at: [https://data.mendeley.com/datasets/skf6w2s2h2/2](https:/data.mendeley.com/datasets/skf6w2s2h2/2).Open asset ↗Mendeley Datahtml-lines:660-689Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Background: Plant disease detection remains a major challenge in agriculture, with direct implications for improving crop productivity and ensuring food security. Seasonal variation significantly influences plant characteristics, making the classification of plant leaves by season-specifically summer and winter-important for optimizing disease detection and management strategies. Methods: In this study, a plant leaf disease detection dataset was developed and categorized based on seasonal conditions. The dataset includes 47 classes representing summer crops and 16 classes for winter crops. To classify plant leaf diseases effectively, we propose a novel dual-encoder Variational Autoencoder (VAE) model that integrates ResNet and VGGNet as parallel encoders. These encoders extract complementary feature maps from the seasonal datasets, which are then concatenated to improve classification accuracy. Result: Experimental evaluation demonstrates the robustness and accuracy of the proposed approach. The dual-encoder VAE achieved a classification accuracy of 98.86% on the summer dataset and 97.53% on the winter dataset, highlighting the model’s ability to generalize effectively across seasonal variations in plant leaf disease detection.
Why it matches plant phenotyping methods植物葉の病徴を画像から検出・分類する新規VAE手法を開発し、季節別データセットで精度評価しているため、病害状態の表現型取得が中心である。
abstractTo classify plant leaf diseases effectively, we propose a novel dual-encoder Variational Autoencoder (VAE) model that integrates ResNet and VGGNet as parallel encoders.
Reproduction assets foundThe paper's plant leaf disease image inputs are publicly available Kaggle datasets explicitly cited as the sources for the summer and winter subsets (Sankalana plant-diseases-training-dataset, Gadde yellow vein mosaic, Kapadnis watermelon, Mir pumpkin). No authors' analysis code, trained model checkpoints, or paper-derDataset · public, G., Rathod, N., Pooja, S., Huligol,
S.N., Channakeshava, R. and Vijaykumar, K.N. (2025).
Artificial intelligence based precise disease detection in
soybean using real-time object detectors. Legume
Research. 48(11): 1878-1883. doi: 10.18805/LR-5431.
Kapadnis, S. (n.d.). Watermelon Disease Recognition Dataset
[Dataset]. Kaggle. https://www.kaggle.com/datasets/sujaykapadnis/watermelon-disease-recognition-dataset.Kashyap, N. and Kashyap, A.K. (2025). Deep learning VGG19
model for precise plant disease detection. Agricultural
Science Digest. 1-8. doi: 10.18805/ag.D-6220.
Dual-encoder Variational Autoencoder for Detection and Classification of Plant Leaf Diseases
Fig 5: Confusion matrix obtainedOpen asset ↗Kaggle · Watermelon Disease Recognition Datasetpdf-raw-page:6 lines:1-85Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Accurate and timely diagnosis of rice leaf diseases is crucial for smart agriculture leveraging vision sensors. However, existing lightweight convolutional neural networks (CNNs) often struggle in complex field environments, where small lesions, cluttered backgrounds, and varying illumination complicate recognition. This paper presents I-GhostNetV3, an incrementally improved GhostNetV3-based network for RGB rice leaf disease recognition. I-GhostNetV3 introduces two modular enhancements with controlled overhead: (1) Adaptive Parallel Attention (APA), which integrates edge-guided spatial and channel cues and is selectively inserted to enhance lesion-related representations (at the cost of additional computation), and (2) Fusion Coordinate-Channel Attention (FCCA), a near-neutral SE replacement that enables efficient spatial-channel feature fusion to suppress background interference. Experiments on the Rice Leaf Bacterial and Fungal Disease (RLBF) dataset show that I-GhostNetV3 achieves 90.02% Top-1 accuracy with 1.831 million parameters and 248.694 million FLOPs, outperforming MobileNetV2 and EfficientNet-B0 under our experimental setup while remaining compact relative to the original GhostNetV3. In addition, evaluation on PlantVillage-Corn serves as a supplementary transfer sanity check; further validation on independent real-field target domains and on-device profiling will be explored in future work. These results indicate that I-GhostNetV3 is a promising efficient backbone for future edge deployment in precision agriculture.
Why it matches plant phenotyping methods画像からイネ葉の病徴を認識・分類する軽量深層学習手法を開発し、複数データセットで精度と計算量を評価しているため、植物フェノタイピング手法が中心である。
abstractThis paper presents I-GhostNetV3, an incrementally improved GhostNetV3-based network for RGB rice leaf disease recognition.
Reproduction assets foundThe paper's phenotyping inputs are two publicly available plant image datasets explicitly linked by the authors: the RLBF rice leaf disease dataset on Mendeley Data (primary evaluation) and the PlantVillage-Corn dataset on GitHub (cross-domain transfer). No author analysis code or trained model checkpoints are stated.Dataset · publicThe Rice Leaf Bacterial and Fungal Disease Dataset can be accessed at https://data.mendeley.com/datasets/hx6f852hw4/2 (accessed on 20 July 2025)Open asset ↗hx6f852hw4lines:688-704Dataset · publicthe PlantVillage-Corn Dataset is available at https://github.com/gabrieldgf4/PlantVillage-Dataset (accessed on 27 August 2025)Open asset ↗github.com/gabrieldgf4/PlantVillage-Datasetlines:688-704Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Field / plotLeafObject detectionStress / disease detectionDisease symptoms / severity
Timely localization and diagnosis of crop lesions are critical for disease control and reducing pesticide use. However, in-field lesions often resemble leaf textures, vary widely in scale, and suffer from lighting and shadow interference-making simultaneous high accuracy and lightweight inference challenging. We propose WGA-YOLO, a lightweight YOLO variant for crop disease recognition. Central to our design is Wavelet Channel Recalibration (WCR), a DWT-based downsampling module: discrete wavelet transform naturally provides multi-resolution, time-frequency localized representations that explicitly separate low-frequency approximations from high-frequency edge/texture details. WCR fuses high- and low-frequency components and enhances feature representation through their frequency-domain complementarity, thereby preserving semantic and fine texture information during resolution reduction with negligible extra cost. We also introduce PS-C2f, which integrates Pinwheel-shaped convolutions into C2f to better capture tiny lesion details via multi-directional, irregular kernels, and replace SPPF with Dynamic Group Attention Pooling (DGAP) for efficient multi-scale context aggregation. On our PlantDoc_boost dataset, WGA-YOLO improves over YOLOv8n by 3.02 and 2.85% points, while reducing parameters and FLOPs by ~ 0.18 M and ~ 0.3G, demonstrating improved inference efficiency and deployment friendliness while maintaining strong detection performance in field scenarios.
Why it matches plant phenotyping methods植物葉の病斑を画像から検出・診断するYOLO改良手法の開発が中心であり、病害状態の画像ベース表現型計測に該当する。
abstractWe propose WGA-YOLO, a lightweight YOLO variant for crop disease recognition.
Reproduction assets foundThe paper's PlantDoc_boost dataset (the annotated crop-disease image dataset constructed and analyzed in this study) is explicitly stated to be publicly released on the authors' GitHub repository. No author analysis code or trained model checkpoints are stated as available. The Roboflow corn and tomato datasets are preDataset · publicThe PlantDoc_boost dataset used and analyzed in this study is publicly available from the project repository at http://github.com/YongChaoLiang/PlantDoc_boost/tree/master.Open asset ↗YongChaoLiang/PlantDoc_boosthtml-lines:693-708Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
The rapid and precise identification of apple leaf diseases is crucial for minimizing yield loss in precision agriculture. However, many existing deep learning methods struggle to be applicable in real-world settings, are not easily interpretable, and often lack sufficient statistical validation. To address these difficulties, we propose our solution approach LeafSightX . This dual-backbone architecture combines features from DenseNet201 and InceptionV3 using Multi-Head Self-Attention (MHSA) techniques, enhancing representational capability and spatial context reasoning. Our extensive procedure includes specialized preprocessing and limited data augmentation, improving model resilience in many scenarios. Furthermore, LeafSightX integrates explainable AI techniques with Grad-CAM visualizations to improve transparency. In assessments of a five-class apple leaf disease dataset featuring field and laboratory images, LeafSightX demonstrates exceptional performance, attaining a test accuracy of 99.64%, an F1-score of 0.9962, and AUC and PR-AUC scores of 1.000, far surpassing all baseline CNNs. Cross-validated Cohen's Kappa (mean = 0.9917, σ = 0.0020) and AUC (mean = 0.9998) indicate a significant level of predictive consistency. Despite its architectural complexity, the model offers real-time inference capabilities, ensuring per-sample latency suitable for edge device deployment. Additionally, the proposed LeafSightX framework was trained and evaluated on an additional independent apple leaf disease dataset, achieving a test accuracy of 99.69%, demonstrating its robustness and generalization. Our approach is a rigorously evaluated, clear, and highly accurate system for identifying plant diseases, providing a reproducible foundation for the actual application of AI in agriculture.
Why it matches plant phenotyping methodsリンゴ葉の病害状態を画像から識別するCNN手法を開発し、複数データセット・交差検証・ベースライン比較で性能を評価しており、植物病害表現型の取得・推定が中心である。
Reproduction assets foundThe paper uses two public Kaggle apple leaf disease image datasets as its phenotyping inputs; both are directly cited with public URLs. No author analysis code, trained model checkpoints, or supplementary code repository is deposited — the data availability statement only offers contact with corresponding authors.Dataset · publicThis research utilizes the Apple Tree Leaf Disease dataset, collected from Kaggle and made available by Nirmal (Kaggle, 2025).Open asset ↗Kagglehtml-lines:128-184Dataset · publicDhar S. (2023). Apple leaf disease classification dataset. Available online at: https://www.kaggle.com/datasets/showravdhar/apple-disease-dataset (Accessed November 1, 2023).Open asset ↗Kaggle · showravdhar/apple-disease-datasethtml-lines:1449-1484Code / dataset availability confirmedEurope PMC · bioRxiv · checked 15 Sept 2026
The leaf blade and vasculature develop together within a shared morphological space. Despite shared molecular patterning pathways, it is unknown if developmental and evolutionary variation affect these tissues separately or together in a coordinated way. Grapevine leaves have a morphometric history and abundant data measuring the shape of the blade and vasculature together. Using a combination of topological data analysis and deep learning, we perform reciprocal semantic segmentation of leaf blade and vasculature. Each tissue contains sufficient information to predict the other. We hypothesize that this is due to a one-to-one relationship between blade and vein. Using thin plate splines to swap and warp different combinations of blade and vein shapes, we show that a set of leaves with a many-to-one relationship of blade and vein are distinguishable from true leaves. We also swap blade and vein across the developmental series and between species and show that only reversing the developmental series disrupts the relationship between blade and vasculature. We end by discussing the evolutionary and developmental implications that there is a unique, one-to-one mapping between blade and vein that allows each to be predicted from the other. Author summary Leaves are made of two closely connected parts: the flat blade that captures light and the network of veins that transports water, nutrients, and developmental signals. Although these tissues grow together and share common molecular patterning pathways, it has remained unclear whether a particular blade shape is uniquely linked to a specific vein pattern. In this study, we use grapevine leaves as a model system and combine mathematical shape analysis with deep learning to examine this relationship. We show that the shape of the blade alone can accurately predict the vein network, and that the vein network can likewise predict the blade. This finding suggests a near one-to-one relationship between these two tissues. To test this idea, we created artificial leaves in which blade and vein shapes were deliberately mismatched. Although these synthetic leaves appeared realistic at a global level, a neural network was able to distinguish them from real leaves based on subtle differences. We further show that this tight coupling is maintained by the developmental sequence of leaf growth rather than by species identity, revealing a conserved constraint linking leaf form and internal structure.
Why it matches plant phenotyping methods葉身と葉脈の形状を深層学習による相互セグメンテーションと形状解析で抽出・予測する方法が研究の中心であり、植物形態表現型の方法開発に該当する。
abstractUsing a combination of topological data analysis and deep learning, we perform reciprocal semantic segmentation of leaf blade and vasculature.
Reproduction assets foundThe Data Availability Statement explicitly lists three public Zenodo deposits containing data and code to reproduce the paper's analyses: reciprocal U-Net blade/vein prediction, the two-tower CNN 1:1 vein:blade relationship, and interspecies/intraspecies developmental series swaps. All URLs are in allowed_urls and the Code · public393
which the relationship between blade and vasculature is conserved or diversified across the
394
spectacular variety of leaf shapes remains to be seen.
395
Data Availability Statement
396
Data and code to reproduce this work can be found for the following analyses: Reciprocal
397
prediction of vein and blade from the other, https://zenodo.org/records/16920155; Two tower CNN
398
1:1 vein:blade relationship, https://zenodo.org/records/17014105; Interspecies and Intraspecies
399
developmental series swaps, https://zenodo.org/records/17013783
400
Conflict of Interest Statement
401
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the aOpen asset ↗zenodo · 16920155pdf-raw-page:22 lines:1-53Code · publicd across the
394
spectacular variety of leaf shapes remains to be seen.
395
Data Availability Statement
396
Data and code to reproduce this work can be found for the following analyses: Reciprocal
397
prediction of vein and blade from the other, https://zenodo.org/records/16920155; Two tower CNN
398
1:1 vein:blade relationship, https://zenodo.org/records/17014105; Interspecies and Intraspecies
399
developmental series swaps, https://zenodo.org/records/17013783
400
Conflict of Interest Statement
401
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuitOpen asset ↗zenodo · 17014105pdf-raw-page:22 lines:1-53Code · publicment
396
Data and code to reproduce this work can be found for the following analyses: Reciprocal
397
prediction of vein and blade from the other, https://zenodo.org/records/16920155; Two tower CNN
398
1:1 vein:blade relationship, https://zenodo.org/records/17014105; Interspecies and Intraspecies
399
developmental series swaps, https://zenodo.org/records/17013783
400
Conflict of Interest Statement
401
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted January 29, 2026.
;
https:Open asset ↗zenodo · 17013783pdf-raw-page:22 lines:1-53Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
This article introduces Agri-Vision Bangladesh, a comprehensive, augmented image dataset designed to advance automated disease diagnosis in four economically vital agricultural crops: Bottle Gourd ( Lagenaria siceraria ), Zucchini ( Cucurbita pepo ), Papaya (Carica papaya), and Tomato ( Solanum lycopersicum ). Addressing the scarcity of region-specific agricultural data, a total of 5266 original images were acquired directly from diverse agricultural fields in Bangladesh using a SONY ALPHA 7 II full-frame camera under natural lighting conditions. The dataset encompasses 28 distinct classes, covering a wide spectrum of biotic stressors including viral (Mosaic Virus, Leaf Curl), fungal (Downy Mildew, Anthracnose, Alternaria Blight), bacterial (Bacterial Blight, Xanthomonas), and pest-induced damage (Insect Hole, White Spot), alongside Healthy samples. To ensure scientific reliability, each image underwent a rigorous two-stage validation process by senior agronomists. To tackle class imbalance and facilitate the training of data-intensive Deep Learning models, the dataset was expanded using a Python-based augmentation pipeline incorporating geometric transformations (rotation, flipping) and photometric adjustments (noise, brightness) resulting in a final repository of 28,000 images (5266 original and 22,734 augmented). All files are standardized to 512×512 pixels in JPG format. This expert-validated resource serves as a critical benchmark for developing robust computer vision algorithms (e.g., CNNs, Vision Transformers) for precision agriculture, enabling research into fine-grained classification, object detection, and cross-crop transfer learning in subtropical farming environments.
Why it matches plant phenotyping methods植物病害症状を画像で分類するための専門家検証済みデータセットを構築し、再利用可能なベンチマークとして提供しているため、植物表現型取得法が中心です。
abstractThis article introduces Agri-Vision Bangladesh, a comprehensive, augmented image dataset designed to advance automated disease diagnosis
Reproduction assets foundThe paper is a Data in Brief article describing the Agri-Vision Bangladesh multi-crop leaf disease image dataset, publicly deposited on Mendeley Data with an explicit direct URL and DOI (10.17632/8t6k37ztxc.2). This is a paper-specific public asset containing the original and augmented plant images used in the study.Dataset · publicRepository name: Mendeley Data
Data identification number: 10.17632/8t6k37ztxc.2
Direct URL to data: https://data.mendeley.com/preview/8t6k37ztxc?a=a88a48f1-a9b0-4354-a081-cc8f1e936364Open asset ↗Mendeley Data · 10.17632/8t6k37ztxc.2html-lines:93-117Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
In real-world agriculture, healthy plant leaves are significantly more common than diseased ones. This natural class imbalance presents challenges in automated plant disease detection, as analyzing each leaf with computationally intensive deep-learning models is problematic, leading to inefficiency and increased resource consumption. To tackle this challenge and promote sustainable AI solutions, this study presents an iterative, hybrid AI approach that boosts computational efficiency, interpretability, and scalability for real-time disease detection. This hybrid system operates in two stages: first, a lightweight traditional machine learning classifier performs binary classification to quickly separate and exclude healthy leaves, followed by a deep learning model (ResNet, DenseNet, MobileNet, and EfficientNet) that classifies the specific disease in the smaller group of diseased leaves. This two-stage method minimizes computational load while maintaining high classification accuracy. Additionally, this study uses Explainable AI (XAI) methods, particularly Gradient-weighted Class Activation Mapping (Grad-CAM), to generate heatmaps. These heatmaps highlight the image areas that most significantly influence the model’s predictions, thereby improving transparency and refining the feature extraction process. The proposed hybrid model, comprising Logistic Regression and Mobilenetv3, offers up to 77.6% faster inference than conventional deep learning models with only about 3% accuracy loss. For a large-scale test of 1,227 images on an entry-level laptop, the hybrid model reduced the total inference time from 4,548 seconds to just 1,010.13 seconds, with minimal CPU load. By addressing class imbalance, optimizing inference efficiency, and incorporating explainable AI, this work contributes a scalable, sustainable, and trustworthy solution for plant disease detection in precision agriculture.
Why it matches plant phenotyping methods葉画像から植物病害状態を推定する二段階AI手法とXAIを開発・評価しており、病害の表現型取得が研究の中心である。
abstractThis two-stage method minimizes computational load while maintaining high classification accuracy.
Reproduction assets foundThe paper uses the public MangoLeafBD dataset (Mendeley Data) and the authors state they made their complete source code, trained model weights, and preprocessing scripts publicly available on GitHub. Both are paper-specific, public, and actionable.Code · publicwe have made the complete source code, trained model weights (for MobileNet and Random Forest), and preprocessing scripts publicly available on GitHub. The repository is accessible at:
3. GitHub - abkafi1234/Disease_Agnostic_Hybrid_classifierOpen asset ↗abkafi1234/Disease_Agnostic_Hybrid_classifierlines:832-873Code / dataset availability confirmedCrossref · Europe PMC · checked 14 Sept 2026
Abstract Time-resolved phenotyping of disease symptoms enables dissection of resistance mechanisms and improves diagnosis, but acquiring phenotypic data at satisfactory scale remains challenging. Advances in imaging and image processing have improved measurement precision, robustness, and throughput, but further improvements are needed for practical application. We present a data set comprising 12,520 high-resolution (~0.03 mm/pixel) RGB images representing 1,032 time series of wheat leaves with developing disease symptoms. All images are geometrically aligned with a median precision of 0.16 mm (≈5 pixels). The dataset includes transformation matrices, symptom segmentation masks, metadata on treatments, weather, crop phenology, and disease occurrence, and a lightweight Python toolkit for loading, aligning, inspecting, and editing image sequences. These resources enable detailed investigation of leaf-level disease dynamics such as lesion, pustule, and fruiting body emergence rates, lesion growth, and dynamic interactions of disease development with spatial and environmental contexts. They offer a broad basis for developing improved methods for image alignment and symptom detection, segmentation, and tracking, possibly by tackling these connected challenges within a single end-to-end framework.
Why it matches plant phenotyping methods葉の病徴を対象とした高解像度時系列画像データセットで、幾何位置合わせ、病徴セグメンテーション、追跡用ツールを提供しており、植物病害表現型の取得・解析基盤が中心である。
abstractWe present a data set comprising 12,520 high-resolution (~0.03 mm/pixel) RGB images representing 1,032 time series of wheat leaves with developing disease symptoms.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicWe provide a lightweight Python toolkit to facilitate loading, inspection, and curation of the image sequences and their associated processing products in the associated Git repository (https://github.com/and-jonas/sympathique-wheat).Open asset ↗github.com/and-jonas/sympathique-wheathtml-lines:317-337Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
Introduction: Accurately segmenting cotton seedling organs from 3D point clouds is fundamental for high-throughput plant phenotyping and digital breeding. However, cotton seedling segmentation remains challenging due to fine-scale and complex organ morphology, uneven point density with noise, and the lack of high-quality annotated datasets. Methods: To address these issues, we propose DCSFormer, a tailored extension of Point Transformer V3 designed for cotton seedling point cloud segmentation. The model introduces the DCS Block, which leverages dynamic sparse expert routing and dual-channel attention to adaptively capture global semantic dependencies and subtle local geometric variations, thereby improving stem-leaf boundary discrimination. In addition, the proposed CLFSkip replaces traditional skip connections with a cross-layer fusion strategy, effectively integrating multi-scale features while preserving organ-level details. We also constructed an annotated cotton seedling dataset to support training and evaluation. Results and Discussion: Experimental results show that DCSFormer achieves 93.67% mIoU, 95.83% mPrec, 97.35% mRec, and 96.56% mF1, outperforming multiple comparison models. Furthermore, when evaluated against baseline models on two public datasets, Crops3D and Pheno4D, DCSFormer exceeds the baseline across all four metrics, further validating its effectiveness and generalizability. This work provides an effective solution for precise cotton seedling organ segmentation.
Why it matches plant phenotyping methods綿花幼苗の3D点群から器官を抽出する手法を開発し、アノテーション済みデータセットの構築と複数データセットでの性能検証を行っており、植物表現型取得が中心である。
abstractAccurately segmenting cotton seedling organs from 3D point clouds is fundamental for high-throughput plant phenotyping and digital breeding.
Reproduction assets foundThe authors constructed an annotated cotton seedling point cloud dataset (100 samples with semantic/instance organ labels and ground-truth traits) used for training and evaluating DCSFormer, and the data availability statement points to a public Kaggle repository containing it. No author analysis code or trained model/Dataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.kaggle.com/datasets/tengfeiliu333/dcsformer-cotton/croissant/download .Open asset ↗Kaggle · tengfeiliu333/dcsformer-cottonlines:957-1015Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Hyperspectral reflectance provides rapid, non-destructive phenotyping of plant leaves. These data have been used to develop machine learning models for predicting diverse plant traits, yet key challenges remain. We collected hyperspectral reflectance data together with 25 anatomical, gas exchange, and chlorophyll fluorescence traits from 320 recombinant inbred lines grown over three seasons. Using these data, we systematically (1) compare the performance of PLSR and SVR across a wide range of traits, including also slow fluorescence kinetics, (2) assess model generalizability and transferability, and (3) investigate how different aggregation strategies affect predictive accuracy. Based on a nested cross-validation framework, single cross-validation with MSE as metric performed comparably to repeated cross-validation or PRESS-based calibration. Optimal performance of trait-specific predictions was found to be dependent on the combination of model and data aggregation levels. Structural and biochemical traits showed the best generalizability and transferability, whereas physiological traits, particularly those derived from gas exchange and fluorescence kinetics, exhibited markedly reduced transferability. Together, these results provide a rigorous benchmark for evaluating machine learning models for trait prediction from hyperspectral reflectance data, and highlight both the opportunities and limitations for achieving robust generalization across diverse environments and genotypes.
Why it matches plant phenotyping methodsハイパースペクトル反射データから植物形質を予測する機械学習手法を、複数形質・環境・遺伝子型で系統的に比較し、一般化性と転移性を厳密にベンチマークしているため、方法論が中心である。
abstractHyperspectral reflectance provides rapid, non-destructive phenotyping of plant leaves.
Reproduction assets foundThe paper's Data availability statement explicitly deposits all code and raw hyperspectral/trait data in a public GitHub repository, matching an allowed URL.Code · publicAll code and raw data to ensure reproducibility of the results can be accessed at: [https://github.com/Rudan-X/HyperspectralML](https:/github.com/Rudan-X/HyperspectralML).Open asset ↗Rudan-X/HyperspectralMLlines:158-246Code / dataset availability confirmedEurope PMC · Crossref · checked 14 Sept 2026
Abstract Mango diseases and pest infestations represent a major challenge to agricultural productivity, making early and accurate diagnosis crucial for reducing crop losses. This study presents a security-preserving ensemble convolutional neural network (CNN) framework for the automated identification and classification of mango leaf diseases using image-based analysis. The proposed system is designed to work with images captured under real field conditions, ensuring its suitability for practical agricultural applications. The dataset includes mango leaf images affected by various diseases and pests such as Gall Midge, Powdery Mildew, Sooty Mould, Die Back, Cutting Weevil, and Anthracnose, each characterized by distinct visual symptoms including discoloration, necrotic spots, fungal growth, leaf deformation, and edge damage. Traditional manual diagnosis of these conditions is often time-consuming, labor-intensive, and susceptible to human error. To overcome these limitations, the proposed framework employs an ensemble of transfer-learning-based CNN models to extract meaningful features related to texture, color distribution, shape, and lesion patterns. A security-preserving learning mechanism is integrated to ensure the safe handling of agricultural image data, minimizing data exposure risks while maintaining high model performance. Additionally, data augmentation techniques are utilized to improve model robustness, reduce overfitting, and address class imbalance commonly found in agricultural datasets. The system is capable of multi-class classification, reflecting real-world scenarios where multiple diseases may exhibit visually similar characteristics. Experimental results indicate that the ensemble CNN framework achieves high classification accuracy and demonstrates strong generalization across varying lighting conditions and complex backgrounds. By effectively capturing disease-specific visual features, the proposed approach enhances detection reliability in real-world field environments. Overall, this system offers a scalable, non-invasive, and security-aware solution for early mango leaf disease detection, contributing to precision agriculture and informed decision-making. The findings highlight the potential of deep learning and computer vision technologies in developing intelligent, secure, and efficient plant health monitoring systems.
Why it matches plant phenotyping methodsマンゴー葉の病徴を画像から分類するCNNフレームワークの開発が研究の中心であり、植物の病害状態を直接推定する画像ベース表現型解析に該当する。
abstractThis study presents a security-preserving ensemble convolutional neural network (CNN) framework for the automated identification and classification of mango leaf diseases using image-based analysis.
Reproduction assets foundThe paper's plant-phenotyping input is the public Kaggle Mango Leaf Disease Dataset of mango leaf images (Gall Midge, Powdery Mildew, Sooty Mould, Die Back, Cutting Weevil, Anthracnose, Healthy), explicitly declared as publicly available with a link. No author code, models, or checkpoints are shared.Dataset · publicdation.
Zahra Maryam handled data curation and resources.
Muhammad Haseeb Zia conducted the formal analysis. All
authors reviewed and approved the final manuscript for
submission.
Funding This research did not receive funding.
Data Availability The dataset used in this study is
publicly available on Kaggle. The dataset link is:
https://www.kaggle.com/datasets/aryashah2k/mango-leaf-disease-dataset.Declarations
Conflict of interest The authors declare that they
have no conflict of interest.
Ethical approval This study utilizes a publicly
available benchmark dataset from Kaggle (Mango
Leaf Disease Dataset:
https://www.kaggle.com/datasets/aryashah2k/mang
o-leaf-disease-dataset ). The dataset is Open asset ↗Kaggle · aryashah2k/mango-leaf-disease-datasetpdf-raw-page:11 lines:1-91Code / dataset availability confirmedCrossref · Europe PMC · checked 14 Sept 2026
Plant diseases remain a significant challenge in global agricultural production. Achieving efficient and accurate disease detection is essential for reducing crop losses, controlling agricultural costs, and improving yields. As agriculture rapidly advances toward digitalization and intelligent transformation, the application of artificial intelligence technologies has become a key pathway to enhancing industrial competitiveness. In this study, Chat Demeter, a multi-agent system for plant disease diagnosis based on deep learning. The system captures real-time leaf images through camera devices. It employs a CNN-Transformer model to perform instance segmentation and object detection, thereby enabling automatic identification of diseased leaves and classification of disease types. To enhance interactivity and practical value, the system incorporates a natural language interface, allowing users to upload images and receive automated diagnostic results and treatment suggestions. Experimental results demonstrate that the system achieves an accuracy of 99.50% and an AUC o f 99.91% on the validation dataset, highlighting its superior performance. Overall, Chat Demeter provides an effective tool for crop health monitoring and disease intervention, while offering a feasible pathway and developmental direction for integrating and optimizing future agricultural multi-agent systems.
Why it matches plant phenotyping methods植物葉の画像から病葉をセグメンテーション・分類する診断システムが研究の中心であり、植物病害状態の画像ベース表現型計測に該当する。
abstractChat Demeter, a multi-agent system for plant disease diagnosis based on deep learning.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe original dataset can be accessed at https://www.kaggle.com/code/anshulm257/rice-disease-detection-using-cnn , which includes four distinct datasets to ensure diversity in data sources: https://www.kaggle.com/datasets/nirmalsankalana/rice-leaf-disease-imageOpen asset ↗Kaggle · nirmalsankalana/rice-leaf-disease-imagelines:304-312Code / dataset availability confirmedEurope PMC · OpenAlex · Crossref · checked 5 Sept 2026
Background Rice plant architecture underpins yield and grain quality, yet two obstacles impede accurate field characterization in dense paddies. First, single-plant reconstruction is constrained by severe inter-plant occlusion, cluttered backgrounds, and limited viewpoints. These factors obscure culms, leaves, basal tillers, and the true physical scale of the plant. Active ranging devices are cumbersome in outdoor plots and can lose accuracy, whereas conventional passive photogrammetry performs poorly under such conditions. Second, delineating panicles within a 3D rice model is intrinsically difficult. Panicles are slender, highly branched, and visually similar to surrounding foliage, often interwoven and partially hidden. These factors result in fragmented boundaries and missing details. Direct point-cloud segmentation struggles with such discontinuous geometry and requires costly 3D annotation, whereas generic image segmentation models trained on natural scenes transfer poorly to paddy imagery. These challenges motivate a field-ready workflow that both reconstructs whole plants at high resolution in dense plantings and reliably segments panicles to enable trait extraction. Results A low-cost, in-field, multi-view pipeline for whole-plant three-dimensional reconstruction, termed One Stop 3D Target Reconstruction And segmentation (OSTRA), operates on color images with a reference-board setup. The pipeline builds detailed three-dimensional models of individual rice plants and automatically segments key organs (in this case, panicles), despite dense surrounding vegetation. When applied to 231 diverse rice landraces grown in a crowded field setting, the method produced high-fidelity plant models with clearly delineated panicle structures. From these reconstructions, three architectural traits were derived: plant height, leaf area, and panicle length. Genome-wide association analysis of the measured traits identified strong genotype-phenotype associations tagging known candidate genes. Natural variants at D2 and RFL/APO2 were associated with plant height variation, variants at FLW7 were linked to differences in leaf area, and allelic variation at AAI1 corresponded to panicle length variation. These loci are established regulators of plant growth and morphology, indicating that this three-dimensional phenotyping pipeline attains accuracy sufficient to rediscover meaningful genetic signals. Conclusions This study provides a practical tool for precise rice phenotyping even under dense field planting conditions, overcoming occlusion and structural complexity. By enabling non-destructive, field-based measurement of complete plant architecture and linking these phenotypes to specific genes, the pipeline bridges field phenomics and genomics. The integrated reconstruction and analysis framework advances the study of rice architecture and offers a general route to connect complex traits with their genetic determinants.
Why it matches plant phenotyping methods密植圃場でのイネ全体3D再構築、器官分割、形質抽出を中核とする画像ベース表現型解析手法の開発・実証であり、明確に収載対象。
abstractA low-cost, in-field, multi-view pipeline for whole-plant three-dimensional reconstruction, termed One Stop 3D Target Reconstruction And segmentation (OSTRA), operates on color images with a reference-board setup.
Reproduction assets foundThe paper explicitly states that the 3D rice plant models (231 landraces) are deposited on Zenodo and the OSTRA source code is publicly available on GitHub. Both are paper-specific, public, and actionable.Code · publicThe source code of OSTRA is available on GitHub at [http://github.com/ganlab/ostra] (http:/github.com/ganlab/ostra).Open asset ↗github · ganlab/ostralines:217-246Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Abstract Agriculture is essential to human civilization, providing food and raw materials. Plant diseases significantly threaten agricultural productivity, making early and accurate detection essential. Despite Recent advances of deep learning in making automatic plant leaf diseases diagnosis systems, some of them depend on simple features fusion methods and lack an efficient method to exploit complementary information. Therefore, this paper proposes a system for diagnosing plant leaf diseases by fusing two powerful deep learning models: EfficientNetV2B0 and Swin Transformer via attention-based feature fusion that adaptively weights each model’s contribution with features. These models extract complementary features: EfficientNetV2B0 extracts fine-grained local features, and the Swin Transformer extracts global contextual information, producing highly and complementary expressive fused features. The high-dimensional fused features demand High-Performance Computing (HPC) resources for efficient parallel processing and accelerated training. Moreover, the Henry Gases Solubility Optimization (HGSO) metaheuristic is applied to select the most discriminative and related features. Unlike previous methods that diagnose diseases affecting only one plant, the proposed approach handles multiple plant species simultaneously, further increasing computational demand. Finally, RBF-kernel SVM is applied for a classification step. The system was implemented on a GPU-based high-performance computing environment using CUDA acceleration to enhance computational efficiency. Experimental evaluation on the PlantVillage benchmark dataset with seven classes achieved a high classification accuracy of 99.2%, outperforming other state-of-the-art methods. These results enhance the model’s practical capability for application in real-world agricultural decision-support systems.
Why it matches plant phenotyping methods葉画像から植物病害を診断する深層学習システムの開発とベンチマーク評価が研究の中心であり、植物の病害状態を直接推定しているため。
abstractTherefore, this paper proposes a system for diagnosing plant leaf diseases by fusing two powerful deep learning models: EfficientNetV2B0 and Swin Transformer via attention-based feature fusion
Reproduction assets foundThe paper's sole experimental input is the PlantVillage leaf-disease image dataset, which the authors explicitly state is publicly available online with a Kaggle URL in the Data availability statement. No author code, models, or other paper-specific assets are disclosed.Dataset · publicuthors have
read and agreed to the published version of the manuscript.
Funding Open access funding provided by The Science, Technology & Innovation Funding Authority
(STDF) in cooperation with The Egyptian Knowledge Bank (EKB). No funding.
Data availability The datasets used during the current research are available online at: https://www.kag-gle.com/datasets/mohitsingh1804/plantvillageOpen asset ↗kag-gle.com/datasets/mohitsingh1804/plantvillagepdf-raw-page:53 lines:1-44Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Malabar spinach is a nutrient-dense leafy vegetable widely cultivated and consumed in Bangladesh. Its productivity is often compromised by Alternaria leaf spot and straw mite infestations. This work proposes an efficient and interpretable deep learning framework for automatic Malabar spinach leaf disease classification. A curated dataset of Malabar spinach images collected from Habiganj Agricultural University and supplemented with public samples was categorized into three classes: Alternaria, straw mite, and healthy leaves. A lightweight SpinachCNN established a strong baseline, while Spinach-ResSENet, enhanced with squeeze-and-excitation modules, improved channel-wise attention and feature discrimination. A customized Vision Transformer (SpinachViT) and SwinV2-Base were further investigated to assess the benefits of transformer-based architectures under limited data. To mitigate annotation scarcity, we employed SimSiam-based self-supervised pretraining on unlabeled images, followed by supervised fine-tuning with cross-entropy or a hybrid objective combining cross-entropy and supervised contrastive loss. The best-performing domain-optimized model, SimSiam-CBAM-ResNet-50, incorporated Convolutional Block Attention Modules and achieved 97.31% test accuracy, 0.9983 macro ROC-AUC, and low calibration error, while maintaining robustness to Gaussian and salt-and-pepper noise. Although a SwinV2-Base benchmark pretrained on ImageNet-22k reached slightly higher accuracy (97.98%, 98.99% with test-time augmentation), its 86.9M parameters and reliance on large-scale pretraining reduce feasibility for edge deployment. In contrast, the SimSiam-CBAM model offers a more parameter-efficient and deployment-friendly solution for real-world agricultural applications. Model decisions are interpretable via Grad-CAM, Grad-CAM++, and LayerCAM, which consistently highlight biologically relevant lesion regions. The spinach dataset used in this study is publicly available on: https://huggingface.co/datasets/saifullah03/malabar_spinach_leaf_disease_dataset.
Why it matches plant phenotyping methods葉画像から病害状態を分類する深層学習手法を開発・比較し、公開データセットと解釈可能性・頑健性も評価しており、植物表現型取得が中心である。
abstractThis work proposes an efficient and interpretable deep learning framework for automatic Malabar spinach leaf disease classification.
Reproduction assets foundThe paper's Malabar spinach leaf disease image dataset (the phenotyping input used for all measurements) is explicitly stated as publicly available on Hugging Face, with the URL given in the abstract and Data Availability Statement. No author analysis code or trained model checkpoints are explicitly deposited.Dataset · publicThe spinach dataset used in this study is publicly available on: https://huggingface.co/datasets/saifullah03/malabar_spinach_leaf_disease_dataset.Open asset ↗huggingface · saifullah03/malabar_spinach_leaf_disease_datasethtml-lines:585-614Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Potato foliar diseases, particularly early and late blight, pose a serious threat to yield and food security, yet reliable visual recognition remains challenging due to cultivar heterogeneity, variable symptom expression, and acquisition noise in field-like imagery. To address these issues, we propose PotatoLeafNet, a two-stage deep learning framework that combines a fixed-sequence image-augmentation pipeline with a compact, task-optimized 11-layer convolutional neural network (CNN) using 3 × 3 kernels for robust, data-efficient classification of potato leaf conditions (Healthy, Early Blight, Late Blight). Methods We construct a dataset of 4,072 labeled potato leaf images from the PlantVillage-Potato subset and standardize all inputs to 224 × 224 RGB tensors with pixel intensities normalized to [0,1]. A balanced, fixed-order augmentation policy-comprising rotation, translation, shear, zoom, horizontal flipping, brightness adjustment, and channel jitter-is applied exclusively to the training split, increasing it to 6,000 images (2,000 per class) while keeping the validation and test sets free of synthetic samples. The second stage consists of an 11-layer CNN implemented in TensorFlow/Keras and trained with categorical cross-entropy loss and the Adam optimizer under a unified training and evaluation protocol. Performance is benchmarked against strong CNN and hybrid baselines, including ResNet-50 + VGG-16, VGG-16 + MobileNetV2, MobileNetV2, and Inception-V3. Results On the PlantVillage-Potato test set, PotatoLeafNet achieves 98.52% accuracy, 98.67% macro-precision, 99.67% macro-recall, 99.16% macro-F1, and 1.00 macro-AUC, outperforming all baseline models under identical preprocessing and training conditions. In particular, PotatoLeafNet surpasses ResNet-50 + VGG-16 (97.10% accuracy, AUC 0.98), VGG-16 + MobileNetV2 (94.80% accuracy, AUC 0.93), MobileNetV2 (93.20% accuracy, AUC 0.92), and Inception-V3 (92.50% accuracy, AUC 0.91). Short 10-epoch runs yield stable convergence (training accuracy 88.22%, validation accuracy 86.91%, test accuracy 88.15%), indicating efficient learning from the augmented distribution. Discussion The results demonstrate that explicitly coupling a fixed sequential augmentation stage with a lightweight 3×3-kernel CNN enables high tri-class accuracy, strong recall for disease classes, and improved generalization relative to deeper or fused architectures, without incurring substantial computational cost. By emphasizing disease-relevant structure while limiting overfitting, PotatoLeafNet provides a practical and resource-efficient solution for automated screening of potato leaf health in real-world agronomic settings, supporting timely and data-driven disease management.
Why it matches plant phenotyping methodsジャガイモ葉の病害状態を画像から分類するCNN手法を開発し、複数モデルとの性能比較で検証しており、植物フェノタイピング手法が研究の中心である。
abstractwe propose PotatoLeafNet, a two-stage deep learning framework
Reproduction assets foundThe paper's potato leaf disease classification is built on two publicly available Kaggle image datasets: the PlantVillage dataset (source of the PlantVillage-Potato subset) and the Potato Leaf Disease Dataset (PLD, 4,072 images across Healthy, Early Blight, Late Blight). Both are cited with public Kaggle URLs in the 3.Dataset · publicPotato Leaf Disease Dataset ( 2025 ). Kaggle dataset 2024. Available online at: https://www.kaggle.com/datasets/rizwan123456789/potato-disease-leaf-datasetpldOpen asset ↗Kaggle · rizwan123456789/potato-disease-leaf-datasetpldlines:788-867Dataset · publicPlant Village Dataset ( 2024 ). Kaggle [dataset]. Available online at: https://www.kaggle.com/datasets/mohitsingh1804/plantvillage (Accessed April 29, 2024).Open asset ↗Kaggle · mohitsingh1804/plantvillagelines:788-867Code / dataset availability confirmedEurope PMC · OpenAlex · checked 5 Sept 2026
Maize-soybean intercropping is a sustainable intensive agroecosystem, though the productivity is constrained by interspecific competition for water and light resources. To enhance the water use efficiency in this intercropping system and understand canopy structure dynamics under the water-limited conditions of arid northwest China, this study proposes a novel optimization strategy that synchronizes deficit irrigation scheduling with crop-specific water requirements during critical phenological phases. Four irrigation regimes were implemented: W1 (full irrigation for both maize and soybean crops), W2 (maize-full and soybean-deficit), W3 (maize-deficit and soybean-full), and W4 (dual deficit). Through UAV-based high-resolution 3D canopy reconstruction (R = 0.98 for plant height validation), 14 spatial-geometric descriptors were quantified. The W2 strategy demonstrated superior competitive coordination, enhancing aggressivity of maize (Ams) by 85.9 % through strategic canopy reconfiguration: 11.8 % reduction in maize maximum leaf layer width position (MLLWP), 28.3 % decrease in inter-specific canopy overlap area (COA), and 40.0 % compression of shading convex hull volume (SCHV). These optimized structural adaptations synergistically enhanced photosynthetically active radiation interception (+13.4 %) while achieving concurrent reductions in crop evapotranspiration (ET, -19.7 %) without yield penalty, thereby elevating irrigation water use efficiency (IWUE) by 14.4 % and water equivalent ratio (WER) by 15.9 %. This work provides mechanistic insights into canopy architecture-mediated resource competition mitigation and establishes a technological framework for sustainable intensification in water-limited environments.
Why it matches plant phenotyping methodsUAVによる3Dキャノピー再構成を用いた植物構造形質の取得と検証が、灌漑試験の主要な解析基盤として明示されているため、実質的なフェノタイピング手法の応用に該当する。
abstractThrough UAV-based high-resolution 3D canopy reconstruction (R = 0.98 for plant height validation), 14 spatial-geometric descriptors were quantified.
Reproduction assets foundThe paper's data availability statement explicitly deposits the authors' analysis source code on a public GitHub repository, which qualifies as a paper-specific public code asset. The study's phenotype data (UAV-derived 3D canopy point clouds, geometric trait measurements, yield/biomass data) are only available upon请求,Code · publicThe source code used in this study is available for noncommercial use and the code can be downloaded from https://github.com/Pepe-oss/3D-Reconstruction-analysis-of-maize-soybean-intercropping-competition-under-water-stress . The data of this study are available from the corresponding author upon request.Open asset ↗Pepe-oss/3D-Reconstruction-analysis-of-maize-soybean-intercropping-competition-under-water-stresslines:320-407Code / dataset availability confirmedEurope PMC · OpenAlex · checked 5 Sept 2026
Three-dimensional (3D) reconstruction technologies for crops are of significant importance in the context of smart breeding and precision agriculture, as they enable accurate characterization of crop spatial architecture and developmental dynamics. Such capabilities provide essential phenotypic information for the rapid selection of breeding materials and informed agronomic decision-making. A critical requirement for the practical application of crop 3D models is high-accuracy organ-level segmentation. However, the absence of a stage-universal segmentation framework capable of operating across complete soybean growth cycle remains a major bottleneck hindering progress in this field. To address this issue, we propose SOY3DSEG-a high-precision framework based on an improved Point Transformer, designed to support the full developmental spectrum of soybean (V1-R7). The framework incorporates a novel down sampling strategy termed Dynamic Multi-Stage Sampling Strategy (DMSS), alongside multi-scale feature enhancement and a local geometry-aware attention mechanism, enhancing segmentation accuracy and efficiency. Performance evaluations across 12 consecutive soybean growth stages (V1 to R7) indicate that SOY3DSEG achieved an average mean Intersection-over-Union (mIoU) of 93.34 % for stem-leaf segmentation-surpassing RandLA-Net, BAAF-Net, PointNet++, and PointConv by over 30 %, and outperforming the baseline Point Transformer by 14.18 %. A moderate accuracy decline appears at R6-R7 due to dense canopies and strong occlusion, yet SOY3DSEG retains clear superiority over the baseline Point Transformer, demonstrating robustness under complex morphology. In cross-crop transfer tests limited to early seedling stages of maize and tomato, the model achieves an mIoU of approximately 99 %, indicating strong early-stage transferability while mature-stage generalization across species remains open for future study. SOY3DSEG thus provides a stage-robust and scalable solution for full-cycle soybean phenotyping and growth monitoring, contributing to precision agricultural practice.
Why it matches plant phenotyping methods大豆の3D点群から器官レベル形態を抽出する分割フレームワークを開発・評価しており、植物表現型取得手法が研究の中心である。
abstractA critical requirement for the practical application of crop 3D models is high-accuracy organ-level segmentation.
Reproduction assets foundThe authors state that the dataset (Soybean-MVS point clouds) and program code used in this study are publicly available at their GitHub repository, which is an allowed URL.Code · publicThe dataset and program code used in this study can be found at the link below: https://github.com/NiuJiarui718/SOY3DSEG .Open asset ↗NiuJiarui718/SOY3DSEGlines:306-323Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Precise and timely identification of cotton leaf diseases is essential for sustaining crop yield and quality, yet manual inspection remains time-consuming, labor-intensive, and prone to error. Existing automated approaches are limited by insufficient dataset diversity, inconsistent evaluation practices, limited use of explainable AI (XAI), and high computational cost. To address these challenges, we propose an attention-enhanced CNN ensemble, namely CottonLeafNet, which integrates lightweight convolutional neural networks for accurate cotton leaf disease classification across two publicly available datasets. CottonLeafNet achieves state-of-the-art performance, obtaining 98.33% accuracy, a macro F1-score of 0.9833, Cohen's kappa of 0.9800, a mean PPV of 0.9838, and an NPV of 0.9967 on Dataset D1, with an inference time of 0.51 s per image. On Dataset D2, it reaches 99.43% accuracy, a macro F1-score of 0.9942, Cohen's kappa of 0.9924, a mean PPV of 0.9943, and an NPV of 0.9981, with a 0.40 s inference time. Moreover, a unified eight-class dataset created by merging both datasets yields a test accuracy of 99.08%. Robustness analysis under artificially induced class imbalance further confirms the model's stability, with consistently strong macro F1-scores. To evaluate the generalization capability of the proposed CottonLeafNet, we conducted cross-dataset experiments, and the results indicate that the model maintains moderate performance even when trained and tested on different datasets. Gradient-Weighted Class Activation Mapping (Grad-CAM) visualizations demonstrate that CottonLeafNet reliably attends to disease-relevant regions, enhancing interpretability. Finally, real-time feasibility is validated through a web-based deployment achieving ≈1 s inference per image. These results establish CottonLeafNet as an accurate, robust, interpretable, and computationally efficient solution for automated cotton leaf disease diagnosis.
Why it matches plant phenotyping methods綿花葉の画像から病害状態を推定する分類手法を開発・評価しており、植物病害フェノタイピングが中心的である。
abstractwe propose an attention-enhanced CNN ensemble, namely CottonLeafNet, which integrates lightweight convolutional neural networks for accurate cotton leaf disease classification across two publicly available datasets.
Reproduction assets foundThe paper's plant-phenotyping inputs are three publicly available Kaggle cotton leaf disease image datasets (D1, D2, and cross-dataset D3) explicitly named in the Data availability statement. No author analysis code, trained model checkpoints, or supplementary code repository is disclosed in the supplied blocks.Dataset · publicThe datasets analyzed during the current study are publicly available in the Kaggle repository. Dataset D1 can
be accessed atOpen asset ↗Kagglepdf-page:17 lines:68-84Dataset · publicThe dataset used for
cross-dataset testing is publicly available at:Open asset ↗pdf-page:17 lines:68-84Code / dataset availability confirmedOpenAlex · checked 5 Sept 2026
Field / plotMultispectral / hyperspectralLeafVisualization / data managementLeaf traitsPhotosynthesis / fluorescence
Abstract. Accurate assessment of leaf functional traits is crucial for a diverse range of applications from crop phenotyping to parameterizing global climate models. Leaf reflectance spectroscopy offers a promising avenue to advance ecological and agricultural research by complementing traditional, time-consuming gas exchange measurements. However, the development of robust hyperspectral models for predicting leaf photosynthetic capacity and associated traits from reflectance data has been hindered by limited data availability across species and environments. Here we introduce the Global Spectra-Trait Initiative (GSTI), a collaborative repository of paired leaf hyperspectral and gas exchange measurements from diverse ecosystems. The GSTI repository currently encompasses over 7500 observations from 397 species and 41 sites gathered from 36 published and unpublished studies, thereby offering a key resource for developing and validating hyperspectral models of leaf photosynthetic capacity. The GSTI database is developed on GitHub (https://github.com/plantphys/gsti, last access: 4 January 2026) and published to ESS-DIVE https://doi.org/10.15485/2530733, Lamour et al., 2025). It includes gas exchange data, derived photosynthetic parameters, and key leaf traits often associated with traditional gas exchange measurements such as leaf mass per area and leaf elemental composition. By providing a standardized repository for data sharing and analysis, we present a critical step towards creating hyperspectral models for predicting photosynthetic traits and associated leaf traits for terrestrial plants.
Why it matches plant phenotyping methods葉のハイパースペクトルとガス交換・光合成形質を標準化して収録するデータベースを構築し、植物フェノタイピングモデルの開発・検証に供することが中心である。
abstractHere we introduce the Global Spectra-Trait Initiative (GSTI), a collaborative repository of paired leaf hyperspectral and gas exchange measurements from diverse ecosystems.
Reproduction assets foundThe paper describes the GSTI database of paired leaf hyperspectral and gas-exchange measurements, with both the data and R processing/model-fitting code publicly available on GitHub and archived releases on ESS-DIVE.Code · publicThe GSTI data and code are available in the public GitHub repository at https://github.com/plantphys/gsti (last access: 4 January 2026)Open asset ↗https://github.com/plantphys/gstilines:537-549Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Abstract Phenotyping is crucial for understanding crop trait variation and advancing research, but is currently limited by expensive, labor-intensive monitoring. New phenotypic trait monitoring methods are being proposed to reduce this so-called phenotyping bottleneck via automation. These methods are often data-driven, requiring a dataset recorded with a specific sensor and corresponding reference values for developing novel methods. To this end, we present the MuST-C (Multi-Sensor, multi-Temporal, multiple Crops) dataset, which contains field data from various sensors collected over a growing season, covering six crop species. All data was georeferenced for alignment across sensors and dates. To collect our dataset, we deployed aerial and ground robotic platforms equipped with RGB cameras, LiDARs, and multispectral cameras, aiming to capture a wide variety of modalities and observations from different viewpoints. In addition to sensor data, we also provide manually collected leaf area index and biomass reference measurements. Our dataset enables the development of novel automatic phenotypic trait estimation methods, allows comparisons across different sensors, and generalizability across crop species.
Why it matches plant phenotyping methods複数センサー・ロボットプラットフォームによる圃場フェノタイピング用データセットを構築・提供し、形質推定法の開発、センサー比較、汎化評価を可能にすることが中心的な貢献である。
abstractwe present the MuST-C (Multi-Sensor, multi-Temporal, multiple Crops) dataset
Reproduction assets foundThe paper's MuST-C multi-sensor, multi-temporal crop phenotyping dataset (RGB/multispectral images, LiDAR point clouds, LAI and biomass reference measurements) is publicly available via the authors' project webpage, and the authors' custom Python processing/loading code is publicly available on GitHub.Dataset · publicThe MuST-C dataset is available via our project webpage https://www.ipb.uni-bonn.de/data/MuST-C/or directly via the bonndata public access repository 10.60507/FK2/OX9XTM34Open asset ↗html-lines:421-440Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Abstract Wheat, the third most widely consumed cereal crop worldwide, faces substantial yield and quality losses as a result of rust disease, notably leaf rust, stem rust, and stripe rust. These rust disease, caused by Puccinia triticina , Puccinia graminis , and Puccinia striiformis , respectively, are capable of causing significant yield losses in wheat in the absence of timely detection. Conventional disease identification relies heavily on manual visual inspection, which is time consuming, labor intensive, and prone to error, especially in large scale agricultural systems. To address these limitations, this study proposes a deep learning-based framework for the early detection and classification of wheat rust diseases. A real-time dataset was developed using field images collected from various wheat-growing regions and augmented with publicly available data. The dataset comprises images of healthy leaves and those affected with the three major rust diseases. A modified convolutional neural network (CNN) architecture was employed for extract features and disease classification. Experimental results demonstrate that the proposed approach achieves high classification accuracy, highlighting its effectiveness as a reliable tool for automated wheat rust detection in precision agriculture. By enabling rapid and accurate disease identification, the system supports timely decision-making, reduces potential yield losses, and improves crop management practices, thereby contributing to food security and sustainable agricultural production.
Why it matches plant phenotyping methods小麦葉の画像からさび病の有無・種類を推定する深層学習手法が研究の中心であり、植物病害状態の画像ベース表現型計測に該当する。
abstractA real-time dataset was developed using field images collected from various wheat-growing regions and augmented with publicly available data.
Reproduction assets foundThe paper's own wheat rust image dataset (field images from North Punjab, Pakistan plus Kaggle-sourced images, with disease severity, GPS, variety, and weather metadata) is publicly deposited on Kaggle via an explicit repository link in Table 1. No author analysis code or trained model checkpoint is publicly released.Dataset · publict, Stripe Rust
Collection Region North Punjab, Pakistan
Collection Period Feb–March 2025
Collection Method Field observation + Kaggle image samples
Plant Growth Stage Tillering to heading
Field Data Includes Disease severity, GPS, wheat variety, weather data
Usage Disease classification, model training, analysis
Repository Link https://www.kaggle.com/datasets/sabaunnisa/wheat-rust-disease
We have divided the datasets 1294 into 962 training
images and 332 testing images. In the current study, a 3:1
ratio was used to create the training, and validation sets for
the image dataset, meaning 75% of the data 722 used to
training and 25% 240 to validation. A fixed random seed
(seed = 42) was used toOpen asset ↗Kaggle · sabaunnisa/wheat-rust-diseasepdf-raw-page:4 lines:1-66Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Potato plants are highly vulnerable to numerous diseases that can substantially affect both yield and quality. Conventional approaches for detecting these diseases are often labor-intensive, slow, and prone to inaccuracies, particularly under variable environmental conditions. This study presents a hybrid deep learning architecture, termed potato leaf diseases DenseNet (PLDNet), which integrates a DenseNet-based convolutional neural network with a Transformer-based attention module to accurately classify potato leaf diseases. Furthermore, an adaptive parametric activation function, referred to as Adaptive Flatten p-Mish (AFpM), is proposed to enhance the model's learning flexibility and representational capacity. When evaluated on the PlantVillage and Mendeley datasets, PLDNet attains classification accuracies of 99.54% and 87.50%, respectively, surpassing contemporary state-of-the-art models and activation techniques. The proposed framework exhibits strong generalization performance and offers a scalable, efficient approach for automated plant disease identification. To highlight the novelty, the proposed AFpM activation function introduces a learnable parameter enabling adaptive nonlinearity, improving over Mish, Swish, and PFpM activation functions through dynamic gradient control. AFpM improves accuracy by 2.52% on Mendeley dataset, and 1.93% on PlantVillage dataset compared to PFpM, and by more than 3% compared to Swish and Mish.
Why it matches plant phenotyping methods葉画像から植物病害状態を推定する深層学習モデルと新規活性化関数を開発・評価しており、植物フェノタイピング手法が研究の中心である。
abstractThis study presents a hybrid deep learning architecture, termed potato leaf diseases DenseNet (PLDNet), which integrates a DenseNet-based convolutional neural network with a Transformer-based attention module to accurately classify potato leaf diseases.
Reproduction assets foundThe paper's phenotyping inputs are two public leaf-image datasets used directly for its classification experiments: the Mendeley Potato Leaf Disease dataset (explicitly deposited with URL) and the PlantVillage dataset via Kaggle (explicitly linked in Data availability). The authors' PLDNet code is only promised 'upon' Dataset · publicsis. A.M initially drafted the paper, and all the authors (A.M, A.C, and N.A) reviewed and edited the paper.
Funding
Open access funding provided by University of Inland Norway. INN have subscription for open-access (OA) publication in Scientific Reports, Nature.
Data availability
The dataset used in this study is available at: https://data.mendeley.com/datasets/ptz377bwb8/1 and https://www.kaggle.com/datasets/emmarex/plantdisease .
Code availabilityOpen asset ↗Mendeley · ptz377bwb8/1lines:1390-1403Dataset · publicthors (A.M, A.C, and N.A) reviewed and edited the paper.
Funding
Open access funding provided by University of Inland Norway. INN have subscription for open-access (OA) publication in Scientific Reports, Nature.
Data availability
The dataset used in this study is available at: https://data.mendeley.com/datasets/ptz377bwb8/1 and https://www.kaggle.com/datasets/emmarex/plantdisease .
Code availabilityOpen asset ↗Kaggle · emmarex/plantdiseaselines:1390-1403Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Jackfruit cultivation is highly affected by leaf diseases that reduce yield, fruit quality, and farmer income. Early diagnosis remains challenging due to the limitations of manual inspection and the lack of automated and scalable disease detection systems. Existing deep-learning approaches often suffer from limited generalization and high computational cost, restricting real-time field deployment. Methods This study proposes CNNAttLSTM, a hybrid deep-learning architecture integrating Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) units, and an attention mechanism for multi-class classification of algal leaf spot, black spot, and healthy jackfruit leaves. Each image is divided into ordered 56×56 spatial patches, treated as pseudo-temporal sequences to enable the LSTM to capture contextual dependencies across different leaf regions. Spatial features are extracted via Conv2D, MaxPooling, and GlobalAveragePooling layers; temporal modeling is performed by LSTM units; and an attention mechanism assigns adaptive weights to emphasize disease-relevant regions. Experiments were conducted on a publicly available Kaggle dataset comprising 38,019 images, using predefined training, validation, and testing splits. Results The proposed CNNAttLSTM model achieved 99% classification accuracy, outperforming the baseline CNN (86%) and CNN-LSTM (98%) models. It required only 3.7 million parameters, trained in 45 minutes on an NVIDIA Tesla T4 GPU, and achieved an inference time of 22 milliseconds per image, demonstrating high computational efficiency. The patch-based pseudo-temporal approach improved spatial-temporal feature representation, enabling the model to distinguish subtle differences between visually similar disease classes. Discussion Results show that combining spatial feature extraction with temporal modeling and attention significantly enhances robustness and classification performance in plant disease detection. The lightweight design enables real-time and edge-device deployment, addressing a major limitation of existing deep-learning techniques. The findings highlight the potential of CNNAttLSTM for scalable, efficient, and accurate agricultural disease monitoring and broader precision agriculture applications.
Why it matches plant phenotyping methods植物葉の病害状態を画像から分類する深層学習手法を提案・比較評価しており、病害フェノタイピングの方法開発が中心である。
abstractThis study proposes CNNAttLSTM, a hybrid deep-learning architecture integrating Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) units, and an attention mechanism for multi-class classification of algal leaf spot, black spot, and healthy jackfruit leaves.
Reproduction assets foundThe paper's core phenotyping input is a publicly available Kaggle jackfruit leaf disease image dataset (38,019 images) explicitly linked in the data availability statement; no author code or model checkpoints are deposited.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://www.kaggle.com/datasets/shuvokumarbasak4004/jackfruit-leaf-diseases .Open asset ↗Kaggle · shuvokumarbasak4004/jackfruit-leaf-diseaseslines:823-830Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 15 Sept 2026
BACKGROUND: Plant developmental plasticity, particularly in root system architecture, is fundamental to understanding adaptability and agricultural sustainability. Existing automated phenotyping solutions face limitations, including binary segmentation approaches, restricted structural analysis capabilities, and text-based interfaces that limit accessibility, with most focusing solely on root structures while overlooking valuable information from simultaneous analysis of multiple plant organs. FINDINGS: ChronoRoot 2.0 builds upon established low-cost hardware while significantly enhancing software capabilities and usability. The system employs nnUNet architecture for multi-class segmentation, demonstrating significant accuracy improvements while simultaneously tracking 6 distinct plant structures encompassing root, shoot, and seed components: main root, lateral roots, seed, hypocotyl, leaves, and petiole. This architecture enables easy retraining and incorporation of additional training data without requiring machine learning expertise. The platform introduces dual specialized graphical interfaces: a Standard Interface for detailed architectural analysis with novel gravitropic response parameters and a Screening Interface enabling high-throughput analysis of multiple plants through automated tracking. Functional principal component analysis integration enables discovery of novel phenotypic parameters through temporal pattern comparison. We demonstrate multi-species analysis, with Arabidopsis thaliana and Solanum lycopersicum, both morphologically distinct plant species. Three use cases in Arabidopsis thaliana and validation with tomato seedlings demonstrate enhanced capabilities: circadian growth pattern characterization, gravitropic response analysis in transgenic plants, and high-throughput etiolation screening across multiple genotypes. CONCLUSIONS: ChronoRoot 2.0 maintains the low-cost, modular hardware advantages of its predecessor while dramatically improving accessibility through intuitive graphical interfaces and expanded analytical capabilities. The open-source platform makes sophisticated temporal plant phenotyping more accessible to researchers without computational expertise. SOFTWARE AVAILABILITY: https://chronoroot.github.io.
Why it matches plant phenotyping methods根・シュート・種子を時系列追跡し、植物形態・成長・重力応答などの表現型を抽出するオープンプラットフォームの開発と検証が中心である。
titleChronoRoot 2.0: An Open AI-Powered Platform for 2D Temporal Plant Phenotyping
Reproduction assets foundThe paper publicly releases its authors' analysis code (GitHub), the annotated plant image dataset used for segmentation training/validation (HuggingFace), a pre-configured Docker image, and a project home page, all with explicit availability statements and URLs matching allowed entries.Code · publicapproach to open science will not only ensure transparency and reproducibility but also allow the system to evolve alongside the changing needs of the plant biology community.
Availability of source code and requirements
Project name: ChronoRoot 2.0.
Project home page: https://chronoroot.github.io .
Main Source Code repository: https://github.com/ChronoRoot/ChronoRoot2 .
Operating system(s): Platform independent.
Programming language: Python.
Other requirements: Conda, Apptainer, or Docker.
License: GNU GPL 3.0.
Additional files
Supplementary Text S1 : Functional PCA. Provides an intuitive explanation of functional principal component analysis (FPCA) for readers without a quantitative backgrOpen asset ↗https://github.com/ChronoRoot/ChronoRoot2lines:439-479Dataset · publicgulates LAZY genes. Plant J. 2025;121:e70016. 10.1111/tpj.70016.
19.
Gaggion N, Boccardo N A, Bonazzola R, et al. ChronoRoot 2.0 Main Source Code Repository. 2026. https://github.com/ChronoRoot/ChronoRoot2 . Accessed 25 February 2026.
20.
Gaggion N, Boccardo N A, Bonazzola R, et al. ChronoRoot 2.0 Annotated Image Dataset. 2026. https://huggingface.co/datasets/ngaggion/ChronoRoot2 . Accessed 25 February 2026.
21.
Gaggion N, Boccardo N A, Bonazzola R, et al. ChronoRoot 2.0 Docker Image. 2026. https://hub.docker.com/r/ngaggion/chronoroot . Accessed 25 February 2026.
22.
Gaggion N, Boccardo N A, Bonazzola R, et al. ChronoRoot 2.0 Project Home Page. 2026. https://chronoroot.github.io . Accessed 2Open asset ↗https://huggingface.co/datasets/ngaggion/ChronoRoot2lines:568-618Code · publicical modules, and experimental protocols. We hope that this approach to open science will not only ensure transparency and reproducibility but also allow the system to evolve alongside the changing needs of the plant biology community.
Availability of source code and requirements
Project name: ChronoRoot 2.0.
Project home page: https://chronoroot.github.io .
Main Source Code repository: https://github.com/ChronoRoot/ChronoRoot2 .
Operating system(s): Platform independent.
Programming language: Python.
Other requirements: Conda, Apptainer, or Docker.
License: GNU GPL 3.0.
Additional files
Supplementary Text S1 : Functional PCA. Provides an intuitive explanation of functional princOpen asset ↗lines:439-479Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Stomatal pores, formed by guard cells, govern the critical trade-off between carbon assimilation and water loss in plants. Their dynamic responses to environmental stresses, such as stomatal oscillations and drought “stress memory” (hysteresis), have lacked a unified mechanistic explanation. While abscisic acid (ABA) is believed to play key roles in water stress responses, no model has linked its core regulatory kinetics to these complex stomatal behaviors. Here, we introduce a coupled hydropassive-hydroactive (HP-HA) model that integrates leaf hydraulics with the biokinetics of guard cell-autonomous ABA regulation and plasma membrane-mediated osmoregulation. We demonstrate that this framework predicts accurate, genotype-specific stomatal regulation across wildtype, ABA-insensitive mutant ( ost1-3 ), and ABA-synthesis mutant ( aao3-2 ) in Arabidopsis thaliana ( At ) and that non-linear feedbacks in ABA autoregulation can drive both stomatal oscillations and hysteresis. This work unifies genetic, signaling, and membrane processes with leaf-scale physiological dynamics, providing a new predictive foundation for understanding and modulating plant management of water use and water stress.
Why it matches plant phenotyping methods葉の水理とABA制御を統合した予測モデルを開発し、遺伝子型別の気孔コンダクタンス制御を検証しており、植物生理表現型の取得・予測手法が中心である。
abstractHere, we introduce a coupled hydropassive-hydroactive (HP-HA) model that integrates leaf hydraulics with the biokinetics of guard cell-autonomous ABA regulation and plasma membrane-mediated osmoregulation.
Reproduction assets foundThe paper's Code Availability section explicitly archives all MATLAB code used to generate the study's stomatal conductance modeling results in a Zenodo repository (DOI 10.5281/zenodo.17888362) and on GitHub (desai-sahil/sys-bio-gs), both listed as allowed URLs. This is author analysis code directly reproducing the hydCode · publicn analysis are provided in SI sections S5. Comprehensive tables listing all model parameters,
362
their sources, and the methodology for parameter fitting are provided in SI section S7.
363
364
Code Availability
365
All MATLAB code used to generate the results in this study is permanently archived in a Zenodo
366
repository at: https://doi.org/10.5281/zenodo.17888362. The most current version of the code is also
367
available on GitHub at: https://github.com/desai-sahil/sys-bio-gs.git. Refer to SI section S7.C for details
368
on steps to run the code to reproduce the results in main text.
369
370
371
Acknowledgements
372
We thank F. E. Rockwell, V. Bacheva, S. Sen, I. Gabay, E. Wu, J. BeldiOpen asset ↗Zenodo · 10.5281/zenodo.17888362pdf-raw-page:9 lines:1-74Code · publicnd the methodology for parameter fitting are provided in SI section S7.
363
364
Code Availability
365
All MATLAB code used to generate the results in this study is permanently archived in a Zenodo
366
repository at: https://doi.org/10.5281/zenodo.17888362. The most current version of the code is also
367
available on GitHub at: https://github.com/desai-sahil/sys-bio-gs.git. Refer to SI section S7.C for details
368
on steps to run the code to reproduce the results in main text.
369
370
371
Acknowledgements
372
We thank F. E. Rockwell, V. Bacheva, S. Sen, I. Gabay, E. Wu, J. Belding, and P. Jain for insightful
373
discussions. This work was supported by the Center for Research on Programmable Open asset ↗GitHub · desai-sahil/sys-bio-gspdf-raw-page:9 lines:1-74Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Crop traits are the integrated outcome of genetic variation, environmental conditions, and their complex interactions, rendering accurate prediction from genetic markers alone a persistent challenge. Here, we present KineticGP, a computational framework that combines genomic prediction with genotype-specific kinetic models of C 4 photosynthesis to make predictions of leaf photosynthetic traits across genotypes from a multiple-parent advanced generation intercross maize population. Using genetic markers and gas exchange measurements from three field seasons, we show that KineticGP outperforms a baseline genomic prediction model in predicting the photosynthetic rate at saturating light by 86% for unseen genotypes across two seen seasons. In addition, KineticGP enabled us to survey genetic variability in enzyme kinetic parameters, which can be used to identify targets for the improvement of photosynthesis. This approach paves the way for interrogating and integrating the dynamic interactions between genotype and environment to improve the accuracy of photosynthetic trait predictions.
Why it matches plant phenotyping methods葉の光合成形質を予測する計算フレームワーク自体が研究の中心であり、遺伝マーカーとガス交換測定を統合した植物生理形質の推定手法を開発・評価している。
abstractHere, we present KineticGP, a computational framework that combines genomic prediction with genotype-specific kinetic models of C 4 photosynthesis to make predictions of leaf photosynthetic traits across genotypes
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicAll codes and data to ensure the reproducibility of the results can be accessed at https://github.com/Rudan-X/KineticGP .Open asset ↗GitHub · Rudan-X/KineticGPlines:231-264Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
Accurate and efficient leaf trait measurement is essential for plant phenotyping, agronomy, and ecological studies. In this work, we introduce Leaf Analyzer, a novel open-source, fully automated computer vision-based tool for high-throughput leaf morphological trait measurement such as leaf area, dimensions, perimeter, count, and percent damage. Unlike existing methods that rely on strong foreground-background contrast or controlled imaging conditions, Leaf Analyzer employs an unsupervised clustering approach based on the K-means++ clustering algorithm and a novel Leaf Background Separation (LBS) feature, which combines the L∗ and b∗ channels from CIEL∗a∗b∗ color space and the saturation channel from HSV color space. The proposed method and the LBS feature can effectively distinguish leaves from the background across varying lighting conditions, leaf colors, and camera orientations. To evaluate the performance of the new software, we conducted comprehensive quantitative and qualitative comparison experiments with two widely used software tools - Petiole Pro and LeafByte, demonstrating that Leaf Analyzer achieves superior accuracy and consistency, particularly under challenging imaging conditions. Additionally, we explore methods to further enhance measurement precision, including leaf flattening and the integration of supplementary leaf features such as texture features and color specific features. Beyond leaf trait measurement, we showcase the versatility of Leaf Analyzer in a range of applications, including nondestructive plant phenotyping, seed counting, root trait analysis, leaf area measurement for petri dish-grown plants, plant projected silhouette area or crown projection area estimation, leaf damage assessment, and broader plant science applications, making it a valuable tool for researchers working in laboratory and field environments.
Why it matches plant phenotyping methods葉形態形質を自動抽出するオープンソース画像解析ツールの開発と、既存ツールとの定量比較検証が研究の中心であるため。
abstractIn this work, we introduce Leaf Analyzer, a novel open-source, fully automated computer vision-based tool for high-throughput leaf morphological trait measurement such as leaf area, dimensions, perimeter, count, and percent damage.
Reproduction assets foundThe authors state that the Leaf Analyzer source code, installer files, and all data (including evaluation images) used in this study are publicly available on their GitHub repository.Code · publicThe Leaf Analyzer source code, platform-specific installer files, and all data used in this study are publicly available on our GitHub repository at https://github.com/squashking/Leaf-Analyzer .Open asset ↗squashking/Leaf-Analyzerlines:239-277Dataset · publicAll the images used in the evaluation have been published on our Github repository ( https://github.com/squashking/Leaf-Analyzer ).Open asset ↗squashking/Leaf-Analyzerlines:134-155Code / dataset availability confirmedCrossref · checked 5 Sept 2026
ABSTRACT Sample preparation is an important first step to obtain high quality mass spectrometry imaging (MSI) data. Preparing plant tissues is especially challenging for MSI of thin tissues along the lateral dimensions. The unique challenges involved with plant tissues, such as fragile cell walls, hydrophobic barriers, and specific tissue structures, often lead to inefficiency and difficulties in sample preparation. Imprinting plant tissues onto porous polytetrafluoroethylene (pPTFE) sheet has been widely used to extract internal metabolites in leaves and petals while keeping spatial resolution for MSI. However, pressure applications were typically made manually using a vise or pliers leading to low reproducibility and resolution in MS images. In this study, we introduce a home‐built pneumatic press (PNP) that has been designed to precisely control the pressure application parameters during imprinting. To evaluate the performance of the new device, Lemna minor fronds, Arabidopsis thaliana , and Bacopa monnieri leaves were imprinted onto the pPTFE with PNP, vise, or pliers, and matrix‐assisted laser desorption/ionization (MALDI) MSI was obtained on the imprints. The PNP showed dramatic improvements in reproducibility and image quality compared to manual pressure application tools.
Why it matches plant phenotyping methods植物組織の空間的な代謝物情報を再現性よく取得するための空気圧式インプリンティング装置を開発し、手動法と性能比較している。植物表現型取得に関わる試料調製・イメージング手法が中心である。
abstractIn this study, we introduce a home‐built pneumatic press (PNP) that has been designed to precisely control the pressure application parameters during imprinting.
Reproduction assets foundThe paper's MALDI-MSI data (imzML files of imprinted Lemna minor, Arabidopsis, and Bacopa tissues) are openly deposited in a paper-specific METASPACE project, as stated in the Data Availability Statement. No author analysis code or trained models are disclosed.Dataset · publicData Availability Statement
The data that support the findings of this study are openly available in METASPACE (https://metaspace2020.eu/project/pnp_ptfe_imprinting_plant).Open asset ↗METASPACE · pnp_ptfe_imprinting_planthtml-lines:230-307Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Timely and precise detection of diseases on plants is crucial for minimizing losses during crop production in order to sustain food supply demands worldwide. In this work, deep learning (DL) was used to develop an automatic disease identification system for the leaves of potato and mango plants using two publicly available datasets, the PlantVillage Potato Leaf Disease (2,152 images) dataset and the Kaggle Mango Leaf Disease dataset (4,000 images). Images were pre-processed, augmented, and split into training and testing datasets (80:20), to enable better model generalization. Four deep learning architectures, namely Convolutional Neural Networks (CNN), AlexNet, Residual Networks (ResNet), and EfficientNet, were evaluated in the context of multi-class disease classification. The baseline CNN achieved a training accuracy of 93.67% and a testing accuracy of 92.61%, with balanced precision and recall (92.5%), thus providing a very strong feature extraction and classification capability. AlexNet showed moderate performance (91.3% training, 90.2% validation), and a very small overfitting was observed. ResNet had an efficient convergence, and attained 96.7% validation accuracy in just a few epochs, thus pointing out the advantage of residual connections in the context of deeper learning. EfficientNet surpassed all the other architectures, since it reached a training accuracy of 98.2% and a validation accuracy of 97.8%, with very small loss (≈ 0.015) and no overfitting, thus proving to have the best generalization ability. The models demonstrated stability and discriminative ability with the support of confusion matrices and accuracy and loss plots produced on an epoch-wise basis. Therefore, the findings indicate that DL models can be adapted for real-time and accurate plant disease diagnosis, establishing a pathway for early remediation, and supporting precision agriculture. The research establishes the opportunity for EfficientNet to be considered a promising solution for scalable smart farming.
Why it matches plant phenotyping methods葉画像から植物病害を分類する深層学習手法を開発・比較し、複数データセットで精度を検証しており、植物の病害状態の取得・推定が研究の中心である。
abstractdeep learning (DL) was used to develop an automatic disease identification system for the leaves of potato and mango plants
Reproduction assets foundThe paper uses two public Kaggle leaf-image datasets (PlantVillage potato, mango leaf disease) and states that all code, preprocessing scripts, dataset splits, and model artifacts are publicly available in a GitHub repository (also archived on Zenodo). All three are paper-specific, public, and actionable.Dataset · publicThe datasets analyzed during the current study are available in (https://www.kaggle.com/datasets/aarishasifkhan/plantvillage-potato-disease-dataset)Open asset ↗html-lines:473-503Code · publicAll code, preprocessing scripts, dataset splits, and model artifacts used in this study are publicly available in the GitHub repository at: [https://github.com/logeswarig/PROJECT_1].Open asset ↗logeswarig/PROJECT_1html-lines:473-503Code / dataset availability confirmedEurope PMC · checked 13 Sept 2026
Field / plotPhotogrammetry / SfM / MVSLiDAR / point cloudMultispectral / hyperspectralLeaf2D/3D reconstructionArchitecture / morphology / geometry
Conifer shoots possess highly complex geometrical structures at a very fine spatial resolution. Accurately characterizing the full architecture of a conifer shoot, which influences how radiation is scattered, has proven challenging. Previous radiative transfer models for coniferous stands have represented these structures in a relatively simplified or coarse manner. This paper presents a dataset that can be used for up-scaling of needle to shoot optical properties and studying the influence of detailed three-dimensional (3D) structure of shoot to light scattering within tree crown. The dataset includes 3D structural information as well optical properties of needles and twigs for 27 shoots of two conifer species present in both locations (3 shoots per species and position in the crown) - Scots pine ( Pinus sylvestris L.) and Norway spruce ( Picea abies L. Karst. ). The samples were collected on 22nd April 2024 in Rájec, the Czech Republic and 17th September 2024 in Järvselja, Estonia. Subsequently blue light 3D photogrammetry scanning technique was used to obtain their high-resolution 3D point cloud representations. Reflectance and transmittance measurements of needles were obtained using a spectroradiometer and an integrating sphere. For each of these samples, the dataset comprises a photo of the sampled shoot, obtained 3D surface reconstruction, and optical properties of conifer needles and twigs (hemispherical-conical reflectance and transmittance factors) in the spectral range of 400-2000 nm. A detailed 3D representation of needle shoots, when combined with radiative transfer modeling, may offer a means to study and compensate for inaccuracies in the measurement of needle optical properties and to enhance the assessment of shoot scattering characteristics.
Why it matches plant phenotyping methods針葉樹シュートの3D構造をフォトグラメトリで取得し、光学特性とともに再利用可能なデータセットとして提供しているため、植物形態・構造の計測手法が中心です。
abstractThis paper presents a dataset that can be used for up-scaling of needle to shoot optical properties and studying the influence of detailed three-dimensional (3D) structure of shoot to light scattering within tree crown.
Reproduction assets foundThe paper is a Data in Brief article describing a public Mendeley Data repository containing the paper's own phenotyping measurements: 3D surface geometry models (.obj) of Scots pine and Norway spruce shoots, sample photos (.jpg), and needle/twig optical property spectra (HCRF/HCTF, .csv, 400-2000 nm). The repository, Dataset · publicRepository name: Mendeley
Data identification number: 10.17632/h39f9t7fjg.1
Direct URL to data: https://data.mendeley.com/datasets/h39f9t7fjg/2Open asset ↗Mendeley · 10.17632/h39f9t7fjg.1lines:47-74Code / dataset availability confirmedCrossref · checked 5 Sept 2026
Callus induction is a complex procedure in plant organ, cell, and tissue culture that underpins processes such as metabolite production, regeneration, and genetic transformation. It is important to monitor callus formation alongside subjective evaluations, which require labor-intensive care. In this research, the first curated lentil (Lens culinaris) callus dataset for instance segmentation was experimentally generated using three genotypes as one data set: Firat-87, Cagil, and Tigris. Leaf explants were cultured on MS medium fortified with different concentrations of gross regulators of BA and NAA to induce callus formation. Three biologically relevant stages, the leaf stage, the green callus, and the necrosis callus, were produced. During this process, 122 high-resolution images were obtained, resulting in 1185 total annotations across them. The dataset was evaluated across four successive generations (v5/7/8/11) of YOLO deep learning models under identical conditions using mAP, Dice coefficient, Precision, Recall, and IoU, together with efficiency metrics including parameter counts, FLOPs, and inference speed. The results show that anchor-based variants (YOLOv5/7) relied on predefined priors and showed limited boundary precision, whereas anchor-free designs (YOLOv8/11) used decoupled heads and direct center/boundary regression that provided clear advantages for callus structures. YOLOv8 reached the highest instance segmentation precision with mAP50@0.855, while it matched the accuracy with greater efficiency and achieved real-time inference with 166 FPS.
Why it matches plant phenotyping methods植物組織培養におけるカルスの形成段階・壊死状態を画像からインスタンスセグメンテーションする手法、データセット、モデル比較を中心に扱っており、植物状態の取得・定量化が本研究の主要な技術貢献である。
titleReal-Time Callus Instance Segmentation in Plant Tissue Culture Using Successive Generations of YOLO Architectures
Reproduction assets foundThe paper's lentil callus image dataset with annotations (122 images, 1185 annotations) is publicly available on Roboflow Universe per the Data Availability Statement. The YOLOv5 GitHub link and Ultralytics docs are generic third-party libraries, not authors' analysis code, and the FAO link is a cited reference, so allDataset · publicThe dataset used in this study, including annotated images for callus detection, is publicly available and can be accessed at Roboflow Universe: https://universe.roboflow.com/yunus-7v2b5/callus-hug7d , accessed on 13 September 2025. This repository contains all images and annotations generated and analyzed during the current study.Open asset ↗Roboflow Universe · callus-hug7dlines:314-345Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Rice leaf diseases pose a significant and escalating threat to global food security. Timely and accurate detection, particularly in the critical early stages characterized by subtle lesions, is paramount for effective disease management. However, existing solutions often struggle with the complexities of real-world field environments (e.g., variable lighting, occlusions, complex backgrounds), computational constraints on edge devices, and limited generalizability across diverse disease types and plant species. To address these challenges, this study proposes a novel lightweight deep learning framework specifically designed for robust rice leaf disease detection. Our key innovations include: (1) A Multi-branch Large-kernel Fusion Depthwise (MLFD) module enhancing multi-scale contextual feature extraction critical for identifying subtle early lesions; (2) A Multi-scale Dilated Transformer Attention (MDTA) module integrating spatial and channel attention mechanisms to improve feature representation under complex conditions; (3) A Lightweight Detection Head (Lo-Head) optimized with grouped and depthwise convolutions, drastically reducing model complexity without sacrificing accuracy. Crucially, extensive experiments demonstrate the framework's superior performance. On a dedicated rice leaf disease dataset, it achieves a mean Average Precision mAP@0.5:0.95 of 62.62%, outperforming state-of-the-art lightweight detectors including YOLOv5n (56.73%), YOLOv8n (57.41%), YOLOv10n (56.14%), and the baseline YOLOv11n (60.85%), while maintaining low computational demands (6.3 GFLOPs, 2.66M parameters). Significantly, rigorous generalization experiments validate the model's exceptional transferability. Evaluated on independent datasets encompassing potato and tomato leaf diseases, the proposed framework consistently surpasses comparable models in mAP@0.5:0.95, demonstrating its robust capability to detect diseases across different plant species. This combination of high accuracy, computational efficiency, and remarkable cross-crop generalizability positions our framework as a highly promising tool for practical deployment on resource-limited edge devices (e.g., drones, field sensors) in smart agriculture systems, enabling proactive disease surveillance and precision control strategies across diverse crops.
Why it matches plant phenotyping methods植物葉の病徴を画像から検出する深層学習手法の開発と、独立データセットによる性能・汎化性検証が中心であり、植物の病害状態を直接推定するフェノタイピング手法に該当する。
abstractthis study proposes a novel lightweight deep learning framework specifically designed for robust rice leaf disease detection.
Reproduction assets foundThe paper's rice leaf disease detection dataset was curated from three publicly available repositories (one Kaggle, two Roboflow), and the cross-species generalization used two additional public Roboflow datasets (tomato and potato leaf diseases). All five URLs are explicitly listed in the article as data sources. No作者Dataset · publicData Sources: The dataset utilized in this study was curated and screened from the following publicly available online repositories:.Open asset ↗html-lines:110-216Dataset · publicThe Tomato Leaf and Potato Leaf disease datasets were acquired from public domain resources. The dataset links are: Tomato Leaf Diseases: https://universe.roboflow.com/dyploma/tomato-leaf-diseases-4xa5iOpen asset ↗dyploma/tomato-leaf-diseases-4xa5ihtml-lines:747-783Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Abstract The research develop an accurate and efficient method for detecting multiple corn leaf diseases to support sustainable agricultural practices in Soppeng Regency, Indonesia. The goal is to design a Convolutional Neural Network (CNN) model capable of classifying corn leaf diseases, including rust, blight, and gray leaf spot, using high-resolution image data. The research employed a balanced dataset sourced from open-access repositories, followed by preprocessing, data augmentation, and CNN model optimization. The model’s performance was evaluated using accuracy, precision, recall, and F1-score to ensure comprehensive assessment. Experimental results show that the proposed CNN achieved high accuracy across all disease classes, with strong per-class metrics, indicating robust performance in distinguishing visually similar symptoms. The classification results with the Convolutional Neural Network algorithm have 95% training data accuracy and 93% test data accuracy in detecting leaf diseases in corn plants. The findings contribute to agricultural technology by offering a scalable and field-deployable disease detection system that can be integrated into mobile or edge-based platforms. Limitations include reliance on publicly available datasets, which may not fully capture the variability of local field conditions. The research concludes that the proposed CNN model can significantly enhance early disease detection, reduce dependency on manual inspections, and support precision agriculture. Future research should focus on expanding the dataset with locally captured images, incorporating real-time image acquisition, and optimizing the model for deployment in low-resource environments to improve adaptability and reliability.
Why it matches plant phenotyping methodsトウモロコシ葉の病害症状を画像から分類するCNN手法の開発・性能評価が中心であり、植物の病害状態を直接推定するため、植物フェノタイピング手法として含める。
abstractThe goal is to design a Convolutional Neural Network (CNN) model capable of classifying corn leaf diseases, including rust, blight, and gray leaf spot, using high-resolution image data.
Reproduction assets foundThe paper's phenotyping analysis is based entirely on a public Kaggle corn leaf disease image dataset, explicitly cited with URL and access date in the Data Availability statement and Methods. No author code or trained model is shared.Dataset · publicrch and innovation in the field of agricultural technology and artificial intelligence
Data Availability
The data used in this study comes from a publicly available and curated image repository. The dataset was obtained from the “Corn or Maize Leaf Disease Dataset” . The data source can be seen in the Kaggle – https://www.kaggle.com/smaranjitghose/corn-or-maize-leaf-disease-dataset
References
Rozi F, et al. Indonesian market demand patterns for food commodity sources of carbohydrates in facing the global food crisis. Heliyon. 2023;9(6):e16809. 10.1016/j.heliyon.2023.e16809 .
Yu B-G, Chen X-X, Zhou C-X, Ding T-B, Wang Z-H, Zou C-Q. Nutritional composition of maize grain assoOpen asset ↗Kaggle · smaranjitghose/corn-or-maize-leaf-disease-datasetlines:224-246Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Digital agriculture and smart farming require crop health monitoring methods that balance detection accuracy with computational cost. Rice leaf diseases threaten yield, while field images often contain small multi-scale lesions, variable illumination and cluttered backgrounds. This paper investigates SCD-YOLOv11n, a lightweight detector designed with these constraints in mind. The model replaces the YOLOv11n backbone with a StarNet backbone and integrates a C3k2-Star module to enhance fine-grained, multi-scale feature extraction. A Detail-Strengthened Cross-scale Detection (DSCD) head is further introduced to improve localization of small lesions. On this architecture, we design a DepGraph-based mixed group-normalization pruning rule and apply channel-wise feature distillation to recover performance after pruning. Experiments on a public rice leaf disease dataset show that the compressed model requires 1.9 MB of storage, achieves 97.4% mAP@50 and 76.2% mAP@50:95, and attains a measured speed of 184 FPS under the tested settings. These results provide a quantitative reference for designing lightweight object detectors for rice disease monitoring in digital agriculture scenarios.
Why it matches plant phenotyping methodsイネ葉の病斑を画像から検出・局在化する軽量モデルを開発し、精度・圧縮性能・速度を評価しており、植物病害状態の取得手法が研究の中心である。
abstractThis paper investigates SCD-YOLOv11n, a lightweight detector designed with these constraints in mind.
Reproduction assets foundThe paper's rice leaf disease image dataset (6715 annotated images) is explicitly stated to be publicly available on Roboflow, and an MDPI supplementary file is provided with additional dataset information. No author analysis code or trained model checkpoints are publicly deposited.Dataset · publicThe rice disease detection dataset used in this study is publicly available at: https://universe.roboflow.com/dreamydaisy-cdagn/rice-dyl9n/dataset/4 (accessed on 10 December 2025).Open asset ↗dreamydaisy-cdagn/rice-dyl9n/dataset/4lines:480-547Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Plants move chloroplasts in response to light, changing the optical properties of leaves. Low irradiance induces chloroplast accumulation, while high irradiance triggers chloroplast avoidance. Chloroplast movements may be monitored through changes in leaf transmittance and reflectance, typically in red light. We present a step-by-step procedure for the detection of chloroplast positioning using reflectance hyperspectral imaging in white light. We show how to employ machine learning methods to classify leaves according to the chloroplast positioning. The convolutional network is a method of choice for the analysis of the reflectance spectra, as it allows low levels of misclassification. As a complementary approach, we propose a vegetation index, called the Chloroplast Movement Index (CMI), which is sensitive to chloroplast positioning. Our method offers a high-throughput, contactless way of chloroplast movement detection. Key features • Protocol for detached leaves handled in laboratory conditions. • Based on differential (dark-adapted versus irradiated) hyperspectral images of plant leaves. • Data analysis includes machine learning methods and the calculation of a vegetation index. • Requires irradiation equipment apart from the hyperspectral camera set.
Why it matches plant phenotyping methods葉の反射ハイパースペクトル画像から葉緑体位置を検出・分類する手法と指標を開発し、高スループット測定として提示しており、植物表現型取得が中心である。
abstractWe present a step-by-step procedure for the detection of chloroplast positioning using reflectance hyperspectral imaging in white light.
Reproduction assets foundThe protocol explicitly deposits its authors' analysis code (HyperspectralImageProcessing.m, including the pretrained CNN classifier for chloroplast positioning) on GitHub and makes the original hyperspectral images of Arabidopsis and Nicotiana leaves used in the paper's figures available on figshare. Both are paper-‐Code · publicAll code has been deposited to GitHub: https://github.com/plantPhotobiologyLab/machine-learning-for-chloroplast-movement-detection (access date, 08/18/2025)Open asset ↗plantPhotobiologyLab/machine-learning-for-chloroplast-movement-detectionhtml-lines:104-130Dataset · publicOriginal files with hyperspectral images of Nicotiana benthamiana and Arabidopsis thaliana (WT and phot2) leaves, including recordings shown in Figure 3 and Figure 4 of this protocol, can be downloaded from https://figshare.com/articles/dataset/Hyperspectral_images_of_Arabidopsis_thaliana_and_Nicotiana_benthamiana_leaves/30402409?file=58898569Open asset ↗html-lines:104-130Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Cassava is one of the most widely cultivated crops worldwide, renowned for its rich natural ingredients and numerous nutritional benefits. However, the complex interdependencies among its features often pose challenges in image restoration and segmentation, particularly when identifying disease regions. In previous work, this manifested as higher false positives and misidentification of non-relevant areas, leading to a decline in precision and accuracy. Methods To address these issues, this study proposed an efficient artificial intelligence-powered image analysis system that leverages optimal feature selection with a HyperCapsInception-ResNet-V2-CNN model to enhance disease detection accuracy. Initially, the dataset was collected from the Kaggle repository, its name was Cassava Leaf Disease Classification, and it comprised 21,367 different images. Our approach began by normalizing cassava plant disease data using adaptive Gaussian Otsu thresholding. Histogram color evaluation and iterative clustering fragmentation were then applied to better isolate disease variations and improve precision. Subsequently, Cascaded Canny Edge Segmentation (CCES) was used to effectively segment the disease region. The disease variation properties were further evaluated using the Optimal Spider Swarm Intelligence Technique (OSSIT) to reduce irrelevant feature dimensions. For classification, the HyperCapsInception-ResNet-V2-CNN model was employed to categorize cassava diseases, including cassava bacterial blight (CBB), cassava mosaic disease (CMD), cassava green mite (CGM) disease, and cassava brown streak disease (CBSD), along with regular and abnormal leaf states. Results The proposed method's simulation results achieved 98.15% accuracy, a 97.22% F1-score, and 96.02% precision, outperforming other traditional methods such as EfficientNetB3, AlexNet, Faster-RCNN, and InceptionV3. Discussion Both optimized feature selection with OSSIT and hybrid HyperCapsInception-ResNet-V2-CNN architecture significantly enhanced the detection reluctance and the classification of the data. These findings indicate that the proposed system is effective in the automated detection of cassava disease and has a high potential of being practical in agricultural practices especially in precision farming and early detection of diseases.
Why it matches plant phenotyping methodsカッサバ葉画像から病変領域を分割・抽出し、病害状態を分類する画像解析手法の開発が研究の中心であるため、植物表現型手法として採用。
abstractCascaded Canny Edge Segmentation (CCES) was used to effectively segment the disease region.
Reproduction assets foundThe paper uses the public Kaggle 'Cassava Leaf Disease Classification' dataset (21,367 cassava leaf images) as its phenotyping input; the dataset is publicly downloadable at the authors' stated URL, which matches an allowed URL.Dataset · publicThe Cassava Leaf Disease Classification dataset is available on Kaggle and comprises 21,367 images. The images have an average resolution of 512 × 512 pixels. The data are split into training and test sets, enabling machine learning algorithms to be trained and tested to accurately detect diseases. The data are available for download from Kaggle: https://www.kaggle.com/datasets/nirmalsankalana/cassava-leaf-disease-classification .Open asset ↗Kaggle · cassava-leaf-disease-classificationlines:523-601Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Rice leaf diseases significantly reduce productivity and cause economic losses, highlighting the need for early detection to enable effective management and improve yields. This study proposes Artificial Neural Network (ANN)-based image-processing techniques for timely classification and recognition of rice diseases. Despite the prevailing approach of directly inputting images of rice leaves into ANNs, there is a noticeable absence of thorough comparative analysis between the Feature Analysis Detection Model (FADM) and the Direct Image-Centric Detection Model (DICDM), specifically when it comes to evaluating the effectiveness of Feature Extraction Algorithms (FEAs). Hence, this research presents initial experiments on the Feature Analysis Detection Model, utilizing various image Feature Extraction Algorithms, Dimensionality Reduction Algorithms (DRAs), Feature Selection Algorithms (FSAs), and Extreme Learning Machine (ELM). The experiments are carried out on datasets encompassing 3829 original rice leaf images across six classes (bacterial leaf blight, brown spot, leaf blast, leaf scald, sheath blight rot, and healthy leaf). A Direct Image-Centric Detection Model is established without the utilization of any FEA, and the evaluation of classification performance relies on different metrics. Ultimately, an exhaustive contrast is performed between the achievements of the Feature Analysis Detection Model and the Direct Image-Centric Detection Model in classifying rice leaf diseases. The results reveal that the highest performance is attained using the Feature Analysis Detection Model. We have also applied Gradient-weighted Class Activation Mapping (Grad-CAM) for visual interpretability of the model's predictions. The adoption of the proposed Feature Analysis Detection Model for detecting rice leaf diseases holds excellent potential for improving crop health, minimizing yield losses, and enhancing the overall productivity and sustainability of rice farming.
Why it matches plant phenotyping methodsイネ葉の病害状態を画像から分類する手法を中心に、特徴抽出モデルと直接画像モデルを比較・評価しており、植物病害フェノタイピング手法の開発・検証に該当する。
abstractThis study proposes Artificial Neural Network (ANN)-based image-processing techniques for timely classification and recognition of rice diseases.
Reproduction assets foundThe paper's rice leaf disease classification experiments are built on a publicly available Kaggle image dataset (3829 rice leaf images across six classes), explicitly linked in the Data Availability Statement. Datasets produced during the study are only available upon request from the corresponding author.Dataset · publicThe dataset is publicly available at https://www.kaggle.com/datasets/vbookshelf/riceleafdiseases.Open asset ↗Kaggle · vbookshelf/riceleafdiseaseshtml-lines:883-951Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
The date palm ( Phoenix dactylifera L.) is a vital crop in arid and semi-arid regions, contributing over $13 billion annually to the global economy. However, it faces significant yield losses due to pests, such as the red palm weevil, and diseases, including Bayoud and Black Scorch. Currently, expert visual inspection is the primary method of management, but it is time-consuming, subjective, and unsuitable for detecting large-scale or early-stage damage. Automated approaches based on classical machine learning offer limited improvements due to their lack of generalizability and environmental sensitivity. Recent deep learning methods, such as CNNs and Vision Transformers, have improved classification accuracy, but treat tasks like classification, detection, segmentation, and severity estimation as separate. This paper proposes an integrated Reveal-Aware Hybrid Vision-Language and Transformer-based AI framework that combines GAN-based augmentations for feature generation, CLIP for multimodal classification, PaliGemma2 for text-based detection, Grounding DINO + SAM 2.1 for zero-shot segmentation, and a Vision Transformer regression model for severity prediction. This end-to-end explainable diagnostic pipeline achieved 98% classification accuracy, 95.8% precision, 91.3% recall, and 94.2% F1-score across two datasets: nine classes of infected date palm leaves and three classes of date palm diseases. The proposed framework demonstrated detection accuracy of 94-98%, high-quality segmentations, and reliable severity estimates. This integrated approach highlights the potential of combining AI, vision-language models, and transformers for scalable, accurate, and sustainable plant disease management.
Why it matches plant phenotyping methods植物病害の分類・検出・セグメンテーション・重症度推定を統合した画像ベース手法の開発と評価が中心であり、感染植物の状態を直接推定するため収録対象。
abstractThis paper proposes an integrated Reveal-Aware Hybrid Vision-Language and Transformer-based AI framework
Reproduction assets foundThe paper analyzes two publicly available date palm disease image datasets (Kaggle and Mendeley Data), both explicitly named in the data availability statement and used directly for the paper's phenotyping measurements. No author analysis code or trained model checkpoints are disclosed.Dataset · public1)Kaggle Dataset: The date palm leaf disease dataset used for this study is available on Kaggle and can be accessed via the following link: https://www.kaggle.com/datasets/hadjerhamaidi/date-palm-data. This dataset contains images of date palm leaves categorized into three classes: healthy leaves, brown spot disease, and white scale infection.Open asset ↗Kaggle · hadjerhamaidi/date-palm-datahtml-lines:510-532Dataset · public2)Mendeley Dataset: Additionally, the second dataset, which includes images of date palm leaves with eight types of diseases, can be accessed via Mendeley Data: https://data.mendeley.com/datasets/g684ghfxvg/2. This dataset was collected from 10 date farms in Madinah, Saudi Arabia.Open asset ↗Mendeley Data · g684ghfxvg/2html-lines:510-532Code / dataset availability confirmedOpenAlex · arXiv · checked 6 Sept 2026
High resolution phenotyping at the level of individual leaves offers fine-grained insights into plant development and stress responses. However, the full potential of accurate leaf tracking over time remains largely unexplored due to the absence of robust tracking methods-particularly for structurally complex crops such as canola. Existing plant-specific tracking methods are typically limited to small-scale species or rely on constrained imaging conditions. In contrast, generic multi-object tracking (MOT) methods are not designed for dynamic biological scenes. Progress in the development of accurate leaf tracking models has also been hindered by a lack of large-scale datasets captured under realistic conditions. In this work, we introduce CanolaTrack, a new benchmark dataset comprising 5,704 RGB images with 31,840 annotated leaf instances spanning the early growth stages of 184 canola plants. To enable accurate leaf tracking over time, we introduce LeafTrackNet, an efficient framework that combines a YOLOv10-based leaf detector with a MobileNetV3-based embedding network. During inference, leaf identities are maintained over time through an embedding-based memory association strategy. LeafTrackNet outperforms both plant-specific trackers and state-of-the-art MOT baselines, achieving a 9% HOTA improvement on CanolaTrack. With our work we provide a new standard for leaf-level tracking under realistic conditions and we provide CanolaTrack - the largest dataset for leaf tracking in agriculture crops, which will contribute to future research in plant phenotyping. Our code and dataset are publicly available at https://github.com/shl-shawn/LeafTrackNet.
Why it matches plant phenotyping methods葉レベルの時系列追跡という植物表現型取得手法を開発し、専用ベンチマークデータセットで評価しているため、方法が中心である。
abstractTo enable accurate leaf tracking over time, we introduce LeafTrackNet, an efficient framework that combines a YOLOv10-based leaf detector with a MobileNetV3-based embedding network.
Reproduction assets foundThe authors explicitly state that the CanolaTrack dataset (5,704 annotated RGB images of 184 canola plants), the LeafTrackNet code, and trained model weights are publicly available at their GitHub repository.Code · publicOur code and dataset are publicly
available at https://github.com/shl-shawn/LeafTrackNet.Open asset ↗shl-shawn/LeafTrackNet · LeafTrackNetpdf-page:1 lines:1-53Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
The development of artificial intelligence (AI) and machine learning (ML) based tools for 3D phenotyping, especially for maize, has been limited due to the lack of large and diverse 3D datasets. 2D image datasets fail to capture essential structural details such as leaf architecture, plant volume, and spatial arrangements that 3D data provide. To address this limitation, we present MaizeField3D (website), a curated dataset of 3D point clouds of field-grown maize plants from a diverse genetic panel, designed to be AI-ready for advancing agricultural research. Our dataset includes 1045 high-quality point clouds of field-grown maize collected using a terrestrial laser scanner (TLS). Point clouds of 520 plants from this dataset were segmented and annotated using a graph-based segmentation method to isolate individual leaves and stalks, ensuring consistent labeling across all samples. This labeled data was then used for fitting procedural models that provide a structured parametric representation of the maize plants. The leaves of the maize plants in the procedural models are represented using Non-Uniform Rational B-Spline (NURBS) surfaces that were generated using a two-step optimization process combining gradient-free and gradient-based methods. We conducted rigorous manual quality control on all datasets, correcting errors in segmentation, ensuring accurate leaf ordering, and validating metadata annotations. The dataset also includes metadata detailing plant morphology and quality, alongside multi-resolution subsampled point cloud data (100k, 50k, 10k points), which can be readily used for different downstream computational tasks. MaizeField3D will serve as a comprehensive foundational dataset for AI-driven phenotyping, plant structural analysis, and 3D applications in agricultural research.
Why it matches plant phenotyping methods3D点群の収集・分割・注釈・手続き型モデル化を中核とする、植物表現型解析向けの再利用可能なデータセットである。
abstractwe present MaizeField3D (website), a curated dataset of 3D point clouds of field-grown maize plants from a diverse genetic panel, designed to be AI-ready for advancing agricultural research.
Reproduction assets foundThe paper's own MaizeField3D dataset (1045 TLS point clouds, 520 segmented/annotated plants, metadata, STL/DAT procedural model outputs) is publicly available on Hugging Face, with a project website and public GitHub code for the procedural NURBS surface generation used in the analysis.Dataset · publicThe MaizeField3D dataset is publicly available on the Hugging Face Datasets platform at https://huggingface.co/datasets/BGLab/MaizeField3D. It includes high-resolution point clouds, segmented plant models, metadata, and reconstructed outputs in STL and DAT formats.Open asset ↗BGLab/MaizeField3Dhtml-lines:343-354Code · publicThe code for procedural NURBS surface generation used in this work is available at https://github.com/baskargroup/ProceduralMaize3D.Open asset ↗baskargroup/ProceduralMaize3Dhtml-lines:343-354Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Grapevine water relations are increasingly influenced by drought under climate change, with significant implications for yield, fruit composition and wine quality. Stable isotopes of hydrogen, oxygen, carbon and nitrogen (δ 2 H, δ 18 O, δ 13 C and δ 15 N) provide sensitive tracers of plant water sources and physiological responses to stress. Here, we combined dual water isotopes (δ 2 H, δ 18 O), carbon and nitrogen isotopes (δ 13 C, δ 15 N), and high-resolution micrometeorological/soil observations to diagnose drought dynamics in Vitis vinifera cv. Sauvignon blanc (Orlești, Romania; 2023-2024). Dual-isotope relationships delineated progressive evaporative enrichment along the soil-plant-atmosphere continuum, with slopes LMWL ≈ 6.41 > stem ≈ 5.0 > leaf ≈ 2.2, consistent with kinetic fractionation during transpiration (leaf) superimposed on source-water signals (stem). Weekly leaf δ 18 O covaried strongly with relative humidity (RH; r = -0.69) and evapotranspiration (ET; r = +0.56), confirming atmospheric control of short-term enrichment, while stem isotopes showed buffered responses to soil water. We integrated Δ 18 O (leaf-stem), RH, ET, and soil matric potential at 60 cm (Soil 60 ) into an Isotopic Drought Index (IDI), which captured the onset, intensity, and persistence of the July-August 2024 drought (IDI 0-100 > 90; RH 40 mm wk -1 , Soil 60 > 100 cb). Carbon and nitrogen isotopes provided complementary, integrative diagnostics: δ 13 C increased (less negative) with drought (r = -0.52 with RH; +0.49 with IDI), reflecting higher intrinsic water-use efficiency, whereas δ 15 N rose with soil dryness and IDI (leaf: r ≈ +0.48 with Soil 60 ; +0.42 with IDI), indicating constraints on N acquisition and enhanced internal remobilization. Together, multi-isotope and environmental data yield a mechanistic, field-validated framework linking atmospheric demand and edaphic limitation to vine physiological and biogeochemical responses and demonstrate the operational value of an isotope-informed drought index for precision viticulture.
Why it matches plant phenotyping methods複数同位体と環境データからブドウの水分状態・干ばつ応答を推定するIsotopic Drought Indexを構築し、圃場で検証した研究であり、植物の生理状態取得手法が中心である。
abstractWe integrated Δ 18 O (leaf-stem), RH, ET, and soil matric potential at 60 cm (Soil 60 ) into an Isotopic Drought Index (IDI), which captured the onset, intensity, and persistence of the July-August 2024 drought
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Supplement · publicTable S1: Isotopic data of leaf and stem of Vitis vinifera cv. Sauvignon Blanc blanc from Orlești-Vâlcea (Romania), during 2023-2024 vintage; Table S2: Meteorological and soil measurements (Romania), during the sampling campaign (Orlești – Vâlcea, Romania; 2023-2024 vintage)Open asset ↗lines:149-204Code / dataset availability confirmedOpenAlex · Crossref · checked 6 Sept 2026
Automated three-dimensional plant phenotyping is an essential tool for non-destructive analysis of plant growth and structure. This paper presents a low-cost system based on stereo vision for depth estimation and morphological characterization of maize plants. The system incorporates an automatic detection stage for the object of interest using deep learning techniques to delimit the region of interest (ROI) corresponding to the plant. The Semi-Global Block Matching (SGBM) algorithm is applied to the detected region to compute the disparity map and generate a partial three-dimensional representation of the plant structure. The ROI delimitation restricts the disparity calculation to the plant area, reducing processing of the background and optimizing computational resource use. The deep learning-based detection stage maintains stable foliage identification even under varying lighting conditions and shadowing, ensuring consistent depth data across different experimental conditions. Overall, the proposed system integrates detection and disparity estimation into an efficient processing flow, providing an accessible alternative for automated three-dimensional phenotyping in agricultural environments.
Why it matches plant phenotyping methods植物の3次元形態を取得・特徴づけるステレオビジョンと深度推定システムの開発が中心であり、明確な植物フェノタイピング手法です。
abstractThis paper presents a low-cost system based on stereo vision for depth estimation and morphological characterization of maize plants.
Reproduction assets foundThe paper's Data Availability Statement openly deposits the original study data (the 544 stereo RGB maize images and related phenotyping data) on OSF at a DOI, which is a paper-specific, publicly actionable asset. No author analysis code repository is explicitly stated.Dataset · publicData Availability Statement: The original data presented in the study are openly available in OSF at
https://doi.org/10.17605/OSF.IO/MN6P9.Open asset ↗OSF · 10.17605/OSF.IO/MN6P9pdf-page:18 lines:1-59Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Background Developing an effective machine vision system is crucial to successfully deploying robotic inspection in open field conditions and controlled environments like greenhouses. Robotic arms with vision-based deep learning models offer an efficient, real-time, non-invasive crop monitoring solution. In agricultural settings, they enable consistent, automated inspection under varying conditions, reduce labor dependency, and support early disease detection, enhancing productivity and sustainability in precision farming. Although considerable progress has been made in computer vision-based approaches, significant challenges persist in developing models that reliably perform under the diverse and variable conditions encountered in real-world agricultural settings. Method Within the domain of precision agriculture, we introduce an advanced robotic system for the detection of plant diseases, utilizing an innovative model based on deep learning principles. This system introduces an algorithm for real-time analysis, called as Strawberry Leaf Disease Inspection (SLDI). The algorithm integrates the use of Receptive Guided Channel Attention (RGCA) alongside a Deep Context Aggregator (DCA), designed to significantly improve the characterization and representation of feature sets, thereby enhancing the overall accuracy and efficiency of disease identification. To optimize the system performance and preserve real-time performance, a Multi-Scale Feature Fusion Module (MSFF) is proposed that facilitates a comprehensive multi-level representation, enabling the model to capture disease symptoms promptly. The SLDI algorithm is deployed on a robotic platform equipped with an RGB camera, enabling real-time, in-field inspection of strawberry crops. Results The proposed system is trained on two publicly available datasets, PlantDoc and PlantVillage. It attains a precision of 91.10% and a recall of 88.50%, while maintaining a real-time processing speed of 76.50 frames per second (fps). Experimental field inspection of strawberry studies demonstrates that the proposed model significantly outperforms existing approaches in accuracy and efficiency.
Why it matches plant phenotyping methodsイチゴ葉の病徴をRGB画像と深層学習でリアルタイム検出するアルゴリズムおよびロボットプラットフォームを開発・評価しており、植物病害状態の表現型取得が中心である。
abstractwe introduce an advanced robotic system for the detection of plant diseases, utilizing an innovative model based on deep learning principles.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe publicly avail- able datasets can be accessed at [ www.plantvillage.org ] and [ https://github.com/pratikkayal/PlantDoc-Dataset ].Open asset ↗pratikkayal/PlantDoc-Datasetlines:271-381Code / dataset availability confirmedCrossref · Europe PMC · checked 6 Sept 2026
Abstract The development of accurate and efficient plant disease classification systems is vital for addressing the challenges of climate change and the growing global demand for food. This study presents $$\hbox {V}^2$$ PlantNet, a novel lightweight multi-class classification model based on a modified MobileNet architecture, designed to detect plant leaf diseases across a diverse range of crop types. $$\hbox {V}^2$$ PlantNet employs depthwise separable convolutions to significantly reduce model complexity without compromising accuracy. The architecture integrates Batch Normalization (BN) and Rectified Linear Unit (ReLU) activation after each convolutional layer, while a multi-stage design enhances feature extraction and overall performance. Despite its compact size, comprising only 389,286 parameters and requiring just 1.46 MB of memory, $$\hbox {V}^2$$ PlantNet achieved up to 99% training accuracy, with validation and test accuracies of 97% and 98%, respectively. Across most classes, precision, recall, and F1-scores ranged from 0.97 to 1.0, demonstrating consistent and robust generalization across diverse plant species. These architectural innovations enable $$\hbox {V}^2$$ PlantNet to outperform larger models such as ResNet-50 and Inception V3 in terms of computational efficiency, owing to its smaller model size (1.46 MB), reduced parameter count (389,286), and faster inference time (0.676 s), offering a scalable solution for real-time plant disease detection in precision agriculture.
Why it matches plant phenotyping methods植物葉の病害状態を画像から分類する軽量な深層学習手法を開発し、精度・計算量・推論速度を比較評価しており、フェノタイピング手法が中心である。
abstractThis study presents $$\hbox {V}^2$$ PlantNet, a novel lightweight multi-class classification model based on a modified MobileNet architecture, designed to detect plant leaf diseases across a diverse range of crop types.
Reproduction assets foundThe paper's Data Availability statement explicitly links the PlantVillage dataset used to train and evaluate PlantNet, hosted publicly on Kaggle. No author code or trained model checkpoint is deposited.Dataset · publicThe datasets generated and analysed during the current study are available in the Kaggle repository (see link to dataset: https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset).Open asset ↗Kaggle · plantvillage-datasethtml-lines:1649-1677Code / dataset availability confirmedEurope PMC · bioRxiv · checked 6 Sept 2026
Citrus farming plays an essential role in agriculture; however, diseases like canker, greening, black spot, and melanose significantly reduce yield and fruit quality. Efficient classification of citrus leaf diseases is important for crop health maintenance and optimal crop yield. Traditional methods for leaf disease detection are slow, labor-intensive, and often inaccurate, which highlights the need for automated solutions. This research presents a novel hybrid approach for identifying citrus diseases by combining a vision transformer with deep learning architectures. Using Bidirectional Encoder Representation from Image Transformers (BEIT) and MobileNetV2 as feature extractors, the proposed model captures distinctive features from images, which are then classified using Support Vector Machine (SVM). The dataset includes four different disease categories and a healthy class. Data augmentation techniques are applied to improve model robustness. The experimental findings demonstrate that CitriBEiTNet achieves a remarkable training accuracy of 99.82% and a testing accuracy of 99.57%, outperforming current leading techniques. This model provides an efficient, scalable, and economical approach for early disease identification, enabling farmers to take preventive measures and improve agricultural yields.
Why it matches plant phenotyping methods柑橘葉画像から病害状態を自動分類する深層学習手法の開発が研究の中心であり、植物の病害表現型を直接推定している。
abstractThis research presents a novel hybrid approach for identifying citrus diseases by combining a vision transformer with deep learning architectures.
Reproduction assets foundThe paper uses a public Kaggle citrus leaf image dataset (1,023 images across black spot, canker, greening, healthy) as its phenotyping input, with an explicit public URL. No author analysis code or trained model checkpoints are reported as publicly available.Dataset · publicThe Kaggle dataset is publicly available at:
https://www.kaggle.com/datasets/sourabh2001/citrus-leaves-dataset/data.Open asset ↗Kaggle · sourabh2001/citrus-leaves-datasetpdf-page:5 lines:1-61Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Owing to changing climatic and environmental conditions, plant diseases are becoming increasingly prevalent, posing a serious threat to global agriculture. Timely and accurate diagnosis remains challenging, especially where scouting still relies on manual inspection. We propose B2-GraftingNet, a deep learning framework for automated detection of grape leaf diseases. B2-GraftingNet is a streamlined variant of our earlier B4-GraftingNet, retaining its strengths while simplifying blocks for faster inference and deployment. The architecture combines a VGG16 backbone with Inception-style blocks inside a custom CNN to extract robust, multi-scale features based on color, size, and shape. To reduce redundancy and improve generalization, Binary Particle Swarm Optimization (BPSO) selects informative features prior to classification. We evaluate Support Vector Machines (SVM) and k-Nearest Neighbors (KNN); a cubic SVM attains 99.56% peak accuracy on the public Kaggle grape-leaf dataset. For context, we also benchmarked standard pretrained CNNs on the same data, observing validation accuracies of 34.04% (VGG16), 34.04% (VGG19), 97.95% (Xception), 94.91% (Darknet), and 98.44% (ResNet-50); B2-GraftingNet matches or exceeds these while remaining lighter and faster to train and deploy. To enhance transparency and actionability, we pair Grad-CAM, LIME, and occlusion-sensitivity visualizations with a local gpt-oss:20b assistant (served via Ollama) that converts evidence into plain, grower-focused guidance and supports interactive chat validated by horticulturists. Results are further checked against expert-annotated ground-truth labels, confirming high accuracy and computational efficiency. Overall, B2-GraftingNet offers a reliable, interpretable, and scalable solution for early grape-leaf disease detection. The complete setup (code, model, web platform, configuration, and assets) is available on Zenodo: https://doi.org/10.5281/zenodo.17353656.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像から検出する深層学習手法を開発し、複数モデル・専門家アノテーションと比較検証しているため、植物フェノタイピング手法が中心である。
abstractWe propose B2-GraftingNet, a deep learning framework for automated detection of grape leaf diseases.
Reproduction assets foundThe paper's Data Availability Statement points to a public Zenodo deposit containing the grape leaf images used for disease classification, which is a paper-specific, publicly actionable asset. The underlying Kaggle source dataset is also cited, but the Zenodo record is the authors' own public deposit matching an exactDataset · publicICCK Journal of Image Analysis and Processing
reproducible runs, API examples for mobile image https://zenodo.org/records/18401218.
uploads and programmatic retrieval of classifications
and explainability overlays, as well as additional Funding
figures and code listings that mirror the production
This work was supported without any funding.
repository.
Conflicts of Interest
4 Conclusion
Syed Adil Hussain Shah is affiliated with the
In this study, we introdOpen asset ↗Zenodo · 18401218pdf-layout-page:16 lines:1-68Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
In this study, we present a combined image dataset created from two distinct plant species: Hibiscus and Tea leaf. The dataset consists of high-resolution images of leaves from both species, captured using a SONY α7 II DSLR camera and a OnePlus 7T lubricant Tea Leaf dataset includes images categorized into five disease classes: Algal Leaf Spot, Brown Blight, Grey Blight, Red Leaf Spot, and Healthy, while the Hibiscus Leaf dataset includes images labeled across eight conditions, including citrus spot, fungal infection, mild edge damage, and healthy foliage. To ensure balanced representation and address class imbalances, extensive data augmentation techniques-such as flipping, rotation, zooming, shifting, noise addition, and brightness adjustment-were applied, resulting in a total of 1,413 combined original images and 13,000 augmented images. The ConvNextTiny deep learning model was fine-tuned on this combined dataset to classify the various leaf conditions, achieving an overall accuracy of 96%. This demonstrates the model's robust performance and high discriminatory power across the diverse set of leaf diseases and conditions. This experiment highlights the utility of combining multiple plant species into a single dataset and utilizing a lightweight yet effective model like ConvNextTiny for plant disease classification. The resulting dataset, along with the model and training scripts, is publicly available to facilitate further research in plant pathology, computer vision, and smart farming applications, enabling more accurate and efficient early-stage disease detection for both Hibiscus and Tea plants.
Why it matches plant phenotyping methods植物葉の病害・健全状態を画像から分類するデータセットを構築し、分類モデルで性能評価しているため、植物フェノタイピング手法・ベンチマークが中心です。
abstractwe present a combined image dataset created from two distinct plant species: Hibiscus and Tea leaf
Reproduction assets foundThe paper's combined Hibiscus and Tea leaf disease image dataset is publicly deposited on Mendeley Data (DOI 10.17632/5bzy89brkv.4), and the authors' augmentation/training scripts are on a public GitHub repository; both are paper-specific, public, and directly actionable.Dataset · publicRepository name: Mendeley Data
Data identification number: 10.17632/5bzy89brkv.4
Direct URL to data: https://data.mendeley.com/datasets/5bzy89brkv/4Open asset ↗Mendeley Data · 10.17632/5bzy89brkv.4lines:1-46Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
This study introduces a hybrid deep learning model that combines CNN-based hierarchical feature extraction with light-efficient vision transformer self-attention to classify multiple types of cotton leaf diseases and fabric defects. Using Explainable AI (XAI) techniques, the framework enhances interpretability, allowing domain experts to better understand the model's decisions. Evaluated on four benchmark datasets, the proposed XCottL-FebViT achieved consistent improvements in accuracy, MCC, and F1 Score compared with leading transformer-based models, while maintaining computational efficiency through hyperparameter optimization. For CottonLeafNet and SAR-CLD, it attained training accuracies of 99.97% and 99.95%, with validation accuracies of 99.93% and 99.91%, respectively. In fabric defect classification, the model achieved 99.97% training accuracy on CottonFabricImageBD and FabricSpotDefect, with validation accuracies of 99.93% and 99.95%, respectively. A lightweight web-based application enables practical deployment for remote disease and defect detection. This work highlights the integration of interpretability, efficiency, and high performance in AI-driven agricultural and textile quality assessment.
Why it matches plant phenotyping methods綿花葉の病害を画像から分類する深層学習手法の開発・比較評価が中心であり、植物の病害状態を直接推定するため、フェノタイピング方法論として採用する。
abstractThis study introduces a hybrid deep learning model that combines CNN-based hierarchical feature extraction with light-efficient vision transformer self-attention to classify multiple types of cotton leaf diseases and fabric defects.
Reproduction assets foundThe paper's cotton leaf disease image datasets (CottonLeafNet, SAR-CLD-2024) are publicly available and directly used as phenotyping inputs, and the authors' analysis code is publicly deposited on GitHub and archived on Zenodo. Fabric defect datasets are excluded as non-plant assets; generic PyPI libraries are excludedCode · publicCode: Source code of the study is available at https://github.com/rezaul-h/CottonVerse.Open asset ↗github · rezaul-h/CottonVersehtml-lines:2058-2083Code / dataset availability confirmedCrossref · checked 15 Sept 2026
The farming of citrus is a crucial component of Pakistan’s fruit-based agricultural economy. But, the foliar diseases citrus canker, black spot, and greening have been posing a constant threat on citrus’s productivity. An optimal solution is an early and accurate detection of these diseases to improve the productivity. Therefore, this paper proposes an automated citrus leaf disease detection and classification framework based on deep convolutional neural networks (DCNNs). The proposed solution has five stages: image acquisition (dataset), preprocessing, data augmentation, deep feature extraction and optimization, and disease classification. Firstly, the images are obtained from a public dataset downloaded from Kaggle. Secondly, preprocessing techniques are used to improve the image quality and shape, thirdly the data augmentation techniques are used to enhance the model generalization, fourthly pre-trained models DenseNet-121, MobileNet, and InceptionV3 with transfer learning technique to extract deep features, and finally Adam optimizer and categorical cross-entropy loss function are used to fine tune the pre-trained models for classifications. The proposed model is evaluated on accuracy, precision, recall, and F1-score metrics. All the models demonstrated robust performance while DenseNet-121 achieved the best performance. The evaluation results assured the robustness of the use of transfer learning-based DCNN in citrus leaf disease detection.
Why it matches plant phenotyping methods柑橘葉の病害状態を画像から直接検出・分類する深層学習ワークフローが研究の中心であり、植物病害フェノタイピング手法に該当する。
abstractTherefore, this paper proposes an automated citrus leaf disease detection and classification framework based on deep convolutional neural networks (DCNNs).
Reproduction assets foundThe paper's phenotyping input is a public Kaggle citrus leaf image dataset (654 RGB images of healthy, blackspot, canker, and greening leaves) explicitly cited with a URL matching an allowed URL. No author code or trained models are reported as publicly available.Dataset · publictaset is essential. Additionally, the dataset must be prepared
so that our model can fully comprehend the data. The model will then be able to effectively
use that dataset for learning. A random sample of infected and healthy leaves images from the
datasets shown in Figure 1. The details of the images are provided in Table 1.
1
https://www.kaggle.com/datasets/sourabh2001/citrus-leaves-dataset?resource=downloadOpen asset ↗Kaggle · sourabh2001/citrus-leaves-datasetpdf-raw-page:3 lines:1-48Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Coffee is a vital agricultural commodity that sustains millions of farmers worldwide, yet its cultivation is increasingly threatened by devastating leaf diseases such as Leaf Rust, Phoma, Cercospora, and Leaf Miner. These diseases reduce photosynthetic efficiency, cause defoliation, and ultimately lower crop yield and quality. Traditional diagnostic methods, including visual inspection and laboratory-based tests such as PCR and ELISA, are often time-consuming, costly, and require expert intervention, making them impractical for large-scale use. To address these challenges, we propose EffResViT-SE FusionNet, a novel hybrid deep learning framework that integrates EfficientNetB3 and ResNet50 enhanced with Squeeze-and-Excitation (SE) blocks for adaptive local feature recalibration, along with a Vision Transformer (ViT) for modeling global contextual dependencies. This fusion design effectively combines CNN-based local feature extraction with transformer-based long-range attention in a unified architecture. The model was trained on a large-scale dataset comprising 58,555 coffee leaf images distributed across five classes: Healthy (18,984), Miner (16,983), Leaf Rust (8336), Cercospora (7681), and Phoma (6571). The dataset was split into 70%, 15%, and 15% testing. Key hyperparameters included the Adam optimizer, a learning rate of 0.001, a batch size of 32, and 80 training epochs, ensuring stable convergence. Experimental results demonstrate the superior capability of the proposed model, achieving an overall classification accuracy of 99%, with precision, recall, and F1-scores all ranging between 98% and 99% across all classes. Comparative analysis confirmed notable improvements over baseline models: ResNet50 (94% accuracy), EfficientNetB3 (95% accuracy), and standalone ViT (97% accuracy). Furthermore, ablation studies validated the critical role of SE blocks and feature fusion with the transformer in achieving optimal performance. These outcomes highlight EffResViT-SE FusionNet as a powerful, precise, and scalable solution for early detection and classification of coffee leaf diseases, supporting timely interventions and promoting sustainable agriculture.
Why it matches plant phenotyping methodsコーヒー葉画像から病害状態を分類する深層学習フレームワークの開発・比較検証が研究の中心であり、植物病害表現型の取得・推定に該当する。
titleEffResViT-SE FusionNet: A Hybrid Deep Learning Framework for Accurate Classification of Coffee Leaf Diseases.
Reproduction assets foundThe paper's plant-phenotyping input is a public Kaggle coffee leaf image dataset (58,555 images across five classes) explicitly linked in the Data Availability Statement. No author analysis code or trained model checkpoints are disclosed.Dataset · publicInterest
The authors declare no conflicts of interest.
Acknowledgments
Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2025R238), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.
Data Availability Statement
The dataset used in this study is publicly available on Kaggle: https://www.kaggle.com/datasets/noamaanabdulazeem/jmuben‐coffee‐dataset/data .
References
Adelaja , O.
, and
B.
Pranggono
. 2025 . “
Leveraging Deep Learning for Real‐Time Coffee Leaf Disease Identification
.” Aǧrı
7 , no. 1 : 13 .
10.3390/agriengineering7010013
.
Alirezazadeh , P.
,
M.
Schirrmann
, and
F.
Stolzenburg
. 2023 . “
Improving Deep Learning‐Based Plant DisOpen asset ↗Kaggle · jmuben‐coffee‐datasetlines:822-895Code / dataset availability confirmedCrossref · Europe PMC · checked 6 Sept 2026
Coffee, the world’s most traded tropical crop, is vital to the economies of many producing countries. However, coffee leaf diseases pose a serious threat to coffee quality and sustainable production. Deep learning has shown strong performance in plant disease identification through automatic image classification. Nevertheless, reliance on a single convolutional neural networks (CNNs) architecture restricts feature variability and real-world generalization. Moreover, limited work has systematically combined feature selection/reduction with CNNs, which constrains the advancement of hybrid models capable of capturing complementary features while ensuring computational efficiency without accuracy loss. This article presents an enhanced deep learning-based framework for coffee disease classification incorporating a hybrid strategy that integrates CNNs and advanced feature selection algorithms. GoogLeNet and ResNet18 are paired for complementary feature extraction, Principal Component Analysis (PCA) and Singular Value Decomposition (SVD) are employed for dimensionality reduction, and ANOVA and Chi-square are applied to select the most informative features. An Adam optimizer (learning rate = 0.001, batch size = 20, epochs = 50) with early stopping is used for training. Experiments on the BRACOL dataset achieved 99.78% accuracy, with precision, recall, and F1-score all exceeding 99% across classes. To the best of our knowledge, this study systematically integrates GoogLeNet and ResNet18 with PCA/SVD dimensionality reduction and analysis of variance (ANOVA)/Chi-square feature selection, for coffee disease classification, thereby addressing a key gap in prior research.
Why it matches plant phenotyping methodsコーヒー葉の病害を画像から分類する深層学習フレームワークが研究の中心であり、植物の病害状態を直接推定する実質的な表現型解析手法である。
abstractThis article presents an enhanced deep learning-based framework for coffee disease classification incorporating a hybrid strategy that integrates CNNs and advanced feature selection algorithms.
Reproduction assets foundThe paper uses the public BRACOL/RoCoLe coffee leaf image dataset (Mendeley) and provides authors' analysis code publicly on GitHub and Zenodo, all explicitly linked in the text.Dataset · publicWe utilized the BRACOL dataset, a publicly available dataset of coffee leaf images. The dataset can be accessed at the following DOI: ( https://data.mendeley.com/datasets/c5yvn32dzg/2 ).Open asset ↗lines:30-49Code · publicAll implementation details, including preprocessing scripts, model training, and evaluation codes, are available in the following GitHub repository: ( https://github.com/DrMaherAlrahhal/coffe-code ).Open asset ↗GitHublines:30-49Code · publicThe code is available at GitHub and Zenodo:
- https://github.com/DrMaherAlrahhal/coffe-code .
- w. (2025). coffee code. Zenodo. https://doi.org/10.5281/zenodo.17470672 .Open asset ↗Zenodo · 10.5281/zenodo.17470672lines:2688-2696Code / dataset availability confirmedOpenAlex · checked 14 Sept 2026
Leaf segmentation plays a crucial role in plant phenotyping and precision agriculture, enabling the monitoring of growth, disease detection, and informed crop management. However, accurate segmentation in natural environments is challenging due to complex backgrounds, overlapping structures, irregular boundaries, and varying illumination. This paper proposes a hybrid six-stage framework that integrates U-Net with the Fast Segment Anything Model (FastSAM) to achieve accurate and efficient leaf segmentation. The pipeline consists of initial U-Net segmentation, largest component filtering, contour extraction with convex hull transformation, bounding box derivation via distance transform, promptable refinement with FastSAM, and final contour selection. The experiments conducted used 633 images from the Pl@ntLeaves database: 333 images for model development with a train/validation split of 266/67 (20% validation), and a held-out test set of 300 images. On the 300-image test set, the proposed framework achieved superior results (Precision = 0.966, Recall = 0.945, Intersection over Union (IoU) = 0.917, Dice = 0.953, HD95 = 27.859), outperforming DeepLabV3 and CLIPSeg. These findings confirm that combining U-Net's fine-grained feature extraction with FastSAM's efficient prompt-based refinement provides a robust and scalable solution for plant phenotyping and precision agriculture, particularly by enhancing boundary accuracy in complex natural scenes.
Why it matches plant phenotyping methods植物葉の画像セグメンテーション手法を開発し、独立テストデータで既存手法と比較検証しており、葉形態の取得を目的とするフェノタイピング手法が中心である。
abstractThis paper proposes a hybrid six-stage framework that integrates U-Net with the Fast Segment Anything Model (FastSAM) to achieve accurate and efficient leaf segmentation.
Reproduction assets foundThe paper's only paper-specific asset is the Pl@ntLeaves leaf image dataset (with ground-truth masks) used for all experiments; the authors explicitly state it is publicly available at the LIRIS REVERES databases page. No author analysis code, trained model checkpoints, or supplementary data deposits are mentioned.Dataset · publicsegmentation, real-time field deployment,
and integration with broader phenotyping pipelines.
ACKNOWLEDGMENT
This research was financially supported by the Silpakorn
University Research, Innovation, and Creative Fund.
DATA AVAILABILITY STATEMENT
The dataset used in this study, the Pl@ntLeaves database, is
publicly available at:
https://liris.univ-lyon2.fr/reves/content/en/databases.php.REFERENCES
[1] J. W. Abe, J. Ilao, and G. Foliente, "Promptable Leaf Segmentation in
Plant Phenotyping: Research Perspectives and Challenges," in 2024 30th
International Conference on Mechatronics and Machine Vision in
Practice, Leeds, UK, Oct. 2024, pp. 1–6,
https://doi.org/10.1109/M2VIP62491.2024.10745998.
[Open asset ↗Pl@ntLeaves databasepdf-raw-page:10 lines:1-83Code / dataset availability confirmedCrossref · checked 6 Sept 2026
Technology-driven agriculture, or precision agriculture (PA), is indispensable in the contemporary world due to its advantages and the availability of technological innovations. Particularly, early disease detection in agricultural crops helps the farming community ensure crop health, reduce expenditure, and increase crop yield. Governments have mainly used current systems for agricultural statistics and strategic decision-making, but there is still a critical need for farmers to have access to cost-effective, user-friendly solutions that can be used by them regardless of their educational level. In this study, we used four apple leaf diseases (leaf spot, mosaic, rust and brown spot) from the PlantVillage dataset to develop an Automated Agricultural Crop Disease Identification System (AACDIS), a deep learning framework for identifying and categorizing crop diseases. This framework makes use of deep convolutional neural networks (CNNs) and includes three CNN models created specifically for this application. AACDIS achieves significant performance improvements by combining cascade inception and drawing inspiration from the well-known AlexNet design, making it a potent tool for managing agricultural diseases. AACDIS also has Region of Interest (ROI) awareness, a crucial component that improves the efficiency and precision of illness identification. This feature guarantees that the system can quickly and accurately identify illness-related areas inside images, enabling faster and more accurate disease diagnosis. Experimental findings show a test accuracy of 99.491%, which is better than many state-of-the-art deep learning models. This empirical study reveals the potential benefits of the proposed system for early identification of diseases. This research triggers further investigation to realize full-fledged precision agriculture and smart agriculture.
Why it matches plant phenotyping methods植物葉の病徴領域を画像から検出・分類する深層学習手法の開発が中心であり、植物の病害状態を直接推定するため、方法論文として採用する。
abstractwe used four apple leaf diseases (leaf spot, mosaic, rust and brown spot) from the PlantVillage dataset to develop an Automated Agricultural Crop Disease Identification System (AACDIS), a deep learning framework for identifying and categorizing crop diseases.
Reproduction assets foundThe paper's phenotyping inputs are PlantVillage apple leaf disease images (leaf spot, mosaic, rust, brown spot), explicitly cited with a public GitHub URL; no author analysis code or trained model checkpoints are released.Dataset · publicPlantVillege Dataset. Available online: https://github.com/spMohanty/PlantVillage-Dataset/tree/master/raw/color (accessed
on 1 December 2024).Open asset ↗PlantVillage-Dataset · raw/colorpdf-page:21 lines:1-59Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Accurate detection of apple leaf diseases remains a critical challenge in precision agriculture, where complex field conditions and subtle symptom variations often degrade model performance. This paper introduces a novel hybrid architecture combining enhanced spatial attention with edge-aware feature extraction to improve disease classification robustness. The proposed model integrates a multi-scale feature fusion module to capture both local lesion patterns and global contextual cues, while a lightweight attention mechanism dynamically prioritizes disease-relevant regions. Experiments on a curated dataset of 12,350 apple leaf images demonstrate the effectiveness of proposed approach, achieving 96.7% classification accuracy across six disease categories - a significant improvement over baseline models like EfficientNet-B4 (94.1%) and ResNet-50 (93.8%). The system particularly excels in detecting early-stage infections, showing 15% higher precision for subtle scab lesions compared to existing methods. With only 3.2 million parameters, the model maintains practical deployment potential for edge devices in orchard environments. These advances address key limitations in current vision-based plant disease detection systems while balancing accuracy and computational efficiency for real-world agricultural applications.
Why it matches plant phenotyping methodsリンゴ葉画像から病害状態を推定するCNN手法の開発と評価が研究の中心であり、植物病害表現型の画像ベース計測に該当する。
abstractThis paper introduces a novel hybrid architecture combining enhanced spatial attention with edge-aware feature extraction to improve disease classification robustness.
Reproduction assets foundThe paper's Data Availability statement points to the public Kaggle Apple Leaf Disease Dataset (ALDD-v2) used for all training/evaluation, matching an allowed URL. No author code or model checkpoints are shared.Dataset · publicThe datasets analysed during the current study is publicly available in the Kaggle repository at https://www.kaggle.com/dsv/2068940.Open asset ↗Kagglehtml-lines:505-539Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
This dataset was generated to support research investigating the use of hyperspectral reflectance for the estimation of foliar nitrogen (N) and phosphorus (P) concentrations in apple ( Malus domestica ) trees. This article and the dataset it describes accompany an original research article submitted to Computers and Electronics in Agriculture entitled "Investigating the limits of spectroscopy for the estimation of foliar N and P in apple" [1]. Data were collected from a controlled potted experiment involving 150 'Golden Delicious' apple trees grown under varying nutrient supply regimes, including full nutrient supply, nitrogen- and phosphorus-deficient treatments, and trees infected with ' Candidatus Phytoplasma mali'. The experiment was conducted over the 2023 growing season at the Laimburg Research Centre in South Tyrol, Italy. All data and the accompanying code for its analysis is freely available in the associated GitHub repository [2]. Spectral data were collected using the Spectral Evolution SR-3500 field spectroradiometer with an attached leaf clip, producing high-resolution hyperspectral reflectance profiles (350-2500 nm) from the adaxial surface of fully expanded leaves. A total of 1189 leaf spectra were recorded and were matched to chemically analysed leaf samples. Corresponding foliar N and P concentrations (and others) were determined through laboratory analysis using the Dumas combustion method for nitrogen and ICP-OES following acid digestion for phosphorus. The dataset includes metadata detailing tree treatments, sampling dates, infection status, and shoot growth metrics. Additionally, R scripts used for data processing, spectral pre-treatment (including multiplicative scatter correction and Savitzky-Golay derivatives), feature selection (VIP and mRMR), and model development are provided. The dataset is suitable for reuse in the development and benchmarking of spectral models for nutrient estimation, especially in the context of field-based or remote sensing applications in horticulture. Its wide range of foliar nutrient values, inclusion of multiple physiological stresses, and detailed documentation make it a valuable resource for researchers working in precision agriculture, plant phenotyping, chemometrics, and hyperspectral data analysis.
Why it matches plant phenotyping methodsリンゴ葉のN・P濃度という植物生理形質を対象に、ハイパースペクトル反射データ、化学分析値、前処理・モデル開発コードを含む再利用可能なデータセットであり、植物フェノタイピング手法の開発・ベンチマークに直接資する。
abstractThis dataset was generated to support research investigating the use of hyperspectral reflectance for the estimation of foliar nitrogen (N) and phosphorus (P) concentrations in apple
Reproduction assets foundThe authors publicly release the paper's own hyperspectral leaf spectra (.sed files), matched foliar N/P concentrations, metadata, and R analysis scripts via a GitHub repository (also archived with Zenodo DOI 10.5281/zenodo.15600557), with explicit public availability and no registration required.Dataset · publicData accessibility
Repository name: Github
Data identification number: DOI 10.5281/zenodo.15600557
Direct URL to data: https://github.com/HyperspectralCameron/Investigating-the-Limits-of-Spectroscopy-for-the-Estimation-of-Foliar-N-and-P-in-Apple.gitInstructions for accessing these data: All data and code are publicly available through the GitHub repository listed above. The repository includes raw spectral files (.sed), metadata files, and R scripts for pre-processing, modelling, and visualisation. No registration or authentication is required.Open asset ↗GitHub · DOI 10.5281/zenodo.15600557html-lines:84-123Code · publicAll data and the accompanying code for its analysis is freely available in the associated GitHub repository [2].Open asset ↗GitHubhtml-lines:1-83Code / dataset availability confirmedCrossref · checked 6 Sept 2026
CoffeeMultispectral / hyperspectralLeafClassificationPhysiological trait estimationWater status / transpiration
Water potential is an important indicator used to study water relations in plants, as it reflects the level of hydration in their tissues. There are different numerical variables that describe plant properties and can be acquired from leaf reflectance. The objective of this study was to estimate water potential in coffee plants using spectral variables. For this, a range of wavelengths that provided analytical flexibility was used. After this, machine learning techniques were employed to build data-driven models. The dataset used presents spectral characteristics (wavelength) of coffee plants, collected through the CI-710 Mini-Leaf Spectrometer equipment and also the water potential of each coffee plant, measured by the Scholander Chamber equipment. The dataset was divided into two crop management groups: irrigated and rainfed. Four machine learning techniques were implemented: Multi-Layer Perceptron (MLP), Decision Tree, Random Forest and K-Nearest Neighbor (KNN). The implementation of machine learning techniques followed two distinct strategies: regression and classification. The results indicate that the decision tree-based model demonstrated superior performance under irrigated conditions for regression tasks. In contrast, the KNN technique achieved the best performance for classification. Under rainfed conditions, the MLP model outperformed the other techniques for regression, while the Random Forest method exhibited the highest accuracy in classification tasks. While no hardware prototype was developed, the machine learning-based methods presented here suggest a possible pathway toward future intelligent, user-friendly, and accessible sensing technologies for coffee plantations.
Why it matches plant phenotyping methodsコーヒー植物の葉スペクトルから水ポテンシャルという生理形質を機械学習で推定・分類する手法が研究の中心であり、植物フェノタイピング手法の開発・評価に該当する。
abstractThe objective of this study was to estimate water potential in coffee plants using spectral variables.
Reproduction assets foundThe paper's Data Availability Statement explicitly states that the study's datasets (coffee leaf spectral reflectance and water potential measurements) and the MATLAB analysis codes are publicly available at the authors' UFLA repository, which is an allowed URL. This is a paper-specific, public, actionable asset.Dataset · publicData Availability Statement: The datasets and MATLAB codes used in this study are available at
http://www.aia.ufla.br/home/filesdatasets/, accessed on 27 November 2025.Open asset ↗aia.ufla.brpdf-page:18 lines:1-54Code / dataset availability confirmedOpenAlex · bioRxiv · Crossref · checked 14 Sept 2026
Abstract Leaf appearance is a crucial plant phenotype. However, traditional methods for extracting this information are inefficient, limiting its full utilization. Deep learning based on convolutional neural networks (CNNs) enables us to capture previously inaccessible information from images. In this study, we made the surprising discovery that the leaf appearance of each individual plant is unique. Using deep learning, leaves from one plant could be efficiently distinguished from those of another plant of the same species and cultivar. We term this phenomenon the “ Plant Face ” and suggest the potential to develop a “plant face recognition system,” analogous to human facial recognition. We also applied similar methods to study the relationship between leaflet appearance and their position on compound leaves, leaf bilateral symmetry, and differences in leaves from twining stems with different chirality. These results collectively indicate that plant genetic characteristics, growth conditions, and developmental features can be stored within their appearance. With appropriate decoding, leaf appearance is poised to play an increasingly important role in phenomics. Significance The saying “no two leaves in the world are identical” holds philosophical significance, as such variation encompasses considerable contingency and randomness. Here, we assert that no two trees have identical leaves ; meaning that even for plants of the same species and cultivar, the leaf morphology of each individual plant is distinct at the population level, even though single leaves may overlap in appearance. Genetic, environmental, and developmental information is recorded in some manner within the phenotypic appearance of leaves. With advancements in computational technologies like artificial intelligence, this information can now be decoded. Highlights The leaves of each individual plant are statistically unique. The relationship between leaflet appearances in compound leaves hints at their developmental patterns. Leaves are not necessarily bilaterally symmetric in a statistical sense. Leaves from stems with different chirality (twining direction) exhibit distinct appearances.
Why it matches plant phenotyping methods葉画像から植物の個体差や形態情報を深層学習で抽出・識別する手法が研究の中心であり、植物フェノタイピングへの応用を明示している。
abstractLeaf appearance is a crucial plant phenotype. However, traditional methods for extracting this information are inefficient
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicCodes are available at git-hub.Open asset ↗pdf-page:13 lines:1-54Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Cotton has, in recent years, become one of the most important cash crops worldwide while being impacted in yield from leaf disease which generally goes unnoticed in the early stage. Detection methods depend on manual efforts producing slow processes and human errors. Automated detection methods establish low accuracies, limited scalability and real time applications. To tackle the research issue, this study proposes the CLD-Net which stands for Cotton Leaf Disease Detection Network a novel deep learning-based framework which combines Faster-RCNN and YOLOv5 algorithms into a single action to achieve ultimately real time detection of accurate diseases the combination helps identify both the high detection speed of YOLOv5 along with Faster-RCNN regional proposal accuracy. The new method is that the compilation of these two modern object detection methods has been compiled and designed specifically for detecting leaf disease across varying environmental conditions. Notable contributions to this method include increases in classification accuracy, processing speed, real time detection making these methods suitable for farmers agronomists and sensor deployment. CLD-Net integrates YOLOv5 and Faster R-CNN, combining real-time detection capability with precise classification, to deliver robust cotton leaf disease identification. Experimental validation on a curated dataset of cotton leaf images demonstrates the superiority of CLD-Net, achieving an accuracy of 96.7%, which surpasses that of traditional models. These results confirm the potential of the proposed approach to revolutionize crop disease detection, leading to timely intervention and increased yield.
Why it matches plant phenotyping methods綿花葉の病害状態を画像から推定する深層学習手法を開発し、画像データセットで性能検証しており、植物フェノタイピング手法が中心です。
abstractthis study proposes the CLD-Net which stands for Cotton Leaf Disease Detection Network a novel deep learning-based framework which combines Faster-RCNN and YOLOv5 algorithms
Reproduction assets foundThe paper's Data Availability statement points to a public Kaggle cotton leaf disease image dataset used for the CLD-Net experiments; no author code or trained model deposit with a public URL is provided despite a mention of 'reproducible code and trained models'.Dataset · publicThe data used in this research are available in the following links: https://www.kaggle.com/datasets/seroshkarim/cotton-leaf-disease-dataset.Open asset ↗Kaggle · seroshkarim/cotton-leaf-disease-datasethtml-lines:541-573Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Field / plotLeafStem / branchMorphology / geometry measurementLeaf traitsStress response / toleranceWater status / transpiration
Background Plant functional traits provide key information about species' ecological strategies and their responses to environmental disturbances such as fire. This dataset documents 14 morpho-functional traits of leaves (specific leaf area, leaf water content and leaf dry matter content), stems (maximum height, bark thickness, diameter at 40 cm, wood density, stem water content and stem dry matter content), one regenerative trait (resprouting capacity), as well as fire-related traits (ignition time, flaming time and flammability) and growth form in 50 woody plant species (27 trees, 22 shrubs and one liana) inhabiting a pine-oak forest in the "Barranca del Cupatitzio" National Park (BCNP), located in Uruapan, Michoacán, Mexico. This dataset is formatted according to the Darwin Core Archive standard and is publicly available for use. New information This dataset is standardised under the Darwin Core framework. It includes 14 morpho-functional and fire-related traits. The data were obtained from 50 woody species with a diameter at breast height (DBH) > 2.5 cm (27 trees, 22 shrubs and one liana), in a pine-oak forest located in the western Trans-Mexican Volcanic Belt, in the Municipality of Uruapan, Michoacán, Mexico. Here, we report flammability-related traits for these species for the first time. The collection of biological material and the measurement of functional traits followed internationally recognised protocols, ensuring methodological consistency and facilitating integration with other global datasets. The dataset includes values for flammability, ignition time, flaming time, specific leaf area, wood density, stem water and dry matter content, bark thickness, leaf water and dry matter content, maximum height, stem diameter at 40 cm above the ground, plant growth form and resprouting capacity. This information is particularly valuable for studies in functional ecology, ecological restoration, the dynamics of woody plant communities and fire management in temperate, fire-prone ecosystems.
Why it matches plant phenotyping methods植物の形態・機能・火災関連形質を体系的に収集し、Darwin Coreで標準化した再利用可能なデータセットであり、形質測定とデータ提供が中心である。
abstractThis dataset documents 14 morpho-functional traits of leaves
Reproduction assets foundThe paper is a data paper whose own trait/flammability dataset is deposited publicly on GBIF via DOI 10.15468/46f8xe, explicitly linked as the data package for this study's measurements.Dataset · publiche Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits use, distribution and reproduction in any medium, provided the original authors are properly credited.
Data resources
Data package title
Functional traits related to fire in woody species from Barranca del Cupatitzio National Park
Resource link
https://doi.org/10.15468/46f8xe
Number of data sets
2
Data set 1.
Data set name
occurrence.txt
Data format
Darwin Core
Data set 1.
Column label
Column description
id
Unique identifier for each occurrence.
institutionID
The identifier for the institution having custody of the specimens.
institutionCode
Full name of the institution having custody of the specimeOpen asset ↗10.15468/46f8xelines:87-297Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
Abstract High‐throughput and noninvasive phenotyping methods are promising technology for improving efficiency in plant research and breeding. In this study, we evaluated the performance of a digital phenotyping system (DPS) based on three‐dimensional (3D) model reconstruction for quantifying key growth traits in rice ( Oryza sativa ). The DPS was used to estimate plant height, biomass, color, leaf morphology, and tiller angle in four rice varieties (Koshihikari, Nipponbare, PL9, and Tachiaoba). The results show high accuracy and correlation between manually measured and DPS‐derived traits. Notably, the 3D volume analysis can quantify biomass accumulation and growth dynamics and revealed distinct differences among varieties. The strong correlation between the green‐red normalized difference index (a red‐green‐blue‐based index) and soil plant analysis development also demonstrated the viability of the system in monitoring leaf color without using a multispectral instrument. The analysis also captured growth patterns over time, including canopy development and senescence, which are often challenging to quantify through manual measurements alone. Furthermore, the tiller angle estimation derived from DPS provided an alternative method to plant architecture evaluation, demonstrating its potential for use in breeding programs aimed to optimize canopy structure. These findings establish DPS as a reliable and scalable tool for a digital phenotyping platform that enables comprehensive trait analysis with reduced labor and increased precision and the capability to continuously monitor plant growth and biomass accumulation. This study shows the potential of this novel digital tool for automating manual measurements, which can increase efficiency and expedite research and breeding in rice and other crops.
Why it matches plant phenotyping methods3Dモデル再構築に基づくデジタル表現型解析システムを開発・評価し、イネの複数形質を手測定と比較検証しているため、方法が研究の中心です。
abstractwe evaluated the performance of a digital phenotyping system (DPS) based on three‐dimensional (3D) model reconstruction for quantifying key growth traits in rice
Reproduction assets foundThe paper's data availability statement explicitly says the analysis code is openly available on GitHub at the authors' repository Rice_VTGa.O, which contains the digital phenotyping/leaf-tracing analysis code for this study. No phenotype dataset or image deposit is stated.Code · publicGrant Number 39 [2023] and 38 [2024]),
and Microbiome and Metabolome Control Project, University
of Miyazaki, Japan.
C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare no conflicts of interest.
DATA AVA I L A B I L I T Y S TAT E M E N T
Codes used for analysis in this study are openly available on
GitHub at https://github.com/sandysan42/Rice_VTGa.O RC I D
SorawichPongpiyapaiboon https://orcid.org/0000-0002-9314-8375
Kenji Aoki https://orcid.org/0000-0001-7003-1994
MasatsuguHashiguchi https://orcid.org/0000-0003-0637-2780
RyoAkashi https://orcid.org/0000-0002-5651-8285
Yuji Kishima https://orcid.org/0000-0002-0942-3371
Hidenori Tanaka https://orcid.org/0000-0002-4237-8154Open asset ↗https://github.com/sandysan42/Rice_VTGa.O · Rice_VTGa.Opdf-raw-page:13 lines:1-84Code / dataset availability confirmedCrossref · OpenAlex · Europe PMC · checked 6 Sept 2026
RiceWheatField / plotMesh / voxelNeRF / 3D Gaussian SplattingLiDAR / point cloudPanicle / ear / spikeLeafRootSeed / grain
Advanced plant phenotyping technologies are vital for trait improvement and accelerating intelligent breeding. Due to the species diversity of plants, existing methods heavily rely on large-scale high-precision manually annotated data. For self-occluded objects at the grain level, unsupervised methods often prove ineffective. This study proposes IPENS, an interactive unsupervised multi-target point cloud extraction method. It utilizes radiance field information to lift 2D masks, segmented by SAM2 (Segment Anything Model 2), into 3D space for target point cloud extraction. A multi-target collaborative optimization strategy addresses the challenge of segmenting multiple targets from a single interaction. On a rice dataset, IPENS achieves a grain-level segmentation mean Intersection over Union (mIoU) of 63.72%. For phenotypic trait estimation, it achieves a grain voxel volume coefficient of determination R 2 = 0.7697 (Root Mean Square Error, RMSE = 0.0025), leaf surface area R 2 = 0.84 (RMSE = 18.93), and leaf length and width prediction accuracies of R 2 = 0.97 and R 2 = 0.87 (RMSE = 1.49 and 0.21). On a wheat dataset, IPENS further improves segmentation performance to a mIoU of 89.68%, with exceptional phenotypic estimation results: panicle voxel volume R 2 = 0.9956 (RMSE = 0.0055), leaf surface area R 2 = 1.00 (RMSE = 0.67), and leaf length and width predictions reaching R 2 = 0.99 and R 2 = 0.92 (RMSE = 0.23 and 0.15). Without requiring annotated data, IPENS rapidly extracts grain-level point clouds for multiple targets within three minutes using single-round image interactions. These features make IPENS a high-quality, non-invasive phenotypic extraction solution for rice and wheat, offering significant potential to enhance intelligent breeding.
Why it matches plant phenotyping methods植物形質抽出のためのNeRF-SAM2融合手法を開発し、作物データセットで分割性能と形質推定精度を検証しているため、方法開発・検証が中心である。
abstractThis study proposes IPENS, an interactive unsupervised multi-target point cloud extraction method.
Reproduction assets foundThe paper's analysis code is publicly available on GitHub. The rice/wheat MMR/MMW phenotype datasets (multi-view images, point clouds, annotations) are only available upon reasonable request, so they are not public.Code · publicCode is available at https://github.com/Vincent-Songwentao/IPENS-Code.git .Open asset ↗https://github.com/Vincent-Songwentao/IPENS-Code.gitlines:472-496Code / dataset availability confirmedCrossref · OpenAlex · Europe PMC · checked 6 Sept 2026
With the continuous progress in micro-optical machine technology, miniature spectral imaging devices have been rapidly developed; however, three-dimensional (3D) imaging measurement technology has become increasingly mature and widely used. The evolution of these technologies has established a robust foundation for the integration of three-dimensional imaging and spectral information. To achieve accurate alignment between 3D data and spectral information to obtain a more comprehensive spectral representation of objects in 3D space, we developed a binocular multispectral stereo imaging (BMSI) system. This system acquires images in synchrony with a binocular multispectral imager, thereby ensuring accurate alignment between 3D data and spectral data at the pixel level and facilitating the construction of a four-dimensional (4D) dataset. The segmentation of leaf regions from shadow backgrounds in two distinct plant species was achieved through optimal band fusion and hue-saturation value (HSV) color space transformation, significantly improving the segmentation accuracy, processing efficiency, and robustness across different plant species. A systematic evaluation was conducted to quantify the reconstruction precision and system stability at different measurement distances. The designed system acquired 4D image spectral data with plants as the objects to be tested. The distribution characteristics of chlorophyll (Chl) on the 3D surface of plants were obtained by first-order derivatives of the spectral data and the normalized difference red edge (NDRE) index. This technique provides a new means for plant phenotyping research and a more effective technical approach for the digitalization and precision monitoring of the agricultural industry.
Why it matches plant phenotyping methods植物の3D・マルチスペクトル画像取得、葉領域分割、再構成精度・安定性評価を中核とする新規フェノタイピングシステムの開発研究である。
abstractwe developed a binocular multispectral stereo imaging (BMSI) system.
Reproduction assets foundThe paper's data availability statement explicitly deposits raw data and essential source code for the BMSI plant phenotyping analysis on a public GitHub repository.Code · publicThe raw data and the essential parts of the source code have been uploaded to Github: https://github.com/wwxsoul1234/BMSI/tree/master.Open asset ↗wwxsoul1234/BMSI · BMSIhtml-lines:278-306Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Overcoming the strong chlorophyll background poses a significant challenge for measuring and optimizing plant growth. This research investigates the novel application of specialized quantum light emitters introduced into intact leaves of tobacco ( Nicotiana tabacum ), a well-characterized model plant system for studies of plant health and productivity. Leaves were harvested from plants cultivated under two distinct conditions: low light (LL), representing unhealthy leaves with reduced photosynthesis and high light (HL), representing healthy leaves with highly active photosynthesis. Higher-order correlation data were collected and analyzed using machine learning (ML) techniques, specifically a Convolutional Neural Network (CNN), to classify the photon emitter states. This CNN efficiently identified unique patterns and created distinct fingerprints for Nicotiana leaves grown under LL and HL, demonstrating significantly different quantum profiles between the two conditions. These quantum fingerprints serve as a foundation for a novel unified analysis of plant growth parameters associated with different photosynthetic states. By employing CNN, the emitter profiles were able to reproducibly classify the leaves as healthy or unhealthy. This model achieved high probability values for each classification, confirming its accuracy and reliability. The findings of this study pave the way for broader applications, including the application of advanced quantum and machine learning technologies in plant health monitoring systems.
Why it matches plant phenotyping methods量子発光体による葉の光子プロファイル取得とCNN解析を組み合わせ、光合成状態および植物の健康状態を分類する手法が研究の中心である。
abstractThis CNN efficiently identified unique patterns and created distinct fingerprints for Nicotiana leaves grown under LL and HL
Reproduction assets foundThe article's Data availability statement explicitly deposits the paper's time-tagged photon correlation data (the raw measurements underlying the quantum fingerprinting and CNN analysis) in the Dryad Digital Repository, a public, paper-specific dataset.Dataset · publicay: conceptualization, funding acqui-
sition, supervision, project administration, visualization,
writing – original draft, writing – review & editing.
Conflicts of interest
There are no conflicts to declare.
Data availability
Data for this article, including time tangled photon data, are
available at Dryad Digital Repository at https://doi.org/10.5061/dryad.1rn8pk15f.Supplementary information is available. See DOI: https://
doi.org/10.1039/d5an00326a.
Acknowledgements
This research was funded in part by the Faculty Industry
Applied Research (FIAR) program, by the University at
Buffalo’s Center of Excellence in Materials Informatics.
References
1 E. Murchie and T. Lawson, J. Exp. Bot., 2013, 6Open asset ↗Dryad Digital Repository · 10.5061/dryad.1rn8pk15fpdf-raw-page:7 lines:1-96Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Lettuce ( Lactuca sativa ), a widely cultivated leafy vegetable, is highly susceptible to bacterial and fungal infections that severely reduce yield and quality. Rapid and accurate disease identification is therefore essential for precision agriculture and sustainable crop management. This study proposes Efficient-FBM-FRMNet, a modular deep learning framework for automated lettuce disease detection. The model integrates EfficientNetB4 with dilated convolutions, a Feature Bottleneck Module (FBM) for redundancy reduction, a Reasoning Engine for higher-order semantic inference, and a Feature Refinement Module (FRM) for enhanced generalization. The framework was trained and validated on a publicly available dataset of 2,813 lettuce leaf images (bacterial, fungal, and healthy classes) using stratified 5-fold cross-validation. The proposed Efficient-FBM-FRMNet achieved an overall accuracy of 97.5%, outperforming baseline CNNs such as EfficientNetB4, ResNet50, and DenseNet121. It demonstrated superior precision (96.0%), recall (96.6%), and F1-score (97.0%), confirming its robustness and consistency across multiple folds. Statistical significance analysis (p
Why it matches plant phenotyping methodsレタス葉画像から病害状態を推定する深層学習フレームワークを開発し、公開データセット上で交差検証・ベースライン比較により性能を評価しており、植物表現型取得手法が中心である。
abstractThis study proposes Efficient-FBM-FRMNet, a modular deep learning framework for automated lettuce disease detection.
Reproduction assets foundThe paper's phenotyping measurements are based entirely on a public Kaggle lettuce plant disease image dataset (2,813 images, bacterial/fungal/healthy), cited with an explicit public URL matching an allowed URL. No author code or model checkpoints are stated as available.Dataset · publiceelwal P.
Dhiman P.
Gulzar Y.
Kaur A.
Wadhwa S.
Onn C. W.
( 2024 ).
A systematic review of deep learning applications for rice disease diagnosis: current trends and future directions
. Front. Comput. Sci.
6 . doi:
10.3389/fcomp.2024.1452961
Shaha, S.
.(n.d.) Lettuce plant Disease Dataset [Data set]. Kaggle. Available online at: https://www.kaggle.com/datasets/santoshshaha/lettuce-plant-disease-dataset (Accessed March 12, 2025 ).
Shoaib M. A.
Lai K. W.
Chuah J. H.
Hum Y. C.
Ali R.
Dhanalakshmi S.
. ( 2022 ).
Comparative studies of deep learning segmentation models for left ventricle segmentation
. Front. Public Health
10 , 981019 . doi:
10.3389/fpubh.2022.981019
, PMID:
36091529
PMC9453312
SuOpen asset ↗Kaggle · lettuce-plant-disease-datasetlines:990-1174Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Field / plotLeafClassificationDisease symptoms / severity
This Data Descriptor presents the Jute Diseases Image Dataset; a curated collection of 1390 high-resolution images aimed at supporting the development of machine learning models for timely identification and accurate diagnosis of jute (Corchorus) plant diseases. The dataset is categorized into five classes: Dieback (300), Holed (300), Mosaic (240), Stem Soft Rot (270), and Fresh (280) representing healthy leaves. Images were captured under varied natural lighting and directional conditions across diverse jute cultivation areas to enhance model generalizability. A rigorous pre-processing pipeline was applied, including uniform resizing to 1024 × 1024 pixels and removal of duplicate images to ensure data integrity. The dataset is organized into two components: a raw, pre-processed set and an augmented train-test split version, enabling immediate use in machine learning workflows. Additionally, Grad-CAM and Guided Grad-CAM techniques were applied to sample images to visualize and validate model attention on disease-relevant regions. This resource addresses the lack of labelled jute disease imagery and supports timely disease management, particularly for stakeholders in Bangladesh and other major jute-producing regions.
Why it matches plant phenotyping methods植物病害症状を画像として収集・ラベル化したデータセットであり、病害状態の画像ベース表現型判定を支援することが中心です。
abstractThis Data Descriptor presents the Jute Diseases Image Dataset; a curated collection of 1390 high-resolution images aimed at supporting the development of machine learning models for timely identification and accurate diagnosis of jute (Corchorus) plant diseases.
Reproduction assets foundThe paper's own jute disease image dataset (1390 labeled images, raw and augmented train/test splits) is publicly deposited in Harvard Dataverse with an explicit DOI and direct URL, matching an allowed URL. No separate analysis code or trained model checkpoint is publicly released.Dataset · publicData accessibility
Repository name: Harvard Dataverse
Data identification number: https://doi.org/10.7910/DVN/FJ1DM1
Direct URL to data: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/FJ1DM1Open asset ↗Harvard Dataverse · doi:10.7910/DVN/FJ1DM1html-lines:1-91Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
In this study, we propose a multi-scale feature fusion network based on an improved RT-DETR model for the efficient detection of tomato leaf disease. Our model combines the multi-scale extended residual module by capturing contextual information at various scales and the multi-scale feature pyramid network by integrating feature information from different levels, which improves feature extraction capability and reduces the interference of complex backgrounds on feature extraction, thereby improving information transmission efficiency and the accuracy of the model. In addition, the novel loss function called adaptive focal loss (AFL) was used, which is based on traditional focal loss with the introduction of attenuation factors to focus the model's attention to high-loss features to alleviate overfitting and of dynamic weight adjustment mechanisms to focus on the more important features during the training process to improve the overall learning performance. Another significant advantage of AFL is that it can more efficiently improve the detection accuracy on imbalanced datasets than on balanced datasets. These innovations optimized the learning strategy of the model, making AP@0.50 up to 97.9% on detecting the categories of tomato diseases. In addition, this model also achieves the high detection accuracy of 85.4% on other crop diseases. These results provide valuable references for agriculture applications.
Why it matches plant phenotyping methodsトマト葉の病害状態を画像から検出する深層学習手法を提案・改良しており、植物病害表現型の抽出法が研究の中心である。
abstractwe propose a multi-scale feature fusion network based on an improved RT-DETR model for the efficient detection of tomato leaf disease.
Reproduction assets foundThe paper constructs its LAB and ENV tomato leaf disease detection datasets by selecting and adjusting samples from two public Roboflow Universe datasets (cited as refs 30 and 31), which are public image inputs directly underlying this paper's phenotyping measurements. Both Roboflow URLs are given in the reference listDataset · public7-024-01188-1
38725014
PMC11080254
29.
Sun H.
Fu R.
Wang X.
Wu Y.
Al-Absi M.A.
Cheng Z.
Chen Q.
Sun Y.
Efficient Deep Learning-Based Tomato Leaf Disease Detection through Global and Local Feature Fusion
BMC Plant Biol. 2025 25 311
10.1186/s12870-025-06247-w
40069604
PMC11895386
30.
Sujansurya
Roboflow Universe
Available online: https://universe.roboflow.com/sujansurya/tomato_object (accessed on 15 August 2024)
31.
Roboflow Universe
Available online: https://universe.roboflow.com/classificationwithyolov8/tomato-leaf-diseases-4xa5i-3ajin-kyo9w (accessed on 15 August 2024)
32.
Singh D.
Jain N.
Jain P.
Kayal P.
Kumawat S.
Batra N.
PlantDoc: A Dataset for Visual Plant Disease Detection
ProceedingOpen asset ↗Roboflow Universelines:265-465Dataset · publicng-Based Tomato Leaf Disease Detection through Global and Local Feature Fusion
BMC Plant Biol. 2025 25 311
10.1186/s12870-025-06247-w
40069604
PMC11895386
30.
Sujansurya
Roboflow Universe
Available online: https://universe.roboflow.com/sujansurya/tomato_object (accessed on 15 August 2024)
31.
Roboflow Universe
Available online: https://universe.roboflow.com/classificationwithyolov8/tomato-leaf-diseases-4xa5i-3ajin-kyo9w (accessed on 15 August 2024)
32.
Singh D.
Jain N.
Jain P.
Kayal P.
Kumawat S.
Batra N.
PlantDoc: A Dataset for Visual Plant Disease Detection
Proceedings of the Proceedings of the 7th ACM IKDD CoDS and 25th COMAD Hyderabad, India 5–7 January 2020
ACM
New York, NY, USA
2020 24Open asset ↗Roboflow Universelines:265-465Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Early and accurate detection of sugarcane leaf diseases is critical for improving crop productivity and reducing economic losses in the agricultural sector. Timely interventions enable sustainable crop management and better resource use. In this study, we propose a deep learning-based approach for sugarcane leaf disease classification that leverages a novel architecture, the Multi-scale Attention-based Dense Residual Network (MADRN). The MADRN model integrates dense residual learning and multi-scale attention mechanisms to effectively capture fine-grained, disease-specific features and address challenges related to domain variability and complex data patterns. Two datasets are used to evaluate the model: a Kaggle dataset and a blended dataset created by combining Kaggle images with those from the Bangladesh Sugarcrop Research Institute (BSRI), simulating real-world conditions. All images undergo preprocessing steps, including resizing, normalization, and data augmentation, before training. Additionally, several baseline models (CNN, VGG16, MobileNetV2, and XceptionNet) are fine-tuned and compared with the MADRN model. Experimental results demonstrate that MADRN consistently outperforms baseline models in accuracy, precision, recall, and F1-score across both datasets, achieving up to 94.78% accuracy on the Kaggle dataset and 92.25% on the blended dataset. These findings highlight MADRN's superior ability to learn discriminative features and generalize effectively across diverse data sources, making it a promising tool for precision agriculture and disease management. To facilitate practical implementation, a web-based application is developed, enabling real-time and user-friendly disease detection. This research lays a strong foundation for the development of accurate, scalable, and practical disease classification tools that can support sustainable agricultural practices.
Why it matches plant phenotyping methodsサトウキビ葉の画像から病害状態を推定する深層学習手法を開発し、複数データセットとベースラインで比較検証しているため、植物フェノタイピング手法が中心です。
abstractwe propose a deep learning-based approach for sugarcane leaf disease classification that leverages a novel architecture, the Multi-scale Attention-based Dense Residual Network (MADRN).
Reproduction assets foundThe paper's Kaggle sugarcane leaf disease image dataset (2521 images, five classes) is a public, paper-specific phenotyping image asset with an explicit URL in the Data Availability statement. The BSRI field images are only available upon request, and no author analysis code or trained model is deposited.Dataset · publicg and preparation.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data availability
The data supporting the findings of this study are publicly available and can be accessed through the following sources. Kaggle Sugarcane Leaf Disease Dataset [https://www.kaggle.com/datasets/nirmalsankalana/sugarcane-leaf-disease-dataset](https:/www.kaggle.com/datasets/nirmalsankalana/sugarcane-leaf-disease-dataset) (accessed Jan. 15, 2025). Bangladesh Sugarcrop Research Institute (BSRI): Data available upon request. For BSRI data inquiries, please contact the corresponding author, Dr. Md. Shamim Reza.
Declarations
CompetinOpen asset ↗Kagglelines:278-299Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
MaizeTomatoLeafPhysiological trait estimationWater status / transpiration
Within the soil-plant-atmosphere continuum, water movement is driven by the water potential gradients between these three domains. To have a comprehensive understanding of such water relations, an examination of how plants respond to variations in soil water availability is required. The methodologies employed for measuring water potential in leaf (Ψ leaf ) and soil (Ψ soil ) have undergone a significant evolution; transitioning from qualitative assessments to the use of high-precision digital sensors over the past few decades. The present protocol aims to provide a comprehensive, step-by-step guide from the germination phase of maize and tomato plants to the installation of two sensors that continuously monitor water potential in the leaf (PSY1 psychrometer) and in the soil (TEROS 21 matric potential sensor). Additionally, we present the code for processing the raw data files in RStudio.
Why it matches plant phenotyping methods葉の水ポテンシャルという植物生理形質を連続測定するセンサー設置手順とデータ処理コードを中心に扱うプロトコルであり、植物フェノタイピング手法が研究の中心である。
abstractThe present protocol aims to provide a comprehensive, step-by-step guide from the germination phase of maize and tomato plants to the installation of two sensors that continuously monitor water potential in the leaf (PSY1 psychrometer) and in the soil (TEROS 21 matric potential sensor).
Reproduction assets foundThe paper deposits its example water-potential datasets (soil matric potential from Teros 21, leaf water potential from PSY1, transpiration from scales) and the authors' data extraction/cleaning/analysis code on Zenodo (10.5281/zenodo.17158115), under CC0/CC-BY. A supplementary installation video is separately on ZenodDataset · public52.
PubMed Abstract | Publisher Full Text
Cotrozzi L, Couture JJ, Cavender-Bares J, et al.: Using foliar spectral properties
References
Figure 9. Example of data cleaning using the algorithm. Green is kept data and red is discarded data.
Data availability
The datasets and codes to analyze the data have been deposited
on Zenodo (https://doi.org/10.5281/zenodo.17158115, D'Agostino
(2025)).
Data are available under the terms of the Creative Commons
Zero v1.0 Universal
An additional explicative video for the psychrometer instal-
lation on leaves is available on Zenodo (https://doi.org/10.5281/zenodo.17510720, Degand et al. (2025)).
The author(s) declare that this video is released under the
CreOpen asset ↗Zenodo · 10.5281/zenodo.17158115pdf-raw-page:11 lines:1-61Code · publicat were missing,
zero, or otherwise aberrant. It was also programmed to iden-
tify and remove inverted day-night cycle patterns, as well as
values that were statistically insignificant. Figure 9 shows appli-
cations of data cleaning on the example dataset. For more
details, please check codes that have been deposited on Zenodo
(https://doi.org/10.5281/zenodo.17158115, D'Agostino, 2025).
Ethics and consent
Ethical approval and consent were not required
Figure 8. Example of the charging effects on the data recordings.
Page 10 of 18
Open Research Europe 2025, 5:363 Last updated: 19 JUN 2026Open asset ↗Zenodo · 10.5281/zenodo.17158115pdf-raw-page:10 lines:1-58Supplement · publicavailability
The datasets and codes to analyze the data have been deposited
on Zenodo (https://doi.org/10.5281/zenodo.17158115, D'Agostino
(2025)).
Data are available under the terms of the Creative Commons
Zero v1.0 Universal
An additional explicative video for the psychrometer instal-
lation on leaves is available on Zenodo (https://doi.org/10.5281/zenodo.17510720, Degand et al. (2025)).
The author(s) declare that this video is released under the
Creative Commons CC0 1.0 Universal Public Domain Dedica-
tion. This means the video is free of all copyright restrictions
and may be copied, modified, distributed, and used without
permission, including for commercial purposes.
Data are availablOpen asset ↗Zenodo · 10.5281/zenodo.17510720pdf-raw-page:11 lines:1-61Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Field / plotLeafClassificationStress / disease detectionDisease symptoms / severity
Existing disease discovery in papaya leaves is most significant in achieving yield and profitability stability in the tropics but has proven difficult in the presence of deficiencies in manual exploration and tailored crop models in crop-AI systems. Therefore, this study introduces PapayaNet, a lightweight attention-guided convolutional network specifically structured for the automated classification of six papaya leaf states, including major diseases and healthy leaves. For real-world deployment in scarce-resource farming contexts, PapayaNet adopts batch norm and hierarchical attention steps in five convolution stages and accelerates both computational celerity and discriminability. Trained on 6618 manually annotated orchard images sourced from orchards in Bangladesh at a very high resolution, it has a 98.79% classification accuracy, all of which was realized using 483,926 parameters and an average infer time of 0.01 s, which is significantly better when evaluated using EfficientNetB6, DenseNet121, and VGG16. XAI methods, including Grad-CAM and LIME, showed model decisions towards the biologically informative parts of the leaf, thus boosting interpretability and user confidence. Systematic ablation analysis also confirmed the importance of distributed attention in ensuring robust generalization towards visually similar disease classes. An in-browser diagnostic portal deployed using Gradio provides intra-browser predictive deployment and interpretability overlay in real time, thus inviting field practicability. Given its low-latency inference and minimal computational footprint, PapayaNet is well-suited for integration into edge devices and drone platforms, offering a scalable solution for real-time in-situ crop health monitoring. This study advances the field of precision agriculture by delivering a crop-specialized, explainable, and deployable AI system for sustainable management of papaya diseases.
Why it matches plant phenotyping methodsパパイヤ葉の病害・健全状態を画像から分類するCNN手法を開発し、データセット、比較評価、アブレーション、実運用ポータルまで中心的に扱っているため、植物病害フェノタイピング手法に該当する。
abstractthis study introduces PapayaNet, a lightweight attention-guided convolutional network specifically structured for the automated classification of six papaya leaf states, including major diseases and healthy leaves.
Reproduction assets foundThe paper's papaya leaf image dataset is publicly deposited on Mendeley Data with an explicit availability statement and authors' URL; no code or model checkpoint availability is stated.Dataset · publiccript.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data availability
The dataset analysed of this study, titled ”Healthy and Unhealthy Papaya Leaf Images from Bangladeshi Orchards,” is publicly available in the Mendeley Data repository at ( https://data.mendeley.com/datasets/44p8v6ywsm/1 ).
Competing interests
The authors declare no competing interests.
References
1. Sandhu, G. K. & Kaur, R. Plant disease detection techniques: A review. In 2019 International Conference on Automation, Computational and Technology Management, ICACTM 2019 34–38 (2019). 10.1109/ICACTM.2019.8776827
2.
Ngugi LC Abelwahab M AboOpen asset ↗Mendeley Datalines:622-669Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
India is an agro-based country. The major goal of agriculture is to produce disease-free healthy crops. For Indian agronomists, cotton is a profitable commercial and fibre crop, it is the world's second-biggest export crop after China. Cotton production is also affected in a negative way by high use of water, authority of soil erosion and the practice of using dangerous fertilizers and pesticides. The two greatest threats to the rapid growth of the crop are the sucking bugs and cotton diseases. Prompt detection and accurate identification of diseases is vital to ensure healthy crop growth and achieve better yields. The primary objective of this research is to build a model by implementing deep learning-based approaches to spot infections in cotton crops. Deep learning is used because of its exceptional results in classification and image processing tasks. To address this issue, we developed CottonNet-MHA a novel deep learning framework to identify pathological symptoms in cotton leaves. The model employs multi-head attention mechanisms to strengthen feature learning and highlight the diseased-affected regions. To evaluate the performance of the proposed model, five pretrained transfer learning architectures-VGG16, VGG19, InceptionV3, Xception, and MobileNet were used as benchmark models. Furthermore, Gradient-weighted Class Activation Mapping (Grad-CAM) visualization was applied to enhance the trustworthiness and interpretability of the model. A web-based application was developed to deploy the trained model for real-world applicability. The performance analysis is carried out on the developed model based on the conventional models and the results indicate that CottonNet-MHA dominates the conventional models with respect to its accuracy as well as efficiency in the detection of diseases. The use of attention mechanisms approach strengthens the model's diagnostic accuracy and overall reliability. Grad-CAM results further demonstrated that the model effectively targets diseased areas, enhancing interpretability and reliability. Discussion: The study shows that CottonNet-MHA not only automates disease detection but also enhances interpretability through Grad-CAM analysis. The developed web platform allows the model to be applied in real-world environments, supporting live disease monitoring. The proposed framework not only improves the accuracy of cotton disease diagnosis but also offers potential for extension to other crop disease detection systems.
Why it matches plant phenotyping methods綿花葉の病徴を画像から検出・分類する深層学習手法を開発し、既存モデルとの比較検証とGrad-CAMによる病変領域の解釈を行っており、植物病害状態の表現型取得が中心である。
abstractThe primary objective of this research is to build a model by implementing deep learning-based approaches to spot infections in cotton crops.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe dataset used in this work is downloaded from Akash Zade (Data Scientist) which is openly accessible and can be found at: https://drive.google.com/drive/folders/1vdr9CC9ChYVW2iXp6PlfyMOGD-4Um1ue.Open asset ↗Akash Zadehtml-lines:312-354