Accurate point-cloud segmentation of cotton organs is essential for precise phenotypic characterization. However, reliable instance segmentation of cotton bolls in field-derived point clouds remains challenging because foliage occlusion, contact between adjacent bolls and incomplete reconstruction obscure instance boundaries. Here we present RQ-PointNeXt, an end-to-end framework that directly maps input point clouds to boll instance masks within a unified trainable network. Built on PointNeXt, it incorporates relative-elevation geometric channel attention in the shallow encoder to fuse global channel context with elevation and surface-normal cues. Its Query–mask branch integrates semantic guidance, center-seeded queries and a center-aware mask prior to suppress background responses, localize instances and constrain mask extent. Hungarian matching and multitask optimization establish one-to-one query–instance assignments, whereas query-based decoding produces instance masks without external geometric clustering. We evaluated the framework on 226 field-grown cotton plants containing 720 annotated boll instances reconstructed from UAV multi-view imagery using neural radiance fields. On the held-out test set, overall accuracy, mean class accuracy and mean intersection over union reached 0.8942, 0.8980 and 0.8076, respectively. AP25, AP50 and AP75 were 0.7359, 0.5585 and 0.2777, yielding an mAP25/50/75 of 0.5240. The framework provides instance-level outputs for boll counting and spatial analysis in high-throughput field phenotyping.
Why it matches plant phenotyping methods綿花ボールの点群インスタンス分割を開発・評価し、計数や空間解析に利用可能な植物器官表現型を抽出する方法が中心である。
abstractHere we present RQ-PointNeXt, an end-to-end framework that directly maps input point clouds to boll instance masks within a unified trainable network.
Reproduction assets foundThe paper's own field cotton point-cloud dataset (226 plants, 720 annotated bolls) is only available upon request. However, the authors directly used the public UGA-BSAIL Cotton Plants with Foliage point-cloud dataset (with their added boll instance annotations) as an evaluation benchmark for RQ-PointNeXt, and it is公开发Dataset · publict to the pointwise overlap between
predicted and ground-truth instances.
To further evaluate the proposed method under conditions of relatively high point-
cloud completeness, experiments were conducted using the public UGA-BSAIL Cot-
ton Plants with Foliage dataset. The point-cloud data are publicly available through
Figshare (https://figshare.com/projects/Cotton_plant_with_foliage/258065, accessed on
8 September 2026), while the associated code and documentation are hosted on GitHub
(https://github.com/UGA-BSAIL/Cotton_plants_with_foliage, accessed on 8 September
2026). The dataset contains relatively complete cotton plant point clouds, surface-normal
attributes, and semantic labels distinOpen asset ↗Figshare · Cotton_plant_with_foliage/258065pdf-raw-page:16 lines:1-52Code / 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 confirmedOpenAlex · Europe PMC · checked 15 Sept 2026
Boll-opening concentration is critical for mechanical cotton harvesting, yet it is still assessed mainly by manual records and single time-point indicators that miss temporal dynamics. To bridge the lack of a unified workflow linking vision foundation models, multi-temporal boll-opening monitoring, and harvest decision-making, we developed a cross-scale UAV high-throughput phenotyping framework centered on DINO-BollGX. DINO-BollGX couples a DINO v3 backbone, a RetinaNet detection head, and an adaptive refinement-and-suppression module for robust open-boll detection under complex field conditions. Using multi-temporal UAV imagery collected over two years for 383 cultivars, we reconstructed plot-scale time series of open-boll counts, derived dynamic features describing progression and intensity changes, and proposed a Cotton Boll-Opening Temporal Stability Index (CTSI) to quantify boll-opening rhythm and concentration; CTSI was further integrated with a time-based risk function to generate harvest decision curves. Under unified data and training settings, DINO-BollGX achieved precision = 0.91, F1 = 0.88, and AP@0.50 = 0.80, and provided accurate boll-count estimation (R 2 =0.98; MAE=3.10), outperforming YOLOv11, YOLOv12, YOLOv13, and RT-DETR. On an independent cross-year dataset acquired at 5 m altitude, it obtained precision = 0.98 and F1 = 0.87. An internal consistency analysis showed that CTSI had the expected negative association with Window_days (r = −0.83) and positive associations with Max_count (r = 0.78) and the boll-opening efficiency index (r = 0.92), reflecting the co-occurrence of temporal compactness and main-phase opening intensity in the cultivar population. CTSI ranged from −2.72 to 4.69 across cultivars, enabling identification of highly synchronized boll-opening. Harvest decision curves indicated that the relative net income index peaked at day 67 after the first observation and a compact optimal harvest window near the end of monitoring; on a fixed harvest date, Kuche 130292 (CTSI=4.69) produced 486 open bolls versus 182 for Xinluzao 36 (CTSI=0.53) and 90 for Andizhan-60 (CTSI=-2.72). Overall, the framework integrates dynamic boll-opening phenotyping with harvest timing optimization, supporting scalable cultivar screening and mechanization-ready deployment, with potential extension to harvest decision scenarios in other crops.
Why it matches plant phenotyping methodsUAV画像と基盤モデルを用いて綿花の開絮を検出・定量し、時系列表現型指標を構築・検証する方法が研究の中心であるため。
abstractwe developed a cross-scale UAV high-throughput phenotyping framework centered on DINO-BollGX.
Reproduction assets foundThe paper publicly releases its cotton boll-opening UAV image dataset (3638 patches, 94,774 YOLO-format bounding-box annotations) on GitHub, directly supporting the paper's phenotyping analysis. No author analysis code or trained model checkpoints are explicitly deposited.Dataset · publicentary information for evaluating cross-scale detection performance and characterizing macroscopic spatial patterns. The 5 m imagery acquired on 18 Sept 2024 is used exclusively for cross-year generalization assessment. All cropped images and the corresponding YOLO-format annotation files have been publicly released on GitHub ( https://github.com/mianchen0529/cotton-boll-dataset/tree/main ) to facilitate further research on cotton phenotyping, agricultural remote sensing, and intelligent analytics.
2.3.
Model construction
2.3.1.
Overall architecture of the DINO-BollGX network
The proposed DINO-BollGX network consists of four stages: image preprocessing, feature extraction, object prediction,Open asset ↗mianchen0529/cotton-boll-datasetlines:63-74Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Cross-sectional analysis is considered the reference method for measuring cotton fiber fineness and maturity; however, its widespread use has been limited by labor-intensive sample preparation and manual image analysis. The objective of this study was to develop and validate a reproducible deep-learning-based workflow for automated segmentation and quantitative analysis of cotton fiber cross-sections. A total of 249 composite light microscopy images of cotton fiber cross-sections were collected and manually annotated to generate training and validation datasets. A YOLO11m instance segmentation model was developed to identify cotton fiber and lumen regions and automatically extract quantitative traits, including fiber area, lumen area, fiber perimeter, and lumen perimeter. The workflow integrates automated image segmentation, post-processing, quantitative trait extraction, and data export to facilitate reproducible cotton fiber phenotyping. Model performance was evaluated using mean Average Precision (mAP), and workflow outputs were validated against Adobe Photoshop using descriptive comparisons of six cross-sectional traits. The model achieved Box mAP50 scores of 0.984 for cotton fiber regions and 0.789 for lumen regions, demonstrating high segmentation accuracy. The automated workflow substantially reduced manual analysis time while producing measurements with central tendencies comparable to those obtained using Adobe Photoshop. To facilitate reproducibility and adoption, the workflow, trained model weights, and supporting documentation are publicly available through GitHub and a Hugging Face web application. The workflow substantially increases analytical throughput while providing a reproducible and publicly accessible method for automated cotton fiber cross-sectional phenotyping, facilitating quantitative analysis for cotton genetics and breeding research.
Why it matches plant phenotyping methods綿繊維横断面の画像分割、形質抽出、検証を目的とした再現可能な深層学習ワークフローの開発であり、植物フェノタイピング手法が研究の中心である。
abstractThe objective of this study was to develop and validate a reproducible deep-learning-based workflow for automated segmentation and quantitative analysis of cotton fiber cross-sections.
Reproduction assets foundThe paper's cotton fiber cross-section phenotyping workflow (YOLO11m segmentation pipeline, trained model weights, example images/outputs) is explicitly stated as publicly available via a GitHub repository and a Hugging Face web application, with URLs matching the allowed list.Code · publicThe complete source code, training scripts, dataset configuration, example input images, example outputs,
and supporting documentation are publicly available through the GitHub repository:
https://github.com/RifeLab/cotton-lumen-microOpen asset ↗RifeLab/cotton-lumen-micropdf-page:11 lines:1-47Code · publicThe cotton fiber image analysis workflow is publicly available through a web-based application hosted on
Hugging Face at: https://huggingface.co/spaces/chaneylc/cotton_fiber_microscopy_measureOpen asset ↗pdf-page:11 lines:1-47Code / 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
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 15 Sept 2026
This paper presents a deep learning model to detect cotton plant pests and classify diseases, which must overcome limited datasets, class imbalance, and feature redundancy. At the preprocessing phase, the Gaussian blur filtering and Contrast Limited Adaptive Histogram Equalization (CLAHE) are used to sharpen images by improving their clarity and contrast. To increase and diversify the data, Spa-GAN-based data augmentation is used to produce realistic synthetic samples. To obtain an accurate Region of Interest (RoI), an Attention-Guided Multi-Scale Residual U-Net (AGMS-U-Net) is considered to segment local and global structural information. The proposed framework makes three major contributions: (i) a new attention-based feature extractor, TriAttnNet, that incorporates spatial, channel, and contextual attention to represent diseases on a fine-grained level (ii) a new optimization strategy, Hybrid Mongoose Ray Chaotic Optimization (HMRCO), which includes chaotic strategies to better tune the parameters and explore the feature space and (iii) classification layer with focal loss for final decision. Experimental analyses prove that the suggested method is much more effective than the current state-of-the-art models, providing a powerful and understandable solution to precision agriculture and sustainable cotton crop health monitoring. Experimental results show that TriAttnNet achieves 98.66% accuracy, 98.71% recall, and 98.81% F1-score, which is better than the state-of-the-art algorithms, such as EfficientNetB1-CBAM (96.38%) and BERT-ResNet-PSO (95.69%). The proposed system is computationally feasible and interpretable, and it is interpretable to provide a practical solution to precision agriculture and sustainable monitoring of the health of cotton crops.
Why it matches plant phenotyping methods綿花植物の画像から病害を分類する深層学習手法を開発・評価しており、植物の病害状態を直接推定する方法が研究の中心である。
abstractThis paper presents a deep learning model to detect cotton plant pests and classify diseases
Reproduction assets foundThe paper's plant-phenotyping input is the public Kaggle Cotton Plant Disease Dataset (Dhamodharan R), explicitly cited as the study's data source with a matching public URL. The authors' model/code is not publicly deposited (available only upon request), so no qualifying code asset exists.Dataset · publicThe dataset of this study is taken from the publicly available Kaggle repository, Cotton Plant Disease Dataset 43Open asset ↗Kagglelines:48-58Code / 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 · 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 confirmedCrossref · OpenAlex · Europe PMC · checked 5 Sept 2026
Attempts to deploy computer vision in agricultural tasks often suffer from a shortage of annotated data. One strategy to alleviate the impact of limited data is Self-Supervised Learning (SSL), which involves pre-training a model on a pretext task that utilizes automatically generated annotations. The primary objective of this study is to leverage a multi-camera view dataset of cotton boll images for contrastive learning in order to enable phenotyping tasks with minimal data annotation. This dataset was collected in the field using six camera views. The efficacy of two contrastive learning frameworks (SimCLR and MoCo) in producing representations when positive examples originate from different cameras was investigated, and a comprehensive study of how the camera positions affect performance was conducted. After self-supervised pre-training, linear evaluation and semi-supervised learning experiments were performed on boll detection and plot status downstream tasks. In general, using multiple camera views with SimCLR and MoCo improves cotton boll detection mean average precision by 14% compared to vanilla SimCLR and MoCo. Through careful investigation using synthetic data, it was determined that relative camera poses with an intermediate amount of overlap seem more likely to perform well. Neither MoCo nor SimCLR was consistently superior to the other in this context. The representations embed meaningful features about the cotton plants, such as overall boll density, but also less meaningful ones, such as lighting variations. This technique could potentially accelerate the development of phenotyping algorithms based on data collected from field robots. • A contrastive learning method based on comparing multi-camera views was developed. • The method was tested with images of cotton bolls from a ground robot. • The method outperformed baseline contrastive learning approaches.
Why it matches plant phenotyping methodsマルチカメラ画像とコントラスト学習による植物表現学習・フェノタイピング手法の開発と評価が中心であり、綿花のボール検出性能を検証している。
abstractThe primary objective of this study is to leverage a multi-camera view dataset of cotton boll images for contrastive learning in order to enable phenotyping tasks with minimal data annotation.
Reproduction assets foundThe paper's data availability statement points to a public GitHub repository containing the authors' code to reproduce the multi-camera contrastive learning phenotyping experiments. A processed-data Zenodo deposit (10.5281/zenodo.18164649) is also mentioned, but its URL is not among the allowed URLs, so only the code资产Code · publicThe code required to reproduce the above findings are available to download from https://github.com/UGA-BSAIL/self-supervised-learning .Open asset ↗UGA-BSAIL/self-supervised-learninglines:200-224Code / 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 · checked 5 Sept 2026
Abstract This paper presents a deep learning model to detect cotton plant pests and classify diseases, which must overcome limited datasets, class imbalance, and feature redundancy. At the preprocessing phase, the Gaussian blur filtering and Contrast Limited Adaptive Histogram Equalization (CLAHE) are used to sharpen images by improving their clarity and contrast. To increase and diversify the data, Spa-GAN-based data augmentation is used to produce realistic synthetic samples. To obtain an accurate Region of Interest (RoI), an Attention-Guided Multi-Scale Residual U-Net (AGMS-UNet) is considered to segment local and global structural information. The proposed framework makes three major contributions: (i) a new attention-based feature extractor, TriAttnNet , that incorporates spatial, channel, and contextual attention to represent diseases on a fine-grained level; (ii) a new optimization strategy, Hybrid Mongoose Ray Chaotic Optimization (HMRCO), which includes chaotic strategies to better tune the parameters and explore the feature space; and (iii) classification layer with focal loss for final decision. Experimental analyses prove that the suggested method is much more effective than the current state-of-the-art models, providing a powerful and understandable solution to precision agriculture and sustainable cotton crop health monitoring. Experimental results show that TriAttnNet achieves 98.66% accuracy, 98.71% recall, and 98.81% F1-score, which is better than the state-of-the-art algorithms, such as EfficientNetB1-CBAM (96.38%) and BERT-ResNet-PSO (95.69%). The proposed system is computationally feasible and interpretable, and it is interpretable to provide a practical solution to precision agriculture and sustainable monitoring of the health of cotton crops.
Why it matches plant phenotyping methods綿植物の病害を画像から検出・分類する深層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して性能比較・検証しているため。
abstractThis paper presents a deep learning model to detect cotton plant pests and classify diseases
Reproduction assets foundThe paper's plant image input is the public Kaggle Cotton Plant Disease Dataset, explicitly cited with URL. Authors' code/models are only available upon request, so no public code asset qualifies.Dataset · publicThe dataset of this study is taken from the publicly available Kaggle repository, Cotton Plant Disease
Dataset [25].Open asset ↗Kagglepdf-page:6 lines:1-48Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
In agricultural automation, precise cotton segmentation is a key step for tasks such as intelligent harvesting and yield estimation. However, in complex field environments, factors such as background interference and irregular target shapes severely affect segmentation accuracy. Existing deep learning methods offer certain advantages but still generally suffer from limitations including insufficient accuracy, over-segmentation, and misidentification. To address these challenges, this study proposes a novel dual-branch cotton segmentation network, Cotton-aware Mamba-enhanced UNet (CMNet), which optimizes the ParaTransCNN architecture by incorporating the 2D Selective Scan (SS2D) module to replace the original Transformer branch, effectively balancing the extraction of local details and global semantic information while reducing computational burden. To enhance the model's perception of irregularly shaped cotton, a Deformable Convolutional Networks v1 (DCNv1) module is integrated into the Vision Mamba (VMamba) branch, further improving the delineation of target boundaries. Additionally, an Atrous Spatial Pyramid Pooling (ASPP) module is introduced at the end of the Convolutional Neural Network (CNN) branch to strengthen multi-scale feature representation. To optimize the fusion of channel and spatial information, the Spatial and Channel Squeeze-and-Excitation (scSE) attention mechanism replaces the original module, enhancing feature modeling capability. Experimental results on an in-field cotton image dataset demonstrate that CMNet outperforms existing mainstream methods, achieving Dice, mIoU, and Accuracy of 91.06%, 84.18%, and 98.10%, respectively, while reducing parameter count and computational complexity, thus exhibiting excellent performance. Furthermore, generalization experiments on multiple other plant datasets also achieved outstanding results, validating the model's adaptability and potential for broader applications in multi-crop segmentation tasks, providing valuable insights for smart agriculture segmentation research. The source code and dataset of this work are publicly available at https://github.com/halidanmu/CMNet.git.
Why it matches plant phenotyping methods綿花画像から植物領域を抽出する新規セグメンテーション手法を中心に開発・検証しており、植物表現型の画像取得・抽出ワークフローに該当する。
abstractthis study proposes a novel dual-branch cotton segmentation network, Cotton-aware Mamba-enhanced UNet (CMNet)
Reproduction assets foundThe authors explicitly state that the source code and dataset for CMNet are publicly available on GitHub. The paper also uses several public Roboflow plant image datasets in its generalization experiments, cited with public URLs in the references.Code · publicThe source code and dataset of this work are publicly available at https://github.com/halidanmu/CMNet.git.Open asset ↗halidanmu/CMNethtml-lines:106-109Dataset · publicELTE (2023). Assignment 2 dataset. Available online at: https://universe.roboflow.com/elte-msgqy/assignment_2-mjhau (Accessed November 5, 2025).Open asset ↗html-lines:754-834Dataset · publicLaola (2024). Defect banana dataset. Available online at: https://universe.roboflow.com/laola/defect-banana-qf4f6 (Accessed November 5, 2025).Open asset ↗html-lines:754-834Dataset · publicLuffy24312 (2023). Cnn dataset. Available online at: https://universe.roboflow.com/luffy24312/cnn-myqtl.Open asset ↗html-lines:835-919Dataset · publicVyuha T. (2025). Rose dataset. Available online at: https://universe.roboflow.com/tech-vyuha/rose-kfpuf (Accessed November 4, 2025).Open asset ↗html-lines:835-919Code / 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
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 confirmedarXiv · OpenAlex · checked 15 Sept 2026
AppleCottonPearField / plotNeRF / 3D Gaussian SplattingFruitCounting2D/3D reconstructionSegmentation
Rigorous crop counting is crucial for effective agricultural management and informed intervention strategies. However, in outdoor field environments, partial occlusions combined with inherent ambiguity in distinguishing clustered crops from individual viewpoints poses an immense challenge for image-based segmentation methods. To address these problems, we introduce a novel crop counting framework designed for exact enumeration via 3D instance segmentation. Our approach utilizes 2D images captured from multiple viewpoints and associates independent instance masks for neural radiance field (NeRF) view synthesis. We introduce crop visibility and mask consistency scores, which are incorporated alongside 3D information from a NeRF model. This results in an effective segmentation of crop instances in 3D and highly-accurate crop counts. Furthermore, our method eliminates the dependence on crop-specific parameter tuning. We validate our framework on three agricultural datasets consisting of cotton bolls, apples, and pears, and demonstrate consistent counting performance despite major variations in crop color, shape, and size. A comparative analysis against the state of the art highlights superior performance on crop counting tasks. Lastly, we contribute a cotton plant dataset to advance further research on this topic.
Why it matches plant phenotyping methodsNeRFと3Dインスタンスセグメンテーションを用いて作物個体・器官数を推定する画像ベース表現型計測手法を開発・検証しており、方法が研究の中心である。
abstractwe introduce a novel crop counting framework designed for exact enumeration via 3D instance segmentation.
Reproduction assets foundThe paper contributes a public infield cotton plant dataset (8 plants, ~150 iPhone images each, ground-truth boll counts, SAM instance masks) and states that source code, dataset, and multimedia are available at the authors' public project page, which is an allowed URL. The spectacularai GitHub URL is a generic third-pDataset · publicthat incorporates crop visibility
and mask consistency, enabling robustness against occlusions and annotation
discrepancies.
•
We release a public infield cotton plant dataset designed for 3D
rendering and cotton boll counting tasks.
The source code, dataset, and multimedia material associated with this project
can be found at
https://robotic-vision-lab.github.io/cropnerf .
II Related Work
II-A Image-Based Techniques
Image-based methods typically employ object detection to identify crops within
images. For example, Chen et al. [ 4 ] utilized multiple
convolutional neural networks (CNNs) to map input images to total fruit counts.
Similarly, Häni et al. [ 5 ] formulated crop counting as a
multOpen asset ↗lines:108-187Code / 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 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 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:312-354Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Cotton diseases and pests pose significant threats to cotton production, necessitating accurate and efficient classification methods. Despite existing advanced methods, there is a research gap in utilizing both local feature extraction and global context capture for enhanced classification accuracy. Hence, this study developed and evaluated three advanced models for cotton disease and pest classification: a convolutional neural network (CNN)-based model, a Vision Transformer (ViT)-based model, and a hybrid CNN-ViT model. These models were trained on a dataset comprising eight classes of cotton diseases and pests, namely aphids, armyworm, bacterial blight, cotton boll rot, green cotton boll, healthy, powdery mildew, and target spot. The results demonstrated that the hybrid CNN-ViT model achieved the highest overall performance with an average test accuracy of 98.5%. The CNN model showed strong performance with an average accuracy of 97.9%. The ViT models, while having self-attention mechanisms to capture context and dependencies, exhibited improved performance with increased depth. The ViT model having four transformer layers outperformed the two-layer variant, achieving an average accuracy of 97.2% compared to 96.3%. The hybrid model effectively combined the strengths of CNN's local feature extraction and ViT's global feature capture, resulting in superior classification accuracy across most classes. Future research should focus on expanding the dataset to include more diverse diseases and pests and integrating the models with autonomous platforms for spraying the chemicals, thus facilitating real-world adoption and application in agricultural settings.
Why it matches plant phenotyping methods綿花の病害・害虫を画像から分類するCNN、ViT、ハイブリッド手法を開発・評価しており、植物の病害状態推定が研究の中心である。
abstractthis study developed and evaluated three advanced models for cotton disease and pest classification
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 for this study was downloaded from Kaggle ( https://www.kaggle.com/datasets/saeedazfar/customized-cotton-disease-dataset ) and comprises images of various cotton diseases and pests.Open asset ↗Kaggle · saeedazfar/customized-cotton-disease-datasetlines:78-87Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Plant diseases pose a severe threat to global agricultural production, significantly challenging crop yield, quality, and food security. Therefore, accurate and efficient disease detection is crucial. Current detection methods have clear limitations: CNN-based methods struggle to model long-range dependencies effectively and have weak generalization abilities. Transformer-based methods, while adept at long-range feature modeling, face issues with large parameter sizes and inefficient calculations due to the quadratic complexity of the self-attention mechanism in relation to image size. To address these challenges, this paper proposes the MamSwinNet model. Its core innovation lies in: using the Efficient Token Refinement module with an overlapping space reduction method, relying on depthwise separable convolutions designed with “stride + 3” convolution kernels to expand the image block overlap area and fully preserve boundary spatial structure. This generates high-quality tokens and converts them into a fixed number of latent tokens, reducing computational complexity while maximizing the retention of key features. It integrates the Spatial Global Selective Perception (SGSP) module and the Channel Coordinate Global Optimal Scanning (CCGOS) module. The SGSP module uses a dual-branch structure (the spatial modeling branch introduces 2D-SSM to scan four directions for capturing long-range dependencies, and the residual compensation branch supplements features to prevent loss; the two branches are combined using Hadamard product to enhance spatial detail modeling). The CCGOS module combines channel and spatial attention by embedding positional information through global average pooling in the height and width dimensions, using the Mamba block for channel-selective scanning and generating an attention map, enabling precise association of key channel features like color with spatial distribution. Experimental results show that the model achieves F1 scores of 79.47%, 99.52%, and 99.38% on the PlantDoc, PlantVillage, and Cotton datasets, respectively. The model has only 12.97M parameters (52.9% less than the Swin-T model) and a computational cost as low as 2.71GMac, significantly improving computational efficiency. This study provides an efficient and reliable intelligent solution for large-scale crop disease detection.
Why it matches plant phenotyping methods植物病害を画像から検出・判定するMamSwinNetモデルの開発とデータセット評価が研究の中心であり、感染植物の病態を直接推定する画像ベース表現型解析に該当する。
abstractExperimental results show that the model achieves F1 scores of 79.47%, 99.52%, and 99.38% on the PlantDoc, PlantVillage, and Cotton datasets, respectively.
Reproduction assets foundThe paper's phenotyping inputs are three public plant-disease image datasets (PlantDoc, PlantVillage, Cotton Disease) used for all experiments. No author analysis code, trained model checkpoints, or paper-specific supplements are disclosed.Dataset · publicwhile enhancing its practical relevance. Owing to these characteristics, PlantDoc has become a key benchmark dataset for evaluating the robustness and applicability of plant disease detection models.
Figure 5
illustrates several representative samples from the dataset. The dataset is publicly accessible via the following link: https://github.com/pratikkayal/PlantDoc-Dataset
Figure 5
Sample images from the PlantDoc dataset.
Grid of images showing various diseased leaves. Top row: Apple rust leaf with red spots, apple scab leaf with black lesions, bell pepper leaf with dark spots. Middle row: Corn gray leaf with discolored areas, two corn leaves with blight showing yellow and brown patterns. BOpen asset ↗PlantDoc-Datasetlines:151-174Dataset · publicility and distinct disease features, PlantVillage is frequently employed for model pre-training and performance benchmarking, and has become an important reference dataset in plant disease detection research.
Figure 6
presents several representative samples from this dataset. PlantVillage can be accessed via the following link: https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset
Figure 6
Sample images from the PlantVillage dataset.
Grid of nine leaves showing various plant diseases. Top row: Apple Black rot, Apple scab, Grape Leaf blight. Middle row: Peach Bacterial spot, Potato Early blight, Squash Powdery mildew. Bottom row: Tomato Early blight, Tomato Septoria Leaf spot, TOpen asset ↗plantvillage-datasetlines:151-174Dataset · publicdataset’s class design not only covers the major and prevalent diseases in cotton production but also provides a reliable benchmark for evaluating models in multi-class disease classification tasks.
Figure 7
presents several representative image samples from this dataset. The dataset is publicly available at the following link: https://www.kaggle.com/datasets/dhamur/cotton-plant-disease .
Figure 7
Sample images from the cotton disease dataset.
Nine images of leaves show different conditions: three with aphids, showing yellowing and damage; two with bacterial blight, displaying dark spots; three healthy with vibrant green; and three with powdery mildew, covered in white residue. Each conditioOpen asset ↗cotton-plant-diseaselines:151-174Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Light intensity and spectral distribution within plant canopies provides insights into the effects of optimizing canopy architecture on light use efficiency. Breeding crop varieties with a "smart" canopy, characterized by erect upper-layer leaves and flat lower-layer leaves, can be supported with a 3D canopy model which can simulate light distribution for a particular canopy architecture. Leaf optical properties are required parameters for such canopy photosynthesis model to accurately predict canopy microclimate and hence photosynthetic efficiency. In this study, we developed a strategy to estimate the leaf optical properties based on leaf anatomical features. We developed a Directional Spectrum Detection Instrument (DSDI) system and associated Bidirectional Reflectance Distribution Function (BRDF) analysis software to precisely describe leaf light distribution. BRDF parameters were quantified with high accuracy ( R2>0.95 ) for adaxial and abaxial surfaces of maize, rice, cotton, and poplar leaves across canopy layers. Leaf phenotypic traits, surface roughness, pigments content, specific leaf weight and thickness were also assessed. Ensemble learning (EL) model showed excellent predictive performance for leaf optical properties based on phenotypic traits with R 2 between 0.83 and 0.99. Compared to existing BRDF measurement systems, the DSDI achieves broader angular coverage (-π/36 to 35π/36) via mechanical rotation design, and the ensemble learning model establishes the first direct predictive relationship between BRDF parameters and leaf phenotypic traits. This work presents a new approach to quantify leaf optical properties and offers predictive models for leaf optical properties, which can support canopy light distribution prediction and hence support design leaf features for higher canopy photosynthesis efficiency.
Why it matches plant phenotyping methods葉の光学特性と表現型形質を取得・予測する測定機器、BRDF解析ソフトウェア、機械学習モデルを開発しており、植物フェノタイピング手法が研究の中心である。
abstractthe ensemble learning model establishes the first direct predictive relationship between BRDF parameters and leaf phenotypic traits.
Reproduction assets foundThe paper's BRDF analysis code (adaptive grid search fitting and Roughness Calculator) is publicly available at github.com/PlantSystemsBiology/brdf, and the modified fastTracer ray tracing software used for canopy light simulations is at github.com/PlantSystemsBiology/fastTracerPublic. Phenotype/measurement data are '…Code · publicAn adaptive grid search algorithm was developed in this study, and this algorithm utilized a 2-layered grid (step sizes of 1 × 10 − 2 and 1 × 10 − 4 respectively) structure to incrementally optimize each parameter, providing a more precise approximation of true values. By iteratively narrowing the search range and increasing resolution, this method gradually converges on the optimal solution. The source code of Python for adaptive grid search algorithm was available at https://github.com/PlantSystemsBiology/brdf .Open asset ↗PlantSystemsBiology/brdflines:212-227Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Cotton, often referred to as "white gold" or the "king of fibers," is one of the most widely used natural fibers in the global textile industry, supporting approximately 250 million people worldwide. However, cotton plants suffer from a variety of diseases, particularly leaf diseases, which can significantly reduce the yield and fiber quality. To overcome this problem, we propose a carefully curated image dataset that enables research toward early and automated disease detection and health monitoring of cotton plants. The dataset comprises 1373 original and 4963 augmented high-resolution images of cotton leaves with healthy, damaged, and infected samples. The images were captured under different environmental conditions from plants grown at the Sher-e-Bangla Agricultural University in Dhaka, Bangladesh to provide natural variability and realism. The dataset considers four common cotton leaf diseases-Fusarium wilt, Alternaria leaf spot, Verticillium wilt, and bacterial blight-each labeled and classified to support machine learning applications. Captured from different angles and devices, the images have rich visual content that enables the development of strong deep learning models for disease classification. The dataset was designed to advance research relevant to precision agriculture by supporting early disease detection studies, crop health monitoring, and sustainable cotton-growing methods.
Why it matches plant phenotyping methods綿花葉の病害・健全状態を画像で分類するためのデータセットであり、植物の病害状態を直接観測する再利用可能なフェノタイピング資源が中心です。
abstractwe propose a carefully curated image dataset that enables research toward early and automated disease detection and health monitoring of cotton plants.
Reproduction assets foundThe paper is a data article describing a cotton leaf image dataset (1373 original + 4963 augmented images) for disease classification, publicly deposited on Mendeley Data with DOI 10.17632/t9hgvk2h9p.1 and a direct URL. This is the paper's own plant-phenotyping (leaf disease image) dataset and is directly actionable.Dataset · publicRepository name: Mendeley Data
Data identification number: 10.17632/t9hgvk2h9p.1
Direct URL to data: https://data.mendeley.com/datasets/t9hgvk2h9p/1Open asset ↗Mendeley Data · 10.17632/t9hgvk2h9p.1lines:1-48Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Pre-harvest defoliation of cotton is a key agricultural measure to improve mechanical harvesting efficiency and raw cotton purity. Collecting data on cotton defoliation traits for genetic localization and thus breeding defoliation-prone varieties is an essential alternative to traditional defoliant spraying. Nevertheless, it is hampered by low throughput and artificial error in manual field surveys. In this study, a framework for collecting high-throughput defoliation data in large fields was established. Three spectral indices (MTCI, VDVI, CI) and leaf area index (LAI) were first screened as core predictors through hierarchical segmentation analysis in three levels: leaf number (LN), leaf number difference (LND), and defoliation rate (DR). Four deep learning architectures (CNN, BiGRU, CNN-BiGRU, and CNN-BiGRU-Attention) were developed, and the CNN-BiGRU-Attention hybrid model demonstrated superior performance at all three levels, with R 2 values exceeding 0.85. Importantly, the inversion accuracy of this model at the LN and LND levels was superior to that at the DR level, which was also confirmed by the results of the genome-wide association study (GWAS). We combined GWAS and transcriptome results to identify a new gene, GhDR_UAV1 , associated with defoliation traits. The overexpression of GhDR_UAV1 significantly promoted the wilting of cotton leaves, indicating that GhDR_UAV1 plays a positive regulatory role in cotton defoliation. This study proposed a strategy to invert cotton defoliation data at three levels using deep learning fusion of UAV remote sensing data and LAI data and confirmed that LND can provide accurate phenotypic data for GWAS analysis. This study provides a new theoretical basis for cotton defoliation regulation and genetic improvement by integrating cotton high-throughput defoliation phenomics and genomics from an innovative perspective.
Why it matches plant phenotyping methodsUAVリモートセンシング、LAI、深層学習を統合し、ワタの落葉形質を高スループット推定する方法を開発・評価しており、表現型取得が研究の中心である。
abstracta framework for collecting high-throughput defoliation data in large fields was established
Reproduction assets foundThe paper's Data Availability statement explicitly deposits the study's data and code (UAV/LAI defoliation phenotyping data and analysis code) in a public GitHub repository, which matches an allowed URL.Code · publicThe data and code utilized in this study are available at GitHub ( https://github.com/xbw322/Data_upload.git ).Open asset ↗https://github.com/xbw322/Data_upload.gitlines:169-205Code / dataset availability confirmedarXiv · checked 15 Sept 2026
Cotton is one of the most important natural fiber crops worldwide, yet harvesting remains limited by labor-intensive manual picking, low efficiency, and yield losses from missing the optimal harvest window. Accurate recognition of cotton bolls and their maturity is therefore essential for automation, yield estimation, and breeding research. We propose Cott-ADNet, a lightweight real-time detector tailored to cotton boll and flower recognition under complex field conditions. Building on YOLOv11n, Cott-ADNet enhances spatial representation and robustness through improved convolutional designs, while introducing two new modules: a NeLU-enhanced Global Attention Mechanism to better capture weak and low-contrast features, and a Dilated Receptive Field SPPF to expand receptive fields for more effective multi-scale context modeling at low computational cost. We curate a labeled dataset of 4,966 images, and release an external validation set of 1,216 field images to support future research. Experiments show that Cott-ADNet achieves 91.5% Precision, 89.8% Recall, 93.3% mAP50, 71.3% mAP, and 90.6% F1-Score with only 7.5 GFLOPs, maintaining stable performance under multi-scale and rotational variations. These results demonstrate Cott-ADNet as an accurate and efficient solution for in-field deployment, and thus provide a reliable basis for automated cotton harvesting and high-throughput phenotypic analysis. Code and dataset is available at https://github.com/SweefongWong/Cott-ADNet.
Why it matches plant phenotyping methods綿花の花・ボール認識を対象とする画像解析手法を開発し、データセット作成、外部検証、性能評価まで行っており、植物器官の表現型取得が中心である。
abstractWe propose Cott-ADNet, a lightweight real-time detector tailored to cotton boll and flower recognition under complex field conditions.
Reproduction assets foundThe paper explicitly states that its code and curated cotton boll/flower detection dataset (4,966 labeled images plus a 1,216-image external validation set) are publicly released at the authors' GitHub repository. The ultralytics repository is a generic third-party library, not a paper-specific asset.Code · publicy 7.5 GFLOPs, maintaining stable performance under multi-scale and rotational variations. These results demonstrate Cott-ADNet as an accurate and efficient solution for in-field deployment, and thus provide a reliable basis for automated cotton harvesting and high-throughput phenotypic analysis. Code and dataset is available at https://github.com/SweefongWong/Cott-ADNet .
† † footnotetext: ∗ * Corresponding author: cuij@wfu.edu
Index Terms :
cotton, cotton boll detection, lightweight object detection, rotational convolution
1 Introduction
Cotton is one of the most critical economic crops worldwide, accounting for nearly 35% of global natural fiber production. It underpins industries such as Open asset ↗SweefongWong/Cott-ADNetlines:1-57Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Abstract This study investigates the complex regulatory mechanisms of genotype-environment interactions (GEI) in cotton phenotype formation and explores the genetic basis of environmental adaptation through an integrated analytical approach. The research methodology encompasses four key components: (1) deep learning model construction, (2) phenotypic plasticity analysis, (3) environmental adaptation assessment, and (4) genome-wide association study (GWAS). Based on the multi-head self-attention mechanism and the deep feature interaction, we constructed the AttGEI-Net deep learning framework. The model demonstrates remarkable predictive performance with an average accuracy of 0.96 in fixed environments, though this decreases to 0.39-0.44 in novel environments, revealing fundamental differences between genotype-dominated and environment-dominated prediction scenarios. A total of 10,215 significant SNP loci is identified by GWAS, including 2,705 Main-SNPs, 41 phenotype plasticity loci (PP-SNPs), and 9,022 environmental adaptation loci (EvA-SNPs). The regulation of phenotypes by these loci has a distinct hierarchical character: the basic genetic architecture (Main-SNPs) maintains the basic expression of traits, the PP-SNPs mediates the immediate response of phenotypes to environmental changes, and the EvA-SNPs constitutes the highest-level adaptive regulatory network that coordinates the expression of multiple traits by integrating environmental signals. Shared loci of interpretability analyses of model and GWAS may be the key genetic basis adapting to different environments. Broadly adapted varieties in the Yellow River basin (e.g., F096, L090, etc.) can be used as the backbone parents for suitability breeding.
Why it matches plant phenotyping methods綿花の表現型を予測する深層学習フレームワークを構築し、環境間で予測性能を評価しているため、計算的な表現型推定手法が研究の中心です。
abstractdeep learning model construction
Reproduction assets foundThe paper's authors publicly released the AttGEI-Net model code used for cotton phenotype prediction and interpretability analysis on GitHub. The phenotype/trait datasets themselves are not publicly deposited (available only on request), and the genomic deposits are molecular omics data, which do not qualify.Code · public841 The code of our model has been made available at https://github.com/hezikang-git/AttGEI-Net,Open asset ↗hezikang-git/AttGEI-Netpdf-page:32 lines:1-53Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Agricultural is the primary source of essential provisions almost all nation and remains as a vital survival tool for human race for past, present and future. The agriculture which remains as the bedrock of the civilizations is frequently being affected by devastating plant diseases leading to the loss of economy and food shortage. The manual testing requires high expertise, time and often leads to human error. Recent advancements in computer vision and AI models have highlighted the potential for building automatic plant disease detection models based on visible signs using image classification tasks. However, the task becomes complex and erroneous due to the complex nature of data. Consequently Deep Learning (DL) models tend to outperform others traditional methods through utilizing network topology with convolution layers in core. Projected system uses high quality Cotton Disease dataset sourced by Kaggle and provide a suitable solution to the aforementioned problem using advanced DL neural network namely Blend Unity Resqueeze Resnet approach which yields high accuracy with modification including Resqueeze layer and blend unity weights applied to do the tedious job of diagnosing plant diseases on image based classification. The proposed research outperforms the conventional methods achieving better accuracy of 92% with better precision and reliability. The outcome of the respective research is analyzed with suitable metrics and compared with the recent developed conventional algorithms in which the proposed model proves to be a better suited for the efficient plant disease classification. Hence, the proposed method is intended to contribute in the plant disease classification and assist the agriculturists significantly to prevent the losses due to crop diseases.
Why it matches plant phenotyping methods植物画像から病害状態を分類する深層学習手法の開発・比較が中心であり、植物表現型（病害状態）の取得・推定に該当する。
abstractbuilding automatic plant disease detection models based on visible signs using image classification tasks
Reproduction assets foundThe paper's plant-phenotyping measurements are based on the public Cotton Disease Dataset from Kaggle, explicitly cited with URL and CC BY 4.0 license. No author analysis code or trained model is shared.Dataset · publicTable 3. Significant Features of the Validation Dataset
S. Features Feature percentage
no %
1. DL 16.97
2. DP 31.17
3. FL 24.70
4. FP 27.16
The utilized dataset is licensed with creative commons attribution 4.0 international. The dataset is acquired from
the following link:
https://www.kaggle.com/datasets/janmejaybhoi/cotton-disease-dataset/data
4.2 Performance Metrics
The metrics used for evaluating the system’s performance are precision, accuracy, f1-score and recall.
1. Precision: The metric signifies the count of accurate positive predictions. Precision is calculated by
taking the ratio of true positives to the total number of positive predictions, and it is expresOpen asset ↗Kaggle · janmejaybhoi/cotton-disease-datasetpdf-layout-page:10 lines:1-51Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Cotton leaf diseases significantly impact global cotton yield and quality, threatening the livelihoods of millions of farmers. Traditional diagnostic methods are often slow, subjective, and unsuitable for large-scale agricultural monitoring. This study proposes an interpretable and efficient deep learning (DL) framework for the accurate classification of cotton leaf diseases using a hybrid architecture that combines EfficientNetB3 and InceptionResNetV2. The system demonstrates excellent performance, achieving 98.0% accuracy, 98.1% precision, 97.9% recall, an F1-score of 98.0%, and an AUC-ROC of 0.9992. Minimal overfitting was observed, with low training and validation losses and high per-class performance, even in visually similar disease cases such as bacterial blight and target spot. In addition to strong predictive accuracy, the framework incorporates explainable AI (XAI) techniques, including LIME and SHAP, to enhance model transparency. These tools highlight the key visual features used in predictions, providing valuable insights for agronomists and improving trust in AI-based systems. The model is lightweight and scalable, making it deployable on mobile or edge devices for real-time field applications. Overall, this research demonstrates the potential of combining transfer learning and XAI to develop reliable, interpretable, and field-ready diagnostic tools for precision agriculture.
Why it matches plant phenotyping methods綿花葉の病害状態を画像から分類する深層学習・説明可能AI手法の開発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。
abstractThis study proposes an interpretable and efficient deep learning (DL) framework for the accurate classification of cotton leaf diseases using a hybrid architecture that combines EfficientNetB3 and InceptionResNetV2.
Reproduction assets foundThe paper's Data Availability Statement explicitly points to two public sources for the cotton leaf disease image dataset used in this study: a GitHub repository and a Kaggle dataset. Both are paper-specific, public, and actionable. No author analysis code or trained model checkpoints are explicitly deposited.Dataset · publicThe dataset used in this study is publicly available at [ https://github.com/gurjot000/cotton‐leaf‐disease/tree/main ] and [ https://www.kaggle.com/datasets/ataher/cotton‐leaf‐disease‐dataset/data ].Open asset ↗lines:650-650Dataset · publicThe dataset used in this study is publicly available at [ https://github.com/gurjot000/cotton‐leaf‐disease/tree/main ] and [ https://www.kaggle.com/datasets/ataher/cotton‐leaf‐disease‐dataset/data ].Open asset ↗gurjot000/cotton‐leaf‐diseaselines:650-650Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Cotton is the most widely cultivated natural fiber crop worldwide, yet it is highly susceptible to various diseases and pests that significantly compromise both yield and quality. To enable rapid and accurate diagnosis of cotton diseases and pests-thus supporting the development of effective control strategies and facilitating genetic breeding research-we propose a lightweight model, the Resource-efficient Cotton Network (RF-Cott-Net), alongside an open-source image dataset, CCDPHD-11, encompassing 11 disease categories. Built upon the MobileViTv2 backbone, RF-Cott-Net integrates an early exit mechanism and quantization-aware training (QAT) to enhance deployment efficiency without sacrificing accuracy. Experimental results on CCDPHD-11 demonstrate that RF-Cott-Net achieves an accuracy of 98.4%, an F1-score of 98.4%, a precision of 98.5%, and a recall of 98.3%. With only 4.9 M parameters, 310 M FLOPs, an inference time of 3.8 ms, and a storage footprint of just 4.8 MB, RF-Cott-Net delivers outstanding accuracy and real-time performance, making it highly suitable for deployment on agricultural edge devices and providing robust support for in-field automated detection of cotton diseases and pests.
Why it matches plant phenotyping methods綿花の病害を画像から分類する軽量深層学習モデルと画像データセットを開発・評価しており、植物の病害状態を抽出するフェノタイピング手法が中心である。
abstractwe propose a lightweight model, the Resource-efficient Cotton Network (RF-Cott-Net), alongside an open-source image dataset, CCDPHD-11, encompassing 11 disease categories.
Reproduction assets foundThe paper's authors publicly released their self-constructed cotton disease/pest image dataset CCDPHD-11 (18,953 images, 11 classes) on GitHub, as stated in the Data Availability Statement. No code or trained model deposit is mentioned.Dataset · publicew and editing, K.C., H.W., P.W.C. and R.-F.W.; visualization, H.-W.Z. and R.-F.W.; supervision, H.W., P.W.C. and R.-F.W.; project administration, H.W., P.W.C. and R.-F.W. All authors have read and agreed to the published version of the manuscript.
Data Availability Statement
The proposed CCDPHD-11 dataset can be found online ( https://github.com/SweefongWong/CCDPHD-11-Dataset , accessed on 9 March 2025).
Conflicts of Interest
The authors declare no conflicts of interest.
Funding Statement
This research received no external funding.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributOpen asset ↗SweefongWong/CCDPHD-11-Datasetlines:384-405Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Cotton, the backbone of global textile production, demands sustainable agricultural practices to ensure fiber, food, and environmental security. Cotton crop play an essential role in farming economies; however, production is sometimes affected by various diseases that harm production. We proposed a methodology that uses formal modeling and verification for requirements confirmation to improve the monitoring and detection of cotton crop diseases. The correct information and requirements about disease symptoms can improve disease monitoring and prediction. The Temporal Logic of Action (TLA+) is used to construct a mathematical model to verify requirements by providing disease symptoms and then model checking to ensure correctness properties. Using model checking in TLA + ensures the reliability and correctness of disease symptom detection. We consequently used deep learning models to predict cotton diseases, i.e., Aphids, Armyworms, Bacterial Blight, Powdery Mildew, Target Spot, and Healthy leaf. Our results show that the Convolutional Neural Network (CNN) model achieved an overall accuracy of 98.7% with class-specific accuracy ranging from with F1-scores across all classes (e.g., 0.90 for Powdery Mildew and 0.87 for Army Worm).
Why it matches plant phenotyping methods綿花葉の病徴を画像から深層学習で分類・検出する手法が研究の中心であり、植物の病害状態を推定するため、植物フェノタイピング手法として収録対象。
abstractWe proposed a methodology that uses formal modeling and verification for requirements confirmation to improve the monitoring and detection of cotton crop 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。Code · publictoolbox modules, are publicly available on GitHub at https://github.com/abdulpk/MuhammadMusa for furtherOpen asset ↗https://github.com/abdulpk/MuhammadMusapdf-page:18 lines:1-64Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Cotton is one of the most valuable non-food agricultural products in the world. However, cotton production is often hampered by the invasion of disease. In most cases, these plant diseases are a result of insect or pest infestations, which can have a significant impact on production if not addressed promptly. It is, therefore, crucial to accurately identify leaf diseases in cotton plants to prevent any negative effects on yield. This paper presents a hybrid deep learning approach based on Bidirectional Encoder Representations from Transformers with Residual network and particle swarm optimization (BERT-ResNet-PSO) for detecting cotton plant diseases. This approach starts with image pre-processing, which they pass to a BERT-like encoder after linearly embedding the image patches. It results in segregating disease regions. Then, the output of the encoded feature is passed to ResNet-based architecture for feature extraction and further optimized by PSO to increase the classification accuracy. The approach is tested on a cotton dataset from the Plant Village dataset, where the experimental results show the effectiveness of this hybrid deep learning approach, achieving an accuracy of 98.5%, precision of 98.2% and recall of 98.7% compared to the existing deep learning approaches such as ResNet50, VGG19, InceptionV3, and ResNet152V2. This study shows that the hybrid deep learning approach is capable of dealing with the cotton plant disease detection problem effectively. This study suggests that the proposed approach is beneficial to help avoid crop losses on a large scale and support effective farming management practices.
Why it matches plant phenotyping methods葉画像から綿花の病害領域を抽出・分類する深層学習手法を開発し、既存手法と比較評価しており、植物病害状態の表現型取得が中心である。
abstractThis paper presents a hybrid deep learning approach based on Bidirectional Encoder Representations from Transformers with Residual network and particle swarm optimization (BERT-ResNet-PSO) for detecting cotton 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 · publicdataset can be obtained from the following link: https://www.kaggle.com/datasets/Open asset ↗Kagglepdf-page:10 lines:1-17Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Cotton is a major cash crop, and increasing its production is extremely important worldwide, especially in agriculture-led economies. The crop is susceptible to various diseases, leading to decreased yields. In recent years, advancements in deep learning methods have enabled researchers to develop automated methods for detecting diseases in cotton crops. Such automation not only assists farmers in mitigating the effects of the disease but also conserves resources in terms of labor and fertilizer costs. However, accurate classification of multiple diseases simultaneously in cotton remains challenging due to multiple factors, including class imbalance, variation in disease symptoms, and the need for real-time detection, as most existing datasets are acquired under controlled conditions. This research proposes a novel method for addressing these challenges and accurately classifying seven classes, including six diseases and a healthy class. We address the class imbalance issue through synthetic data generation using conventional methods like scaling, rotating, transforming, shearing, and zooming and propose a customized StyleGAN for synthetic data generation. After preprocessing, we combine features extracted from MobileNet and VGG16 to create a comprehensive feature vector, passed to three classifiers: Long Short Term Memory Units, Support Vector Machines, and Random Forest. We propose a StackNet-based ensemble classifier that takes the output probabilities of these three classifiers and predicts the class label among six diseases-Bacterial blight, Curl virus, Fusarium wilt, Alternaria, Cercospora, Greymildew-and a healthy class. We trained and tested our method on publicly available datasets, achieving an average accuracy of 97%. Our robust method outperforms state-of-the-art techniques to identify the six diseases and the healthy class.
Why it matches plant phenotyping methods綿花の病害症状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、植物の病害状態を直接推定しているため。
abstractThis research proposes a novel method for addressing these challenges and accurately classifying seven classes, including six diseases and a healthy class.
Reproduction assets foundThe paper's plant-phenotyping inputs are public cotton leaf disease image datasets. The Data Availability statement names four public datasets (Kaggle Serosh Karim, Mendeley Cotton Plant Disease, and two Roboflow datasets), and the external validation section cites a fifth public Mendeley dataset. No author analysis/trDataset · publicmodels, which affirms the ability of the model to handle diverse data in the real world and validates its potential for reliable use in practical agricultural applications.
Data Availability
The data underlying the results presented in the study are available from the following sources: (1) Kaggle: Cotton Leaf Disease Dataset ( https://www.kaggle.com/datasets/seroshkarim/cotton-leaf-disease-dataset ); (2) Mendeley Data: Cotton Plant Disease Dataset ( https://data.mendeley.com/datasets/6hm6pc3y43/2 ); (3) Roboflow: Cotton Plant Disease Prediction Dataset ( https://universe.roboflow.com/national-college-of-ireland/cotton-plant-disease-prediction-igthk/dataset/3 ); (4) Roboflow: Cotton Plant DiOpen asset ↗Kagglelines:545-557Dataset · publicble use in practical agricultural applications.
Data Availability
The data underlying the results presented in the study are available from the following sources: (1) Kaggle: Cotton Leaf Disease Dataset ( https://www.kaggle.com/datasets/seroshkarim/cotton-leaf-disease-dataset ); (2) Mendeley Data: Cotton Plant Disease Dataset ( https://data.mendeley.com/datasets/6hm6pc3y43/2 ); (3) Roboflow: Cotton Plant Disease Prediction Dataset ( https://universe.roboflow.com/national-college-of-ireland/cotton-plant-disease-prediction-igthk/dataset/3 ); (4) Roboflow: Cotton Plant Disease Dataset ( https://universe.roboflow.com/roboflow-100/cotton-plant-disease/dataset/2) .
Funding Statement
This work was Open asset ↗Mendeley Datalines:545-557Dataset · publicin the study are available from the following sources: (1) Kaggle: Cotton Leaf Disease Dataset ( https://www.kaggle.com/datasets/seroshkarim/cotton-leaf-disease-dataset ); (2) Mendeley Data: Cotton Plant Disease Dataset ( https://data.mendeley.com/datasets/6hm6pc3y43/2 ); (3) Roboflow: Cotton Plant Disease Prediction Dataset ( https://universe.roboflow.com/national-college-of-ireland/cotton-plant-disease-prediction-igthk/dataset/3 ); (4) Roboflow: Cotton Plant Disease Dataset ( https://universe.roboflow.com/roboflow-100/cotton-plant-disease/dataset/2) .
Funding Statement
This work was supported by KUCARS, Department of Mechanical and Nuclear Engineering, Khalifa University under Award numberOpen asset ↗Roboflowlines:545-557Dataset · publich healthy and diseased cotton leaves across different conditions, including Bacterial blight (250 images), Cotton curl virus (431 images), Herbicide growth damage (280 images), Leaf hopper Jassids (225 images), Leaf reddening (578 images), Leaf variegation (116 images), and Healthy leaf (257 images). The dataset is available at https://data.mendeley.com/datasets/b3jy2p6k8w/2
Each image captures critical disease-specific features such as leaf discoloration, curling, wilting, necrosis, and other symptomatic indicators. The dataset is particularly valuable as it includes images collected from real field environments during different growth stages of the cotton plant. These were taken under varyOpen asset ↗lines:417-499Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
High resolution three-dimensional (3D) point clouds enable the mapping of cotton boll spatial distribution, aiding breeders in better understanding the correlation between boll positions on branches and overall yield and fiber quality. This study developed a segmentation workflow for point clouds of 18 cotton genotypes to map the spatial distribution of bolls on the plants. The data processing workflow includes two independent approaches to map the vertical and horizontal distribution of cotton bolls. The vertical distribution was mapped by segmenting bolls using PointNet++ and identifying individual instances through Euclidean clustering. For horizontal distribution, TreeQSM segmented the plant into the main stem and individual branches. PointNet++ and Euclidean clustering were then used to achieve cotton boll instance segmentation. The horizontal distribution was determined by calculating the Euclidean distance of each cotton boll relative to the main stem. Additionally, branch types were classified using point cloud meshing completion and the Dijkstra shortest path algorithm. The results highlight that the accuracy and mean intersection over union (mIoU) of the 2-class segmentation based on PointNet++ reached 0.954 and 0.896 on the whole plant dataset, and 0.968 and 0.897 on the branch dataset, respectively. The coefficient of determination (R 2 ) for the boll counting was 0.99 with a root mean squared error (RMSE) of 5.4. For the first time, this study accomplished high-granularity spatial mapping of cotton bolls and branches, but directly predicting fiber quality from 3D point clouds remains a challenge. This method provides a promising tool for 3D cotton plant mapping of different genotypes, which potentially could accelerate plant physiological studies and breeding programs.
Why it matches plant phenotyping methods3D点群の分割・個体抽出ワークフローを開発し、綿花の果実数と枝・果実の空間分布という植物形質を定量化・検証しており、フェノタイピング手法が研究の中心である。
abstractThis study developed a segmentation workflow for point clouds of 18 cotton genotypes to map the spatial distribution of bolls on the plants.
Reproduction assets foundThe authors explicitly state that the code, data, and trained PointNet++ weights for cotton boll and branch mapping are publicly available in their GitHub repository, which directly reproduces this paper's phenotyping analysis.Code · publicThe code, data, and training weights for cotton boll and branch mapping are available at https://github.com/UGA-BSAIL/cotton_organ_mapping.git .Open asset ↗UGA-BSAIL/cotton_organ_mappinglines:101-109Code / dataset availability confirmedEurope PMC · OpenAlex · Crossref · checked 6 Sept 2026
Plant height (PH) is a key agronomic trait influencing plant architecture. Suitable PH values for cotton are important for lodging resistance, high planting density, and mechanized harvesting, making it crucial to elucidate the mechanisms of the genetic regulation of PH. However, traditional field PH phenotyping largely relies on manual measurements, limiting its large-scale application. In this study, a high-throughput phenotyping platform based on UAV-mounted RGB and light detection and ranging (LiDAR) was developed to efficiently and accurately obtain time series PHs of 419 cotton accessions in the field. Different strategies were used to extract PH values from two sets of sensor data, and the extracted values were used to train using linear regression and machine learning methods to obtain PH predictions. These predictions were consistent with manual measurements of the PH for the LiDAR (R 2 = 0.934) and RGB (R 2 = 0.914) data. The predicted PH values were used for GWAS analysis, and 34 PH-related genes, two of which have been demonstrated to regulate PH in cotton, namely, GhPH1 and GhUBP15 , were identified. We further identified significant differences in the expression of a new gene named GhPH_UAV1 in the stems of the G. hirsutum cultivar ZM24 harvested on the 15th, 35th, and 70th days after sowing compared with those from a dwarf mutant ( pag1 ), which presented shortened stem and internode phenotypes. The overexpression of GhPH_UAV1 significantly promoted cotton stem development, whereas its knockout by CRISPR-Cas9 dramatically inhibited stem growth, suggesting that GhPH_UAV1 plays a positive regulatory role in cotton PH. This field-scale high-throughput phenotype monitoring platform significantly improves the ability to obtain high-quality phenotypic data from large populations, which helps overcome the imbalance between massive genotypic data and the shortage of field phenotypic data and facilitates the integration of genotype and phenotype research for crop improvement.
Why it matches plant phenotyping methodsUAV搭載RGB・LiDARによる綿花草丈の高スループット取得・推定プラットフォームの開発と精度検証が研究の中心であり、GWASや遺伝子機能解析は応用部分です。
abstracta high-throughput phenotyping platform based on UAV-mounted RGB and light detection and ranging (LiDAR) was developed to efficiently and accurately obtain time series PHs of 419 cotton accessions in the field
Reproduction assets foundThe paper's Data availability statement explicitly deposits the authors' source code, UAV-captured images, and analysis datasets in a public GitHub repository, which directly supports this paper's cotton plant-height phenotyping measurements and computational analysis.Code · publicThe source code, images captured by UAVs, data obtained from the analysis, and other datasets supporting the results presented here are available at https://github.com/Liqiangfan/419-cotton-plant-height-datasets .Open asset ↗Liqiangfan/419-cotton-plant-height-datasetslines:142-154Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Diagnosis of cotton plant diseases is essential to maintain agricultural sustainability and output. This study proposes a YOLO-based deep learning model for leaf disease detection to maximize cotton plant leaf disease detection accuracy. This method ensures a comprehensive evaluation of cotton plant health by combining various image processing techniques, improving the accuracy of disease identification. This study provides a viable path to improve crop health monitoring and management in cotton farming systems and emphasizes the importance of utilizing cutting-edge image processing techniques in agricultural activities. ROC curve performance and classification metrics were better for YOLOv5 than for VGG16 and ResNet50, as it had the highest F1 score (99.21%), recall, and precision. Consistent performance in classification tests was demonstrated by all models, which showed balanced precision, recall, and F1 scores. ResNet50 marginally outperformed VGG16 in terms of true positive rates, F1 score (98.88% vs. 98.65%), recall, and precision. More sophisticated models, such as YOLOv5 and ResNet50, showed higher efficiency and accuracy than VGG16, which makes them more appropriate for applications demanding low false positive rates and high precision. The proposed YOLO-based method improves the accuracy of disease identification, ensuring a thorough assessment of cotton plant health using image processing techniques. The results show that the proposed approach is quite successful in correctly detecting and classifying a variety of diseases that affect cotton plants.
Why it matches plant phenotyping methods綿花葉の病害状態を画像から検出・分類するYOLOベース手法の提案とモデル比較が中心であり、植物病害フェノタイピング手法に該当する。
abstractThis study proposes a YOLO-based deep learning model for leaf disease detection to maximize cotton plant leaf disease detection accuracy.
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 leaf images of cotton plants used in this research were
collected from a public Kaggle dataset [21, 22].Open asset ↗Kagglepdf-raw-page:2 lines:1-83Code / dataset availability confirmedCrossref · checked 6 Sept 2026
Published27 Jan 2025Journal of Information Systems Engineering and ManagementCited by 13 · OpenAlex ↗
Plant disease detection is a critical task in agriculture, directly impacting crop yield, food security, and sustainable farming practices. This study proposes FourCropNet, a novel deep learning model designed to detect diseases in multiple crops, including CottonLeaf, Grape, Soybean, and Corn. The model leverages an advanced architecture comprising residual blocks for efficient feature extraction, attention mechanisms to enhance focus on disease-relevant regions, and lightweight layers for computational efficiency. These components collectively enable FourCropNet to achieve superior performance across varying datasets and class complexities, from single-crop datasets to combined datasets with 15 classes. The proposed model was evaluated on diverse datasets, demonstrating high accuracy, specificity, sensitivity, and F1 scores. Notably, FourCropNet achieved the highest accuracy of 99.7% for Grape, 99.5% for Corn, and 95.3% for the combined dataset. Its scalability and ability to generalize across datasets underscore its robustness. Comparative analysis shows that FourCropNet consistently outperforms state-of-the-art models, such as MobileNet, VGG16, and EfficientNet, across various metrics. FourCropNet’s innovative design and consistent performance make it a reliable solution for real-time disease detection in agriculture. This model has the potential to assist farmers in timely disease diagnosis, reducing economic losses and promoting sustainable agricultural practices.
Why it matches plant phenotyping methods植物葉の病害状態を画像から推定するCNNモデルを開発し、複数作物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す。
abstractThis study proposes FourCropNet, a novel deep learning model designed to detect diseases in multiple crops, including CottonLeaf, Grape, Soybean, and Corn.
Reproduction assets foundThe paper evaluates FourCropNet on a public Kaggle multi-class crop disease image dataset (reference [28]), which is the plant image input used for the paper's phenotyping/disease-detection measurements. No author code or trained model checkpoints are reported as publicly available.Dataset · public[28] “20k+ Multi-Class Crop Disease Images.” Accessed: Jan. 28, 2024. [Online]. Available:
https://www.kaggle.com/datasets/jawadali1045/20k-multi-class-crop-disease-imagesOpen asset ↗jawadali1045/20k-multi-class-crop-disease-imagespdf-page:10 lines:1-59Code / dataset availability confirmedbioRxiv · checked 15 Sept 2026
Leaf osmotic potential at full turgor ({pi}0) has been used frequently to indicate turgor loss point of plant leaves. However, even a rapid measurement of{pi} 0 using osmometry is time-consuming, if numerous leaf samples need to be measured. Because of this, researchers tend to use a small sample size to determine{pi} 0 and relate it to indices of crop performance. Yet the statistical and agronomic significance of using a small sample size of{pi} 0 to indicate crop performance is not known. We address this question using field measurements and statistical resampling. Six mature leaf samples were collected at the peak bloom stage from each of the 54 cotton plots in Texas, USA in 2024. The{pi} 0 of the collected leaves were measured using an osmometer. Seed cotton yields from the field plots were measured near the end of cotton season. To test the effect of sample size on strength of the linear relation between{pi} 0 and cotton yield, 1-6 resamples of{pi} 0 were randomly drawn with replacement from the original 6 measurements per plot for the 54 plots. The resampled data of{pi} 0 were then used as independent variable to predict cotton yield. We found that, considering the labor and cost, sampling 3 or 6 leaves per plot may not make a significant difference for the linear regression between{pi} 0 and cotton yield.
Why it matches plant phenotyping methods葉の浸透ポテンシャル測定におけるサンプル数の妥当性と、収量との関係に対する影響を再サンプリングで評価しており、測定プロトコルの技術的検証が中心である。
titleDoes sample size of leaf osmotic potential affect its relationship with cotton yield?
Reproduction assets foundThe paper's field-measured leaf osmotic potential and seed cotton yield dataset, plus the authors' resampling/regression computer code, are explicitly deposited publicly on Zenodo (record 14635663), as stated in the Data availability section.Dataset · publicect 9574-
2, is appreciated. We thank Jose Teran and Joe Gonzalez,
Farm Manager and Farm Foreman, respectively, at Uvalde
Research Center, and collaborating farmer Rick Kruger for
time/efforts invested in crop management.
Data availability
The data and computer code for reproduc-
ing the results of this paper are available from
https://zenodo.org/records/14635663.Bibliography
1. Megan K. Bartlett, Ya Zhang, Christine Scoffoni, Shanwen Sun, Rico Ardy, Kunfang Cao,
and Lawren Sack. Rapid determination of comparative drought tolerance traits: using an
osmometer to predict turgor loss point. Methods in Ecology and Evolution, 3:880–888, 2012.
2. Y. N. S. Cheung, M. T. Tyree, and J. Dainty. WOpen asset ↗Zenodo · 14635663pdf-raw-page:3 lines:1-85Code / dataset availability confirmedCrossref · Europe PMC · checked 6 Sept 2026
Introduction Cotton, being a crucial cash crop globally, faces significant challenges due to multiple diseases that adversely affect its quality and yield. To identify such diseases is very important for the implementation of effective management strategies for sustainable agriculture. Image recognition plays an important role for the timely and accurate identification of diseases in cotton plants as it allows farmers to implement effective interventions and optimize resource allocation. Additionally, deep learning has begun as a powerful technique for to detect diseases in crops using images. Hence, the significance of this work lies in its potential to mitigate the impact of these diseases, which cause significant damage to the cotton and decrease fibre quality and promote sustainable agricultural practices. Methods This paper investigates the role of deep transfer learning techniques such as EfficientNet models, Xception, ResNet models, Inception, VGG, DenseNet, MobileNet, and InceptionResNet for cotton plant disease detection. A complete dataset of infected cotton plants having diseases like Bacterial Blight, Target Spot, Powdery Mildew, Aphids, and Army Worm along with the healthy ones is used. After pre-processing the images of the dataset, their region of interest is obtained by applying feature extraction techniques such as the generation of the biggest contour, identification of extreme points, cropping of relevant regions, and segmenting the objects using adaptive thresholding. Results and Discussion During experimentation, it is found that the EfficientNetB3 model outperforms in accuracy, loss, as well as root mean square error by obtaining 99.96%, 0.149, and 0.386 respectively. However, other models also show the good performance in terms of precision, recall, and F1 score, with high scores close to 0.98 or 1.00, except for VGG19. The findings of the paper emphasize the prospective of deep transfer learning as a viable technique for cotton plant disease diagnosis by providing a cost-effective and efficient solution for crop disease monitoring and management. This strategy can also help to improve agricultural practices by ensuring sustainable cotton farming and increased crop output.
Why it matches plant phenotyping methods綿花の画像から植物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手法が中心である。
abstractThis paper investigates the role of deep transfer learning techniques such as EfficientNet models, Xception, ResNet models, Inception, VGG, DenseNet, MobileNet, and InceptionResNet for cotton plant disease detection.
Reproduction assets foundThe paper analyzed a public Kaggle cotton plant disease image dataset, explicitly linked in its data availability statement. No author code or trained models are shared.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://www.kaggle.com/datasets/dhamur/cotton-plant-disease .Open asset ↗Kaggle · dhamur/cotton-plant-diseaselines:1296-1311Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Cotton is one of the major cash crop in the agriculture led economies across the world. Cotton leaf diseases affects its yield globally. Determining cotton lesions on leaves is difficult when the area is big and the size of lesions is varied. Automated cotton lesion detection is quite useful; however, it is challenging due to fewer disease class, limited size datasets, class imbalance problems, and need of comprehensive evaluation metrics. We propose a novel deep learning based method that augments the data using generative adversarial networks (GANs) to reduce the class imbalance issue and an ensemble-based method that combines the feature vector obtained from the three deep learning architectures including VGG16, Inception V3, and ResNet50. The proposed method offers a more precise, efficient and scalable method for automated detection of diseases of cotton crops. We have implemented the proposed method on publicly available dataset with seven disease and one health classes and have achieved highest accuracy of 95% and F-1 score of 98%. The proposed method performs better than existing state of the art methods.
Why it matches plant phenotyping methods綿花葉の病変・病害状態を画像から推定する深層学習手法の開発と評価が中心であり、植物病害フェノタイピングに該当する。
abstractWe propose a novel deep learning based method that augments the data using generative adversarial networks (GANs) to reduce the class imbalance issue and an ensemble-based method that combines the feature vector obtained from the three deep learning architectures including VGG16, Inception V3, and ResNet50.
Reproduction assets foundThe authors publicly released their analysis code (GitHub + Zenodo DOI) and used two publicly available Kaggle cotton leaf image datasets for their phenotyping/disease-detection analysis; all are paper-specific and actionable.Code · publicon-disease-dataset/data .
The code is available at GitHub and Zenodo:
- https://github.com/FrnazAkbar/Cotton-Lesion-Detection/tree/991640ddd25ad2fee85ee41f1bc92d1ea406a55b
- FrnazAkbar. (2024). FrnazAkbar/Cotton-Lesion-Detection: Automated Lesion Detection in Cotton Leaf Visuals using Deep Learning: Code Release (v1.0). Zenodo. https://doi.org/10.5281/zenodo.13324708 .
References
Abdalla et al. (2024) Abdalla A, Wheeler TA, Dever J, Lin Z, Arce J, Guo W. Assessing fusarium oxysporum disease severity in cotton using unmanned aerial system images and a hybrid domain adaptation deep learning time series model. Biosystems Engineering. 2024;237:220–231. doi: 10.1016/j.biosystemseng.2023.12.014.Open asset ↗Zenodo · 10.5281/zenodo.13324708lines:551-578Code · public:
The cotton plant disease data is available at Kaggle:
https://www.kaggle.com/datasets/dhamur/cotton-plant-disease/data , DOI: 10.34740/kaggle/dsv/5127834 .
The Cotton Disease Dataset is available at Kaggle: https://www.kaggle.com/datasets/janmejaybhoi/cotton-disease-dataset/data .
The code is available at GitHub and Zenodo:
- https://github.com/FrnazAkbar/Cotton-Lesion-Detection/tree/991640ddd25ad2fee85ee41f1bc92d1ea406a55b
- FrnazAkbar. (2024). FrnazAkbar/Cotton-Lesion-Detection: Automated Lesion Detection in Cotton Leaf Visuals using Deep Learning: Code Release (v1.0). Zenodo. https://doi.org/10.5281/zenodo.13324708 .
References
Abdalla et al. (2024) Abdalla A, Wheeler TA, Dever J, Lin ZOpen asset ↗GitHub · FrnazAkbar/Cotton-Lesion-Detectionlines:551-578Dataset · publice dataset to conduct an analysis that involved the application of various deep learning models, namely Inception V3, ResNet50, VGG16, and a Transfer Learning approach. This led to the development of a comprehensive ensemble of pre-trained models through training procedures. Datasets used in this study are publicly available at: https://www.kaggle.com/datasets/dhamur/cotton-plant-disease/data and https://www.kaggle.com/datasets/janmejaybhoi/cotton-disease-dataset/data .
By incorporating a diverse range of models, there is a potential to encompass a broader array of leaf attributes compared to relying solely on a singular paradigm. Inception V3, VGG 16 and ResNet 50 results are combined on theOpen asset ↗Kaggle · dhamur/cotton-plant-diseaselines:410-480Dataset · publicious deep learning models, namely Inception V3, ResNet50, VGG16, and a Transfer Learning approach. This led to the development of a comprehensive ensemble of pre-trained models through training procedures. Datasets used in this study are publicly available at: https://www.kaggle.com/datasets/dhamur/cotton-plant-disease/data and https://www.kaggle.com/datasets/janmejaybhoi/cotton-disease-dataset/data .
By incorporating a diverse range of models, there is a potential to encompass a broader array of leaf attributes compared to relying solely on a singular paradigm. Inception V3, VGG 16 and ResNet 50 results are combined on the bases of voting in order to extract a wide range of leaf features frOpen asset ↗Kaggle · janmejaybhoi/cotton-disease-datasetlines:410-480Code / dataset availability confirmedCrossref · checked 13 Sept 2026
As an essential component of the global economic crop, cotton is highly susceptible to the impact of diseases on its yield and quality. In recent years, artificial intelligence technology has been widely used in cotton crop disease recognition, but in complex backgrounds, existing technologies have certain limitations in accuracy and efficiency. To overcome these challenges, this study proposes an innovative cotton disease recognition method called CANnet, and we independently collected and constructed an image dataset containing multiple cotton diseases. Firstly, we introduced the innovatively designed Reception Field Space Channel (RFSC) module to replace traditional convolution kernels. This module combines dynamic receptive field features with traditional convolutional features to effectively utilize spatial channel attention, helping CANnet capture local and global features of images more comprehensively, thereby enhancing the expressive power of features. At the same time, the module also solves the problem of parameter sharing. To further optimize feature extraction and reduce the impact of spatial channel attention redundancy in the RFSC module, we connected a self-designed Precise Coordinate Attention (PCA) module after the RFSC module to achieve redundancy reduction. In the design of the classifier, CANnet abandoned the commonly used MLP in traditional models and instead adopted improved Kolmogorov Arnold Networks-s (KANs) for classification operations. KANs technology helps CANnet to more finely utilize extracted features for classification tasks through learnable activation functions. This is the first application of the KAN concept in crop disease recognition and has achieved excellent results. To comprehensively evaluate the performance of CANnet, we conducted extensive experiments on our cotton disease dataset and a publicly available cotton disease dataset. Numerous experimental results have shown that CANnet outperforms other advanced methods in the accuracy of cotton disease identification. Specifically, on the self-built dataset, the accuracy reached 96.3%; On the public dataset, the accuracy reached 98.6%. These results fully demonstrate the excellent performance of CANnet in cotton disease identification tasks.
Why it matches plant phenotyping methods綿花の病徴を画像から認識するCNN手法を開発し、自作および公開データセットで性能検証しており、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 an innovative cotton disease recognition method called CANnet
Reproduction assets foundThe paper evaluates CANnet on a self-built Xinjiang cotton disease dataset (no public availability statement) and on a public Kaggle cotton disease dataset explicitly cited with a URL. No author code, models, or supplementary data deposits are mentioned.Dataset · public29. Dhamodharan. Cotton Plant Disease. 2023. Available online: https://www.kaggle.com/datasets/dhamur/cotton-plant-disease
(accessed on 6 May 2024).Open asset ↗Kaggle · dhamur/cotton-plant-diseasepdf-page:21 lines:1-56Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
The creation and use of a comprehensive cotton leaf disease dataset offer significant benefits in agricultural research, precision farming, and disease management. This dataset enables the development of accurate machine learning models for early disease detection, reducing manual inspections and facilitating timely interventions. It serves as a benchmark for testing algorithms and training deep learning models, aiding in automated monitoring and decision support tools in precision agriculture. This leads to targeted interventions, reduced chemical use, and improved crop management. Global collaboration is fostered, contributing to the development of disease-resistant cotton varieties and effective management strategies, ultimately reducing economic losses and promoting sustainable farming. Field surveys conducted from October 2023 to January 2024 ensured meticulous image capture under diverse conditions. The images are categorized into eight classes, representing specific disease manifestations, pests, or environmental stress in cotton plants. The dataset comprises 2137 original images and 7000 augmented images, enhancing deep learning model training. The Inception V3 model demonstrated high performance, with an overall accuracy of 96.03 %. This underscores the dataset's potential in advancing automated disease detection in cotton agriculture.
Why it matches plant phenotyping methods綿花葉の病徴を画像で分類するデータセットを構築し、深層学習モデルのベンチマークとして評価しており、植物病害表現型の取得・解析が中心である。
abstractThe creation and use of a comprehensive cotton leaf disease dataset offer significant benefits in agricultural research, precision farming, and disease management.
Reproduction assets foundThe paper is a Data in Brief article describing the authors' own cotton leaf disease image dataset (SAR-CLD-2024), publicly deposited on Mendeley Data with a direct URL and DOI provided in the article.Dataset · publicining and evaluating machine learning models aimed at accurately classifying and diagnosing cotton leaf diseases.
Data source location
The National Cotton Research Institute field in Gazipur, Dhaka, Bangladesh
Data accessibility
Repository name: Mendeley Data
Data identification number: 10.17632/b3jy2p6k8w.2
Direct URL to data: https://data.mendeley.com/datasets/b3jy2p6k8w/2
Related research article
None
1.
Value of the Data
•
The presence of diseases such as Cotton Leaf Curl Disease and leaf hopper in cotton plants poses significant challenges to farmers worldwide, leading to substantial yield losses, reduced crop quality, and economic hardships. Timely detection and effective management ofOpen asset ↗Mendeley Data · 10.17632/b3jy2p6k8w.2lines:1-44Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Background Cotton accounts for 80% of the global natural fibre production. Its leaf hairiness affects insect resistance, fibre yield, and economic value. However, this phenotype is still qualitatively assessed by visually attributing a Genotype Hairiness Score (GHS) to a leaf/plant, or by using the HairNet deep-learning model which also outputs a GHS. Here, we introduce HairNet2, a quantitative deep-learning model which detects leaf hairs (trichomes) from images and outputs a segmentation mask and a Leaf Trichome Score (LTS). Results Trichomes of 1250 images were annotated (AnnCoT) and a combination of six Feature Extractor modules and five Segmentation modules were tested alongside a range of loss functions and data augmentation techniques. HairNet2 was further validated on the dataset used to build HairNet (CotLeaf-1), a similar dataset collected in two subsequent seasons (CotLeaf-2), and a dataset collected on two genetically diverse populations (CotLeaf-X). The main findings of this study are that (1) leaf number, environment and image position did not significantly affect results, (2) although GHS and LTS mostly correlated for individual GHS classes, results at the genotype level revealed a strong LTS heterogeneity within a given GHS class, (3) LTS correlated strongly with expert scoring of individual images. Conclusions HairNet2 is the first quantitative and scalable deep-learning model able to measure leaf hairiness. Results obtained with HairNet2 concur with the qualitative values used by breeders at both extremes of the scale (GHS 1-2, and 5-5+), but interestingly suggest a reordering of genotypes with intermediate values (GHS 3-4+). Finely ranking mild phenotypes is a difficult task for humans. In addition to providing assistance with this task, HairNet2 opens the door to selecting plants with specific leaf hairiness characteristics which may be associated with other beneficial traits to deliver better varieties.
Why it matches plant phenotyping methods綿花葉の毛状突起という植物形質を画像から定量抽出する深層学習モデルを開発し、複数データセットで検証しており、フェノタイピング手法が研究の中心である。
abstractwe introduce HairNet2, a quantitative deep-learning model which detects leaf hairs (trichomes) from images and outputs a segmentation mask and a Leaf Trichome Score (LTS).
Reproduction assets foundThe paper's Availability of data and materials section explicitly deposits the four paper-specific image/annotation datasets (AnnCoT, CotLeaf-1, CotLeaf-2, CotLeaf-X) with public CSIRO DOIs, all present in allowed_urls. No code or model checkpoint deposit is stated.Dataset · publicThe CotLeaf-1 image dataset is available at https://doi.org/10.25919/9vqw-7453 .Open asset ↗10.25919/9vqw-7453lines:256-289Dataset · publicThe CotLeaf-2 image dataset is available at https://doi.org/10.25919/v0qb-er50 .Open asset ↗10.25919/v0qb-er50lines:256-289Dataset · publicThe CotLeaf-X image dataset is available at https://doi.org/10.25919/eqhx-1x73 .Open asset ↗10.25919/eqhx-1x73lines:256-289Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
CottonGreenhouseThermalLeafClassificationPhotosynthesis / fluorescenceStress response / toleranceWater status / transpiration
Introduction Drought detection, spanning from early stress to severe conditions, plays a crucial role in maintaining productivity, facilitating recovery, and preventing plant mortality. While handheld thermal cameras have been widely employed to track changes in leaf water content and stomatal conductance, research on thermal image classification remains limited due mainly to low resolution and blurry images produced by handheld cameras. Methods In this study, we introduce a computer vision pipeline to enhance the significance of leaf-level thermal images across 27 distinct cotton genotypes cultivated in a greenhouse under progressive drought conditions. Our approach involved employing a customized software pipeline to process raw thermal images, generating leaf masks, and extracting a range of statistically relevant thermal features (e.g., min and max temperature, median value, quartiles, etc.). These features were then utilized to develop machine learning algorithms capable of assessing leaf hydration status and distinguishing between well-watered (WW) and dry-down (DD) conditions. Results Two different classifiers were trained to predict the plant treatment-random forest and multilayer perceptron neural networks-finding 75% and 78% accuracy in the treatment prediction, respectively. Furthermore, we evaluated the predicted versus true labels based on classic physiological indicators of drought in plants, including volumetric soil water content, leaf water potential, and chlorophyll a fluorescence, to provide more insights and possible explanations about the classification outputs. Discussion Interestingly, mislabeled leaves mostly exhibited notable responses in fluorescence, water uptake from the soil, and/or leaf hydration status. Our findings emphasize the potential of AI-assisted thermal image analysis in enhancing the informative value of common heterogeneous datasets for drought detection. This application suggests widening the experimental settings to be used with deep learning models, designing future investigations into the genotypic variation in plant drought response and potential optimization of water management in agricultural settings.
Why it matches plant phenotyping methods葉の熱画像からマスクと熱特徴量を抽出し、機械学習で水分状態・乾燥処理を判定する画像解析パイプラインが中心であり、植物表現型の取得・推定手法に該当する。
abstractOur approach involved employing a customized software pipeline to process raw thermal images, generating leaf masks, and extracting a range of statistically relevant thermal features
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 · publicSupplementary Table S3
Single measurements of volumetric soil water content across all collected images.Open asset ↗lines:440-465Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Introduction The study addresses challenges in detecting cotton leaf pests and diseases under natural conditions. Traditional methods face difficulties in this context, highlighting the need for improved identification techniques. Methods The proposed method involves a new model named CFNet-VoV-GCSP-LSKNet-YOLOv8s. This model is an enhancement of YOLOv8s and includes several key modifications: (1) CFNet Module. Replaces all C2F modules in the backbone network to improve multi-scale object feature fusion. (2) VoV-GCSP Module. Replaces C2F modules in the YOLOv8s head, balancing model accuracy with reduced computational load. (3) LSKNet Attention Mechanism. Integrated into the small object layers of both the backbone and head to enhance detection of small objects. (4) XIoU Loss Function. Introduced to improve the model's convergence performance. Results The proposed method achieves high performance metrics: Precision (P), 89.9%. Recall Rate (R), 90.7%. Mean Average Precision (mAP@0.5), 93.7%. The model has a memory footprint of 23.3MB and a detection time of 8.01ms. When compared with other models like YOLO v5s, YOLOX, YOLO v7, Faster R-CNN, YOLOv8n, YOLOv7-tiny, CenterNet, EfficientDet, and YOLOv8s, it shows an average accuracy improvement ranging from 1.2% to 21.8%. Discussion The study demonstrates that the CFNet-VoV-GCSP-LSKNet-YOLOv8s model can effectively identify cotton pests and diseases in complex environments. This method provides a valuable technical resource for the identification and control of cotton pests and diseases, indicating significant improvements over existing methods.
Why it matches plant phenotyping methods綿花葉の病害・害虫を自然条件下の画像から識別する新規YOLOベースモデルを開発し、精度・速度・メモリを比較検証しており、植物の病害状態の取得・推定が中心である。
abstractThe proposed method involves a new model named CFNet-VoV-GCSP-LSKNet-YOLOv8s.
Reproduction assets foundThe paper's cotton pest/disease detection model was trained on images aggregated from two public Kaggle datasets, both explicitly linked in the text and data availability statement. No author code or trained model is deposited.Dataset · publicThe data used in this study were sourced from six publicly available cotton pest and disease datasets on KAGGLE ( https://www.kaggle.com/datasets/saeedazfar/customized-cotton-disease-datasetOpen asset ↗Kaggle · saeedazfar/customized-cotton-disease-datasetlines:340-382Dataset · publicThis data can be found here: https://www.kaggle.com/datasets/paridhijain02122001/cotton-crop-disease-detection and https://www.kaggle.com/datasets/saeedazfar/customized-cotton-disease-datasetOpen asset ↗Kaggle · paridhijain02122001/cotton-crop-disease-detectionlines:635-688Code / dataset availability confirmedOpenAlex · Europe PMC · checked 7 Sept 2026
Abstract Senescence is a degenerative biological process that affects most organisms. Timing of senescence is critical for annual and perennial crops and is associated with yield and quality. Tracking time-series senescence data has previously required expert annotation and can be laborious for large-scale research. Here, a convolutional neural network (CNN) was trained on unoccupied aerial system (UAS, drone) images of individual plants of cotton (Gossypium hirsutum L.), an early application of single-plant analysis (SPA). Using images from 14 UAS flights capturing most of the senescence window, the CNN achieved 71.4% overall classification accuracy across six senescence categories, with class accuracies ranging between 46.8–89.4% despite large imbalances in numbers of images across classes. For example, the number of images ranged from 109 to 1,129 for the lowest-performing class (80% senesced) to the highest-performing class (fully healthy). The results demonstrate that minimally pre-processed UAS images can enable translatable implementations of high-throughput phenotyping using deep learning methods. This has applications for understanding fundamental plant biology, monitoring orchards and other spaced plantings, plant breeding, and genetic research.
Why it matches plant phenotyping methodsCNNとドローン時系列画像により個体ごとの綿花の老化状態を推定する手法が研究の中心であり、精度評価も実施しているため。
abstracta convolutional neural network (CNN) was trained on unoccupied aerial system (UAS, drone) images of individual plants of cotton
Reproduction assets foundThe authors state that all CNN analysis code, evaluation metrics, figure generation scripts, and the raw single-plant UAS images are publicly available in their GitHub repository, directly reproducing this paper's phenotyping analysis.Code · publicAll of the code used to assess the CNN, calculate evaluation metrics, and generate figures are available at the GitHub repository associated with this manuscript (55): https://github.com/ajdesalvio/cotton-chronology/tree/main. All files necessary to run the script, including the raw images, are available in the repository.Open asset ↗github.com/ajdesalvio/cotton-chronology · cotton-chronologylines:106-141Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Leaf gas exchange measurements are an important tool for inferring a plant's photosynthetic biochemistry. In most cases, the responses of photosynthetic CO 2 assimilation to variable intercellular CO 2 concentrations (A/C i response curves) are used to model the maximum (potential) rate of carboxylation by ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco, V cmax ) and the rate of photosynthetic electron transport at a given incident photosynthetically active radiation flux density (PAR; J PAR ). The standard Farquhar-von Caemmerer-Berry model is often used with default parameters of Rubisco kinetic values and mesophyll conductance to CO 2 (g m ) derived from tobacco that may be inapplicable across species. To study the significance of using such parameters for other species, here we measured the temperature responses of key in vitro Rubisco catalytic properties and g m in cotton (Gossypium hirsutum cv. Sicot 71) and derived V cmax and J 2000 (J PAR at 2000 µmol m -2 s -1 PAR) from cotton A/C i curves incrementally measured at 15°C-40°C using cotton and other species-specific sets of input parameters with our new automated fitting R package 'OptiFitACi'. Notably, parameterisation by a set of tobacco parameters produced unrealistic J 2000 :V cmax ratio of cmax above 15°C, up to 2.3-fold higher estimates of J 2000 and more variable estimates of V cmax and J 2000 , for our cotton data compared to model parameterisation with cotton-derived values. We determined that errors arise when using a g m,25 of 2.3 mol m -2 s -1 MPa -1 or less and Rubisco CO 2 -affinities in 21% O 2 (K C 21%O2 ) at 25°C outside the range of 46-63 Pa to model A/C i responses in cotton. We show how the A/C i modelling capabilities of 'OptiFitACi' serves as a robust, user-friendly, and flexible extension of 'plantecophys' by providing simplified temperature-sensitivity and species-specificity parameterisation capabilities to reduce variability when modelling V cmax and J 2000 .
Why it matches plant phenotyping methods植物のガス交換から光合成形質を推定する新規Rパッケージを開発し、種特異的パラメータによる推定性能を検証しているため、方法が研究の中心である。
abstractwith our new automated fitting R package 'OptiFitACi'
Reproduction assets foundThe paper's authors publicly released the OptiFitACi R package containing the fitacis4 function used for all A/Ci curve fitting analyses in this study, with an explicit GitHub URL. The phenotype data (A/Ci response measurements) are stated to be in the article's Supporting Information, which is part of the article and,Code · public(Walker et al., 2013). KC
21%O2
and Γ* were calculated as
described above for tobacco. Equation (2) in Walker et al. (2013) was
used to calculate the gm of antirbcS Arabidopsis at each temperature.
2.6 | Design and implementation of function
fitacis4 in R package ‘OptiFitACi’
A new R function fitacis4 (in package ‘OptiFitACi’; https://github.com/jsamthor/OptiFitACi/tree/master/R) was designed to enhance the A/
Ci analysis capabilities of functions fitaci, fitacis, fitacis2 in the packages
‘plantecophys’ (Duursma, 2015) and ‘plantecowrap’. The function
fitacis4 is used for the batch analysis of leaf photosynthetic gas
exchange data to estimate Vcmax and J2000 using the FvCB C3 model of
leaf pOpen asset ↗jsamthor/OptiFitACi · OptiFitACipdf-raw-page:5 lines:1-114Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Abstract Background The lint percentage of seed cotton is one the most important parameters in evaluation the seed cotton quality, which affects the price of the seed cotton during the purchase and sale. The traditional method of measuring lint percentage is labor-intensive and time-consuming, and thus there is a need for an efficient and accurate method. In recent years, classification-based machine learning and computer vision have shown promise in solving various classification tasks. Results In this study, we propose a new approach for detecting lint percentage using MobileNetV2 and transfer learning. The model is deployed on the Lint Percentage detection instrument, which can rapidly and accurately determine the lint percentage of seed cotton. We evaluated the performance of the proposed approach using a dataset of 66924 seed cotton images from different regions of China. The results from the experiments showed that the model achieved an average accuracy of 98.43% in classification with an average precision of 94.97%, an average recall of 95.26%, and an average F1-score of 95.20%. Furthermore, the proposed classification model also achieved an average ac-curacy of 97.22% in calculating the lint percentage, showing no significant difference from the performance of experts (independent-samples t test, t = 0.019, p = 0.860). Conclusions This study demonstrates the effectiveness of the MobileNetV2 model and transfer learning in calculating the lint percentage of seed cotton. The proposed approach is a promising alternative to the traditional method, offering a rapid and accurate solution for the industry.
Why it matches plant phenotyping methods種子綿のリント率という植物器官・収量関連形質を、画像と深層学習で非破壊推定する手法を開発し、専門家およびデータセットで性能評価しており、表現型取得法が研究の中心である。
abstractwe propose a new approach for detecting lint percentage using MobileNetV2 and transfer learning.
Reproduction assets foundThe authors explicitly state their LPOSC dataset of 66,924 seed cotton images in six categories is available online via a Baidu Netdisk link, which matches an allowed URL. This is a paper-specific public phenotype image dataset. No code or model checkpoint availability is stated; the declarations say data available on,Dataset · publicThe proposed dataset for LPOSC, which consists of 66924 seed cotton images and six distinct categories, was collected in a real-life scenario. This dataset is unique in its scarcity of available data sets for the study of lint percentage, making it a valuable resource for the development of algorithms for the calculation of lint percentage and a potential stimulus for further research in this area. The dataset is available online at the following link:
https://pan.baidu.com/s/12pnAShYJbaFxMItiF6KdQw?pwd=juq9Open asset ↗pan.baidu.comlines:548-820Code / dataset availability confirmedOpenAlex · checked 14 Sept 2026
Abstract Abstract. Senescence is a highly ordered biological process involving resource redistribution away from ageing tissues that affects yield and quality in annuals and perennials. Images from 14 unmanned/unoccupied/uncrewed aerial system/vehicle (UAS, UAV and drone) flights captured the senescence window across two experiments while functional principal component analysis effectively reduced the dimensionality of temporal visual senescence ratings (VSRs) and two vegetation indices: the red chromatic coordinate (RCC) index and the transformed normalized difference green and red (TNDGR) index. Convolutional neural networks trained on temporally concatenated, or ‘sandwiched’, UAS images of individual cotton plants (Gossypium hirsutum L.), allowed single-plant analysis. The first functional principal component scores (FPC1) served as the regression target across six CNN models (M1–M6). Model performance was strongest for FPC1 scores from VSRs (R2 = 0.857 and 0.886 for M1 and M4), strong for TNDGR (R2 = 0.743 and 0.745 for M3 and M6), and strong-to-moderate for RCC index (R2 = 0.619 and 0.435 for M2 and M5), with deep learning attention of each model confirmed by activation of plant pixels within saliency maps. Single-plant UAS image analysis across time enabled translatable implementations of high-throughput phenotyping by linking deep learning with functional data analysis. This has applications for fundamental plant biology, monitoring orchards or other spaced plantings, plant breeding, and genetic research.
Why it matches plant phenotyping methods単一個体の綿花について、時系列UAS画像とCNN・機能的データ解析を用いて老化表現型を推定する方法が研究の中心であり、性能評価も実施している。
abstractConvolutional neural networks trained on temporally concatenated, or ‘sandwiched’, UAS images of individual cotton plants (Gossypium hirsutum L.), allowed single-plant analysis.
Reproduction assets foundThe paper's DATA AVAILABILITY section states that all code for FPCA, ANOVA and CNN regression, plus all files needed to run the scripts including the raw single-plant UAS images, are publicly available in the authors' GitHub repository. This is a paper-specific, public, actionable asset covering both the phenotyping (cCode · public)
graduate program. OGR and SMD were partially supported
by Cotton Incorporated Awards 18-201 and 20-724, and NSF
Award 1739092.
DATA AVAILABILITY
All of the code used for FPCA, ANOVA and CNN regres-
sion is available at the GitHub repository [see Supporting
Information—Notes S1] associated with this manuscript
(DeSalvio 2024): https://github.com/ajdesalvio/cotton-sand-wiches. All files necessary to run the scripts, including the raw
images, are available in the repository.
NSF STATEMENT
Any opinion, findings and conclusions or recommendations
expressed in this material are those of the authors(s) and do not
necessarily reflect the views of the National Science Foundation.
REFERENCES
Adak A,Open asset ↗ajdesalvio/cotton-sand-wichespdf-raw-page:15 lines:1-93Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Salt stress is considered one of the primary threats to cotton production. Although cotton is found to have reasonable salt tolerance, it is sensitive to salt stress during the seedling stage. This research aimed to propose an effective method for rapidly detecting salt stress of cotton seedlings using multicolor fluorescence-multispectral reflectance imaging coupled with deep learning. A prototyping platform that can obtain multicolor fluorescence and multispectral reflectance images synchronously was developed to get different characteristics of each cotton seedling. The experiments revealed that salt stress harmed cotton seedlings with an increase in malondialdehyde and a decrease in chlorophyll content, superoxide dismutase, and catalase after 17 days of salt stress. The Relief algorithm and principal component analysis were introduced to reduce data dimension with the first 9 principal component images (PC1 to PC9) accounting for 95.2% of the original variations. An optimized EfficientNet-B2 (EfficientNet-OB2), purposely used for a fixed resource budget, was established to detect salt stress by optimizing a proportional number of convolution kernels assigned to the first convolution according to the corresponding contributions of PC1 to PC9 images. EfficientNet-OB2 achieved an accuracy of 84.80%, 91.18%, and 95.10% for 5, 10, and 17 days of salt stress, respectively, which outperformed EfficientNet-B2 and EfficientNet-OB4 with higher training speed and fewer parameters. The results demonstrate the potential of combining multicolor fluorescence-multispectral reflectance imaging with the deep learning model EfficientNet-OB2 for salt stress detection of cotton at the seedling stage, which can be further deployed in mobile platforms for high-throughput screening in the field.
Why it matches plant phenotyping methods綿実生の塩ストレス状態を画像から検出する撮像プラットフォームと深層学習手法を開発しており、植物状態の取得・抽出が研究の中心である。
abstractThis research aimed to propose an effective method for rapidly detecting salt stress of cotton seedlings using multicolor fluorescence-multispectral reflectance imaging coupled with deep learning.
Reproduction assets foundThe paper's Data Availability statement explicitly deposits the authors' EfficientNet-OB2 code and training script on GitHub at the allowed URL. No public phenotype/image dataset is stated.Code · publication and technical support for the project. H.W., B.Z., and D.Y. provided suggestions on the experiment design and discussion sections.
Competing interests: The authors declare that they have no competing interests.
Data Availability
The code and training script of EfficientNet-OB2 has been hosted to GitHub and is available at https://github.com/foddcus/EfficientNetOB .
Supplementary Materials
Supplementary 1
Figs. S1 and S2
Tables S1 and S2
Click here for additional data file.
References
1. Noreen S, Ahmad S, Fatima Z, Zakir I, Iqbal P, Nahar K, Hasanuzzaman M. Abiotic stresses mediated changes in morphophysiology of cotton plant. In: Ahmad S, Hasanuzzaman M, editors. Cotton production andOpen asset ↗foddcus/EfficientNetOBlines:363-403Code / dataset availability confirmedCrossref · checked 7 Sept 2026
Crop height is a vital indicator of growth conditions. Traditional drone image-based crop height measurement methods primarily rely on calculating the difference between the Digital Elevation Model (DEM) and the Digital Terrain Model (DTM). The calculation often needs more ground information, which remains labour-intensive and time-consuming. Moreover, the variations of terrains can further compromise the reliability of these ground models. In response to these challenges, we introduce G-DMD, a novel method based on Gated Recurrent Units (GRUs) using DEM and multispectral drone images to calculate the crop height. Our method enables the model to recognize the relation between crop height, elevation, and growth stages, eliminating reliance on DTM and thereby mitigating the effects of varied terrains. We also introduce a data preparation process to handle the unique DEM and multispectral image. Upon evaluation using a cotton dataset, our G-DMD method demonstrates a notable increase in accuracy for both maximum and average cotton height measurements, achieving a 34% and 72% reduction in Root Mean Square Error (RMSE) when compared with the traditional method. Compared to other combinations of model inputs, using DEM and multispectral drone images together as inputs results in the lowest error for estimating maximum cotton height. This approach demonstrates the potential of integrating deep learning techniques with drone-based remote sensing to achieve a more accurate, labour-efficient, and streamlined crop height assessment across varied terrains.
Why it matches plant phenotyping methodsドローンのDEM・マルチスペクトル画像から作物高を推定する手法を開発し、従来法と精度比較・検証しており、植物表現型取得が中心である。
abstractwe introduce G-DMD, a novel method based on Gated Recurrent Units (GRUs) using DEM and multispectral drone images to calculate the crop height.
Reproduction assets foundThe paper's crop-height phenotyping analysis is built on a public cotton UAV multispectral/DEM dataset deposited by Xu et al. on Figshare, which qualifies as a paper-specific, publicly actionable phenotyping input. The authors' own G-DMD code and processed data are only available upon request, so that component is not公Dataset · public47. Xu, R.; Li, C.; Paterson, A.H. UAV Multispectral. Figshare. Dataset. 2018. Available online: https://figshare.com/articles/Open asset ↗Figsharepdf-page:22 lines:1-20Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Abstract This research presents an innovative approach called Tenacious Fish Swarm Optimization based Hidden Markov Model (TFSO-HMM) for augmented accurate cotton leaf disease identification and yield prediction. Cotton leaf diseases significantly threaten crop productivity, requiring timely detection and precise prediction for effective disease management. The proposed TFSO-HMM framework combines the strengths of Tenacious Fish Swarm Optimization (TFSO) and the Hidden Markov Model (HMM) to address the challenges associated with disease identification and yield prediction in cotton plants. TFSO, a nature-inspired optimization algorithm, optimizes the classification process, enhancing the accuracy of disease identification. By harnessing the collective intelligence of fish swarms, TFSO intelligently explores the search space to identify the optimal solution. The selected information is then incorporated into the HMM framework, which captures the temporal dependencies in disease progression and yield prediction. HMM's sequential modelling approach facilitates understanding the dynamic behaviour of cotton leaf diseases over time, leading to more accurate predictions. Experimental results on a comprehensive dataset demonstrate the superior performance of the TFSO-HMM method over existing approaches in terms of accuracy and predictive capability. The augmented accuracy achieved through TFSO-HMM enables early detection and precise prediction of cotton leaf diseases, enabling timely interventions for disease management and maximizing crop yield.
Why it matches plant phenotyping methods綿花葉の病害状態を対象に、TFSO-HMMという計算手法を開発・評価して病害識別を行っており、植物の病害表現型の抽出が中心的です。
abstractThis research presents an innovative approach called Tenacious Fish Swarm Optimization based Hidden Markov Model (TFSO-HMM) for augmented accurate cotton leaf disease identification and yield prediction.
Reproduction assets foundThe paper uses the public Kaggle 'Cotton Plant Disease Dataset' as its phenotyping image dataset, with an explicit public URL. The authors' analysis code is only available on request, so it does not qualify as a public asset.Dataset · publicThe “Cotton Plant Disease Dataset” available at
https://www.kaggle.com/datasets/dhamur/cotton-plant-diseaseOpen asset ↗kaggle.com · dhamur/cotton-plant-diseasepdf-page:44 lines:1-24Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
The root is an important organ for crops to absorb water and nutrients. Complete and accurate acquisition of root phenotype information is important in root phenomics research. The in situ root research method can obtain root images without destroying the roots. In the image, some of the roots are vulnerable to soil shading, which severely fractures the root system and diminishes its structural integrity. The methods of ensuring the integrity of in situ root identification and establishing in situ root image phenotypic restoration remain to be explored. Therefore, based on the in situ root image of cotton, this study proposes a root segmentation and reconstruction strategy, improves the UNet model, and achieves precise segmentation. It also adjusts the weight parameters of EnlightenGAN to achieve complete reconstruction and employs transfer learning to implement enhanced segmentation using the results of the former two. The research results show that the improved UNet model has an accuracy of 99.2%, mIOU of 87.03%, and F1 of 92.63%. The root reconstructed by EnlightenGAN after direct segmentation has an effective reconstruction ratio of 92.46%. This study enables a transition from supervised to unsupervised training of root system reconstruction by designing a combination strategy of segmentation and reconstruction network. It achieves the integrity restoration of in situ root system pictures and offers a fresh approach to studying the phenotypic of in situ root systems, also realizes the restoration of the integrity of the in situ root image, and provides a new method for in situ root phenotype study.
Why it matches plant phenotyping methods根の画像から表現型情報を抽出・復元するセグメンテーション／再構成手法の開発が中心であり、植物表現型計測法に該当する。
abstractthis study proposes a root segmentation and reconstruction strategy, improves the UNet model, and achieves precise segmentation.
Reproduction assets foundThe authors explicitly state their analysis code (improved UNet segmentation and EnlightenGAN reconstruction pipeline) has been uploaded to a public GitHub repository, and the wheat segmentation/reconstruction results were also uploaded there. The root image dataset itself is only available upon reasonable request fromCode · publicThe code has been uploaded to github: https://github.com/jiwd123/improved_unet .Open asset ↗https://github.com/jiwd123/improved_unetlines:248-271Code / dataset availability confirmedEurope PMC · Crossref · checked 7 Sept 2026
Background Plant architecture can influence crop yield and quality. Manual extraction of architectural traits is, however, time-consuming, tedious, and error prone. The trait estimation from 3D data addresses occlusion issues with the availability of depth information while deep learning approaches enable learning features without manual design. The goal of this study was to develop a data processing workflow by leveraging 3D deep learning models and a novel 3D data annotation tool to segment cotton plant parts and derive important architectural traits. Results The Point Voxel Convolutional Neural Network (PVCNN) combining both point- and voxel-based representations of 3D data shows less time consumption and better segmentation performance than point-based networks. Results indicate that the best mIoU (89.12%) and accuracy (96.19%) with average inference time of 0.88 s were achieved through PVCNN, compared to Pointnet and Pointnet++. On the seven derived architectural traits from segmented parts, an R 2 value of more than 0.8 and mean absolute percentage error of less than 10% were attained. Conclusion This plant part segmentation method based on 3D deep learning enables effective and efficient architectural trait measurement from point clouds, which could be useful to advance plant breeding programs and characterization of in-season developmental traits. The plant part segmentation code is available at https://github.com/UGA-BSAIL/plant_3d_deep_learning .
Why it matches plant phenotyping methods3D深層学習による綿花の器官分割と建築形質抽出ワークフローを開発・比較検証しており、植物フェノタイピング手法が中心である。
abstractThe goal of this study was to develop a data processing workflow by leveraging 3D deep learning models and a novel 3D data annotation tool to segment cotton plant parts and derive important architectural traits.
Reproduction assets foundThe paper's plant part segmentation code is explicitly stated as publicly available in the authors' GitHub repository. The underlying datasets are only available on request, so they are noted as request-only.Code · publicThe plant part segmentation code is available at https://github.com/UGA-BSAIL/plant_3d_deep_learning .Open asset ↗UGA-BSAIL/plant_3d_deep_learninglines:1-72Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Crop phenology is crucial information for crop yield estimation and agricultural management. Traditionally, phenology has been observed from the ground; however Earth observation, weather and soil data have been used to capture the physiological growth of crops. In this work, we propose a new approach for the within-season phenology estimation for cotton at the field level. For this, we exploit a variety of Earth observation vegetation indices (derived from Sentinel-2) and numerical simulations of atmospheric and soil parameters. Our method is unsupervised to address the ever-present problem of sparse and scarce ground truth data that makes most supervised alternatives impractical in real-world scenarios. We applied fuzzy c-means clustering to identify the principal phenological stages of cotton and then used the cluster membership weights to further predict the transitional phases between adjacent stages. In order to evaluate our models, we collected 1,285 crop growth ground observations in Orchomenos, Greece. We introduced a new collection protocol, assigning up to two phenology labels that represent the primary and secondary growth stage in the field and thus indicate when stages are transitioning. Our model was tested against a baseline model that allowed to isolate the random agreement and evaluate its true competence. The results showed that our model considerably outperforms the baseline one, which is promising considering the unsupervised nature of the approach. The limitations and the relevant future work are thoroughly discussed. The ground observations are formatted in an ready-to-use dataset and will be available at https://github.com/Agri-Hub/cotton-phenology-dataset upon publication.
Why it matches plant phenotyping methods綿花の生育段階・遷移を衛星観測指標と環境データから推定する手法を開発し、地上観測で評価している。さらに再利用可能なデータセットも提供するため、植物フェノタイピング手法が中心である。
abstractwe propose a new approach for the within-season phenology estimation for cotton at the field level.
Reproduction assets foundThe paper's ground-observation phenology dataset (1,285 field observations with photos, labels, field geometries) is publicly released on GitHub, and the data are additionally deposited on Zenodo. No author analysis code is explicitly shared.Dataset · publicData relevant to this paper are available from Zenodo at DOI: 10.5281/zenodo.7646864 ( https://doi.org/10.5281/zenodo.7646864 ).Open asset ↗zenodo · 10.5281/zenodo.7646864lines:153-165Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Background Virtual plants can simulate the plant growth and development process through computer modeling, which assists in revealing plant growth and development patterns. Virtual plant visualization technology is a core part of virtual plant research. The major limitation of the existing plant growth visualization models is that the produced virtual plants are not realistic and cannot clearly reflect plant color, morphology and texture information. Results This study proposed a novel trait-to-image crop visualization tool named CropPainter, which introduces a generative adversarial network to generate virtual crop images corresponding to the given phenotypic information. CropPainter was first tested for virtual rice panicle generation as an example of virtual crop generation at the organ level. Subsequently, CropPainter was extended for visualizing crop plants (at the plant level), including rice, maize and cotton plants. The tests showed that the virtual crops produced by CropPainter are very realistic and highly consistent with the input phenotypic traits. The codes, datasets and CropPainter visualization software are available online. Conclusion In conclusion, our method provides a completely novel idea for crop visualization and may serve as a tool for virtual crops, which can assist in plant growth and development research.
Why it matches plant phenotyping methods与えられた表現型情報から作物画像を生成するGANベースの手法とソフトウェアを開発しており、植物表現型の可視化・再現が研究の中心である。
abstractThis study proposed a novel trait-to-image crop visualization tool named CropPainter, which introduces a generative adversarial network to generate virtual crop images corresponding to the given phenotypic information.
Reproduction assets foundThe paper explicitly states that supplementary files including datasets, trained models, software, and source code are publicly available at the authors' HZAU plant phenotyping download site and a GitHub repository. These directly reproduce the paper's phenotyping datasets and CropPainter analysis.Code · publicSupplementary files for this article, which include datasets, trained models, software as well as the source codes used in this study, are available on website: http://plantphenomics.hzau.edu.cn/usercrop/Rice/download and https://github.com/zhwang-hzau/CropPainter-master .Open asset ↗zhwang-hzau/CropPainter-masterlines:157-211Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Premise The shape of young cotton ( Gossypium ) fibers varies within and between commercial cotton species, as revealed by previous detailed analyses of one cultivar of G. hirsutum and one of G. barbadense . Both narrow and wide fibers exist in G. hirsutum cv. Deltapine 90, which may impact the quality of our most abundant renewable textile material. More efficient cellular phenotyping methods are needed to empower future research efforts. Methods We developed semi-automated imaging methods for young cotton fibers and a novel machine learning algorithm for the rapid detection of tapered (narrow) or hemisphere (wide) fibers in homogeneous or mixed populations. Results The new methods were accurate for diverse accessions of G. hirsutum and G. barbadense and at least eight times more efficient than manual methods. Narrow fibers dominated in the three G. barbadense accessions analyzed, whereas the three G. hirsutum accessions showed a mixture of tapered and hemisphere fibers in varying proportions. Discussion The use or adaptation of these improved methods will facilitate experiments with higher throughput to understand the biological factors controlling the variable shapes of young cotton fibers or other elongating single cells. This research also enables the exploration of links between early cell shape and mature cotton fiber quality in diverse field-grown cotton accessions.
Why it matches plant phenotyping methods若い綿繊維の形状を対象に、半自動イメージングと機械学習による細胞形状検出法を開発し、精度と効率を検証しているため、植物表現型取得法が中心である。
abstractMore efficient cellular phenotyping methods are needed to empower future research efforts.
Reproduction assets foundThe paper publicly releases its authors' analysis code/workflow on GitHub and the cotton fiber images of six accessions used for phenotyping on USDA Ag Data Commons, both explicitly stated in the Data Availability statement and Open Data badge sections.Code · publicComputational tools and code supporting the project analysis are available through GitHub ( https://github.com/USDA-ARS-GBRU/Cotton_Fiber_Computer_Vision/ )Open asset ↗USDA-ARS-GBRU/Cotton_Fiber_Computer_Visionlines:217-287Dataset · publicimages of the six cotton accessions used are available through USDA Ag Data Commons ( https://data.nal.usda.gov/dataset/data-efficient-imaging-and-computer-vision-detection-two-cell-shapes-young-cotton-fibers )Open asset ↗lines:217-287Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
The automatic, and accurate plant phenotyping plays important role to improve the crop yield through enabling efficient plant analysis and plant breeding studies. The 3d deep learning has allows automatic segmentation of plant parts from point cloud data. However, the network architecture is designed manually and performance is limited to prior experience. The aim of this study is to search for optimal 3d deep networks to perform the plant part segmentation. We perform the 3d neural architecture search by training a super network composed of candidate networks. Using the trained super network, the evolutionary searching is used to search for top performing architecture. The results demonstrate the searched architecture outperforms manually designed architectures by attaining mean IoU and accuracy of more than 90% and 96%, respectively. The searched architecture achieves more than 83% class-wise IoU for all main stem, branches, and boll class. These plant part segmentation method shows promising results and holds potential to be utilized by plant breeders for enhancing the production quality.
Why it matches plant phenotyping methods植物点群から茎・枝・ボールなどの器官を自動分割する3Dニューラルネットワーク探索手法を開発・評価しており、植物表現型取得が研究の中心です。
abstractThe aim of this study is to search for optimal 3d deep networks to perform the plant part segmentation.
Reproduction assets foundThe paper's cotton plant LiDAR point cloud dataset (with plant part annotations) is publicly available via a DOI in the data availability statement. No author analysis code or trained model checkpoints are reported as publicly available.Dataset · publicThe collected dataset used in this study is available at https://doi.org/10.25739/vnr9-xt59.
ACKNOWLEDGMENTS
Authors gratefully thank Dr. Shangpeng Sun and Javier Rodriguez for data collection. Authors
additionally thank Bio-sensing and Instrumentation Lab (BSAIL) members for their helpful discussions.
Authors further gratefully thank for computing resources and technical expertise from Georgia Advanced
Computing ResoOpen asset ↗10.25739/vnr9-xt59pdf-layout-page:6 lines:1-29Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Pollen grains, the male gametophytes for reproduction in higher plants, are vulnerable to various stresses that lead to loss of viability and eventually crop yield. A conventional method for assessing pollen viability is manual counting after staining, which is laborious and hinders high-throughput screening. We developed an automatic detection tool (PollenDetect) to distinguish viable and nonviable pollen based on the YOLOv5 neural network, which is adjusted to adapt to the small target detection task. Compared with manual work, PollenDetect significantly reduced detection time (from approximately 3 min to 1 s for each image). Meanwhile, PollenDetect can maintain high detection accuracy. When PollenDetect was tested on cotton pollen viability, 99% accuracy was achieved. Furthermore, the results obtained using PollenDetect show that high temperature weakened cotton pollen viability, which is highly similar to the pollen viability results obtained using 2,3,5-triphenyltetrazolium formazan quantification. PollenDetect is an open-source software that can be further trained to count different types of pollen for research purposes. Thus, PollenDetect is a rapid and accurate system for recognizing pollen viability status, and is important for screening stress-resistant crop varieties for the identification of pollen viability and stress resistance genes during genetic breeding research.
Why it matches plant phenotyping methods植物花粉の生存性という状態を画像から自動推定する深層学習ツールを開発し、手動計数および染色法と精度・速度を比較検証しているため、方法が研究の中心である。
abstractWe developed an automatic detection tool (PollenDetect) to distinguish viable and nonviable pollen based on the YOLOv5 neural network, which is adjusted to adapt to the small target detection task.
Reproduction assets foundThe paper's PollenDetect source code and model details are openly available on the authors' GitHub repository, as stated in the Data Availability Statement. The supplement describes dataset composition and annotation but does not explicitly state it contains the pollen images/annotations themselves, so only the code/reCode · publicDetails and source code of the PollenDetect model used in this study are openly available at https://github.com/Tanzhihao1998/Identification-of-pollen-activity.git/ (accessed on 1 April 2022).Open asset ↗https://github.com/Tanzhihao1998/Identification-of-pollen-activity.git/lines:99-142Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
The total boll count from a plant is one of the most important phenotypic traits for cotton breeding and is also an important factor for growers to estimate the final yield. With the recent advances in deep learning, many supervised learning approaches have been implemented to perform phenotypic trait measurement from images for various crops, but few studies have been conducted to count cotton bolls from field images. Supervised learning models require a vast number of annotated images for training, which has become a bottleneck for machine learning model development. The goal of this study is to develop both fully supervised and weakly supervised deep learning models to segment and count cotton bolls from proximal imagery. A total of 290 RGB images of cotton plants from both potted (indoor and outdoor) and in-field settings were taken by consumer-grade cameras and the raw images were divided into 4350 image tiles for further model training and testing. Two supervised models (Mask R-CNN and S-Count) and two weakly supervised approaches (WS-Count and CountSeg) were compared in terms of boll count accuracy and annotation costs. The results revealed that the weakly supervised counting approaches performed well with RMSE values of 1.826 and 1.284 for WS-Count and CountSeg, respectively, whereas the fully supervised models achieve RMSE values of 1.181 and 1.175 for S-Count and Mask R-CNN, respectively, when the number of bolls in an image patch is less than 10. In terms of data annotation costs, the weakly supervised approaches were at least 10 times more cost efficient than the supervised approach for boll counting. In the future, the deep learning models developed in this study can be extended to other plant organs, such as main stalks, nodes, and primary and secondary branches. Both the supervised and weakly supervised deep learning models for boll counting with low-cost RGB images can be used by cotton breeders, physiologists, and growers alike to improve crop breeding and yield estimation.
Why it matches plant phenotyping methods綿花ボール数という植物表現型を画像からセグメンテーション・計数する深層学習手法を開発し、教師あり・弱教師ありモデルの精度とアノテーションコストを比較しており、表現型取得手法が研究の中心である。
abstractThe goal of this study is to develop both fully supervised and weakly supervised deep learning models to segment and count cotton bolls from proximal imagery.
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 can be accessed at the following link: https://doi.org/10.6084/m9.figshare.19665096.v1 .Open asset ↗figshare · 10.6084/m9.figshare.19665096.v1lines:228-245Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Estimation of cotton yield before harvest offers many benefits to breeding programs, researchers and producers. Remote sensing enables efficient and consistent estimation of cotton yields, as opposed to traditional field measurements and surveys. The overall goal of this study was to develop a data processing pipeline to perform fast and accurate pre-harvest yield predictions of cotton breeding fields from aerial imagery using machine learning techniques. By using only a single plot image extracted from an orthomosaic map, a Support Vector Machine (SVM) classifier with four selected features was trained to identify the cotton pixels present in each plot image. The SVM classifier achieved an accuracy of 89%, a precision of 86%, a recall of 75%, and an F1-score of 80% at recognizing cotton pixels. After performing morphological image processing operations and applying a connected components algorithm, the classified cotton pixels were clustered to predict the number of cotton bolls at the plot level. Our model fitted the ground truth counts with an R 2 value of 0.93, a normalized root mean squared error of 0.07, and a mean absolute percentage error of 13.7%. This study demonstrates that aerial imagery with machine learning techniques can be a reliable, efficient, and effective tool for pre-harvest cotton yield prediction.
Why it matches plant phenotyping methods航空画像と機械学習による綿花の収量・果球数推定パイプラインを開発し、画素分類と地上計数で性能検証しており、植物表現型の取得・抽出が研究の中心である。
abstractThe overall goal of this study was to develop a data processing pipeline to perform fast and accurate pre-harvest yield predictions of cotton breeding fields from aerial imagery using machine learning techniques.
Reproduction assets foundThe paper's cotton boll classification/counting pipeline is publicly available: a Dockerized web app on Docker Hub and code with sample test images on GitHub, both explicitly stated by the authors. Raw aerial imagery/ground truth data are only available on request.Code · publicAdditionally, we will provide the code and some sample images for testing at https://github.com/Javi-RS/Cotton_Yield_Estimation .Open asset ↗Javi-RS/Cotton_Yield_Estimationlines:394-495Code / dataset availability confirmedOpenAlex · Europe PMC · checked 8 Sept 2026
BACKGROUND: Leaf hairiness (pubescence) is an important plant phenotype which regulates leaf transpiration, affects sunlight penetration, and provides increased resistance or susceptibility against certain insects. Cotton accounts for 80% of global natural fibre production, and in this crop leaf hairiness also affects fibre yield and value. Currently, this key phenotype is measured visually which is slow, laborious and operator-biased. Here, we propose a simple, high-throughput and low-cost imaging method combined with a deep-learning model, HairNet, to classify leaf images with great accuracy. RESULTS: A dataset of [Formula: see text] 13,600 leaf images from 27 genotypes of Cotton was generated. Images were collected from leaves at two different positions in the canopy (leaf 3 & leaf 4), from genotypes grown in two consecutive years and in two growth environments (glasshouse & field). This dataset was used to build a 4-part deep learning model called HairNet. On the whole dataset, HairNet achieved accuracies of 89% per image and 95% per leaf. The impact of leaf selection, year and environment on HairNet accuracy was then investigated using subsets of the whole dataset. It was found that as long as examples of the year and environment tested were present in the training population, HairNet achieved very high accuracy per image (86-96%) and per leaf (90-99%). Leaf selection had no effect on HairNet accuracy, making it a robust model. CONCLUSIONS: HairNet classifies images of cotton leaves according to their hairiness with very high accuracy. The simple imaging methodology presented in this study and the high accuracy on a single image per leaf achieved by HairNet demonstrates that it is implementable at scale. We propose that HairNet replaces the current visual scoring of this trait. The HairNet code and dataset can be used as a baseline to measure this trait in other species or to score other microscopic but important phenotypes.
Why it matches plant phenotyping methods綿花葉の毛茸という植物形質を対象に、画像取得法と深層学習モデルHairNetを開発・精度評価しており、表現型取得手法が研究の中心である。
abstractCurrently, this key phenotype is measured visually which is slow, laborious and operator-biased. Here, we propose a simple, high-throughput and low-cost imaging method combined with a deep-learning model, HairNet, to classify leaf images with great accuracy.
Reproduction assets foundThe paper publicly releases its de-identified cotton leaf hairiness image dataset (~13,600 leaf images, 27 genotypes) via the CSIRO data access portal and the HairNet analysis code via a public Bitbucket repository, both with explicit availability statements and URLs.Dataset · publicThe HairNet image dataset is available at https://doi.org/10.25919/9vqw-7453 .Open asset ↗10.25919/9vqw-7453lines:242-283Code · publicThe HairNet code is available at https://bitbucket.csiro.au/scm/sth/hairnet.git .Open asset ↗bitbucket.csiro.au/scm/sth/hairnetlines:242-283Code / dataset availability confirmedEurope PMC · checked 9 Sept 2026
The study of phenotypes that reveal mechanisms of adaptation to drought and heat stress is crucial for the development of climate resilient crops in the face of climate uncertainty. The leaf metabolome effectively summarizes stress-driven perturbations of the plant physiological status and represents an intermediate phenotype that bridges the plant genome and phenome. The objective of this study was to analyze the effect of water deficit and heat stress on the leaf metabolome of 22 genetically diverse accessions of upland cotton grown in the Arizona low desert over two consecutive years. Results revealed that membrane lipid remodeling was the main leaf mechanism of adaptation to drought. The magnitude of metabolic adaptations to drought, which had an impact on fiber traits, was found to be quantitatively and qualitatively associated with different stress severity levels during the two years of the field trial. Leaf-level hyperspectral reflectance data were also used to predict the leaf metabolite profiles of the cotton accessions. Multivariate statistical models using hyperspectral data accurately estimated ( R 2 > 0.7 in ∼34% of the metabolites) and predicted ( Q 2 > 0.5 in 15-25% of the metabolites) many leaf metabolites. Predicted values of metabolites could efficiently discriminate stressed and non-stressed samples and reveal which regions of the reflectance spectrum were the most informative for predictions. Combined together, these findings suggest that hyperspectral sensors can be used for the rapid, non-destructive estimation of leaf metabolites, which can summarize the plant physiological status.
Why it matches plant phenotyping methods葉のハイパースペクトル反射から代謝物プロファイルを非破壊推定する手法を統計モデルで評価しており、植物の生理状態の推定が中心的な方法的貢献として記述されている。
abstractLeaf-level hyperspectral reflectance data were also used to predict the leaf metabolite profiles of the cotton accessions.
Reproduction assets foundThe article's Supplementary Data 1 publicly provides best linear unbiased estimators for all fiber, metabolite, hyperspectral, and vegetation index measurements of this study, accessible via the Frontiers supplementary-material page. No author analysis code or trained model deposit is mentioned.Dataset · publicSupplementary Data 1
Best linear unbiased estimators of single accessions in the 2 years of the field experiment for all the fiber yield/quality data, metabolites, hyperspectral data, and vegetation indices.Open asset ↗lines:577-642Supplement · publicSupplementary Table 2
Repeatability values and significance of fixed effects from the linear mixed models for the fiber traits of the 22 cotton accessions in 2018.Open asset ↗lines:577-642Code / dataset availability confirmedCrossref · checked 14 Sept 2026
Abstract Fractional vegetation cover (FVC) is the key trait of interest for characterizing crop growth status in crop breeding and precision management. Accurate quantification of FVC among different breeding lines, cultivars, and growth environments is challenging, especially because of the large spatiotemporal variability in complex field conditions. This study presents an ensemble modeling strategy for phenotyping crop FVC from unmanned aerial vehicle (UAV)-based multispectral images by coupling the PROSAIL model with a gap probability model (PROSAIL-GP). Seven field experiments for four main crops were conducted, and canopy images were acquired using a UAV platform equipped with RGB and multispectral cameras. The PROSAIL-GP model successfully retrieved FVC in oilseed rape (Brassica napus L.) with coefficient of determination, root mean square error (RMSE), and relative RMSE (rRMSE) of 0.79, 0.09, and 18%, respectively. The robustness of the proposed method was further examined in rice (Oryza sativa L.), wheat (Triticum aestivum L.), and cotton (Gossypium hirsutum L.), and a high accuracy of FVC retrieval was obtained, with rRMSEs of 12%, 6%, and 6%, respectively. Our findings suggest that the proposed method can efficiently retrieve crop FVC from UAV images at a high spatiotemporal domain, which should be a promising tool for precision crop breeding.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像から作物のFVCという形態・生育形質を推定するモデルを開発し、複数作物・圃場実験で精度と頑健性を検証しており、表現型取得手法が研究の中心である。
abstractThis study presents an ensemble modeling strategy for phenotyping crop FVC from unmanned aerial vehicle (UAV)-based multispectral images by coupling the PROSAIL model with a gap probability model (PROSAIL-GP).
Reproduction assets foundThe paper's Data availability statement explicitly deposits the authors' PROSAIL-GP model code and all datasets (UAV-derived canopy reflectance/FVC measurements) in a public GitHub repository, plus detailed protocols on protocols.io. The PROSAIL model itself is a generic prior tool and is excluded.Code · publicle.
Conflict of interest
The authors declare no conflict of interest.
Data availability
Data supporting this work,such as details and source code of the PROSAIL
model used in this study,are openly available at http://teledetection.ipgp.jussieu.fr/prosail/.The code of the PROSAIL-GP model and all of the
datasets are available at https://github.com/WanLiangZJU/Crop-FVC-retrieval. The detailed protocols can be found at protocols.io (https://
dx.doi.org/10.17504/protocols.io.btmynk7w).
References
Aballa A, Cen H, Wan L, Mehmood K, He Y. 2020. Nutrient status diag-
nosis of infield oilseed rape via deep learning-enabled dynamic model. IEEE
Transactions on Industrial Informatics 17, 4379–4389.
BacOpen asset ↗WanLiangZJU/Crop-FVC-retrievalpdf-raw-page:15 lines:1-89Code / dataset availability confirmedEurope PMC · Crossref · checked 9 Sept 2026
Cotton is a significant economic crop. It is vulnerable to aphids ( Aphis gossypii Glovers) during the growth period. Rapid and early detection has become an important means to deal with aphids in cotton. In this study, the visible/near-infrared (Vis/NIR) hyperspectral imaging system (376-1044 nm) and machine learning methods were used to identify aphid infection in cotton leaves. Both tall and short cotton plants (Lumianyan 24) were inoculated with aphids, and the corresponding plants without aphids were used as control. The hyperspectral images (HSIs) were acquired five times at an interval of 5 days. The healthy and infected leaves were used to establish the datasets, with each leaf as a sample. The spectra and RGB images of each cotton leaf were extracted from the hyperspectral images for one-dimensional (1D) and two-dimensional (2D) analysis. The hyperspectral images of each leaf were used for three-dimensional (3D) analysis. Convolutional Neural Networks (CNNs) were used for identification and compared with conventional machine learning methods. For the extracted spectra, 1D CNN had a fine classification performance, and the classification accuracy could reach 98%. For RGB images, 2D CNN had a better classification performance. For HSIs, 3D CNN performed moderately and performed better than 2D CNN. On the whole, CNN performed relatively better than conventional machine learning methods. In the process of 1D, 2D, and 3D CNN visualization, the important wavelength ranges were analyzed in 1D and 3D CNN visualization, and the importance of wavelength ranges and spatial regions were analyzed in 2D and 3D CNN visualization. The overall results in this study illustrated the feasibility of using hyperspectral imaging combined with multi-dimensional CNN to detect aphid infection in cotton leaves, providing a new alternative for pest infection detection in plants.
Why it matches plant phenotyping methods綿葉のアブラムシ感染状態を、ハイパースペクトル画像とCNNで直接推定する画像ベースの植物状態フェノタイピング手法を開発・比較しており、手法が中心的である。
abstractthe visible/near-infrared (Vis/NIR) hyperspectral imaging system (376-1044 nm) and machine learning methods were used to identify aphid infection in cotton leaves.
Reproduction assets foundThe paper's data availability statement points to a public figshare deposit (DOI 10.6084/m9.figshare.13668314) containing the original contributions — the cotton leaf hyperspectral images/spectra used for the 1D/2D/3D CNN aphid-infection analysis. No code availability is stated.Dataset · publicThe original contributions presented in the study are publicly available. This data can be found here: https://doi.org/10.6084/m9.figshare.13668314 .Open asset ↗figshare · 10.6084/m9.figshare.13668314lines:721-728Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Accurate phenotype prediction of quantitative traits is paramount to enhanced plant research and breeding. Here, we report the accurate prediction of cotton fiber length, a typical quantitative trait, using 474 cotton ( Gossypium ssp.) fiber length ( GFL ) genes and nine prediction models. When the SNPs/InDels contained in 226 of the GFL genes or the expressions of all 474 GFL genes was used for fiber length prediction, a prediction accuracy of r = 0.83 was obtained, approaching the maximally possible prediction accuracy of a quantitative trait. This has improved by 116%, the prediction accuracies of the fiber length thus far achieved for genomic selection using genome-wide random DNA markers. Moreover, analysis of the GFL genes identified 125 of the GFL genes that are key to accurate prediction of fiber length, with which a prediction accuracy similar to that of all 474 GFL genes was obtained. The fiber lengths of the plants predicted with expressions of the 125 key GFL genes were significantly correlated with those predicted with the SNPs/InDels of the above 226 SNP/InDel-containing GFL genes ( r = 0.892, P = 0.000). The prediction accuracies of fiber length using both genic datasets were highly consistent across environments or generations. Finally, we found that a training population consisting of 100-120 plants was sufficient to train a model for accurate prediction of a quantitative trait using the genes controlling the trait. Therefore, the genes controlling a quantitative trait are capable of accurately predicting its phenotype, thereby dramatically improving the ability, accuracy, and efficiency of phenotype prediction and promoting gene-based breeding in cotton and other species.
Why it matches plant phenotyping methods綿花繊維長という植物形質を遺伝子情報から予測する計算手法を開発・比較し、予測精度と訓練集団サイズを検証しており、形質予測法が研究の中心である。
abstractHere, we report the accurate prediction of cotton fiber length, a typical quantitative trait, using 474 cotton ( Gossypium ssp.) fiber length ( GFL ) genes and nine prediction models.
Reproduction assets foundThe paper's fiber length phenotyping measurements and prediction inputs are reproduced in its own publicly available supplementary material hosted at the Frontiers article page: Supplementary Table 4 (TPM expression profiles of the 474 GFL genes in the RIL population), Supplementary Tables 6 and 7 (SNP/InDel genotypes)Dataset · publicoundation Collaborative Research grant (DBI-1458515). The open access publishing fees for this article have been partially covered by the Texas A&M University Open Access to Knowledge Fund (OAKFund), supported by the University Libraries.
Supplementary Material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2020.583277/full#supplementary-material
Supplementary Figure 1
Examples of validation of cotton GFL SNPs by allele-specific PCR.
Click here for additional data file.
Supplementary Figure 2
Selection of key GFL genes for GBB.
Click here for additional data file.
Supplementary Figure 3
Prediction of fiber length using diOpen asset ↗lines:130-209Code / dataset availability confirmedCrossref · checked 14 Sept 2026
CottonMaizeLaboratory / benchtopRootStem / branch2D/3D reconstructionRoot system architecture
Abstract Aims The flow of electric current in the root-soil system relates to the pathways of water and solutes, its characterization provides information on the root architecture and functioning. We developed a current source density approach with the goal of non-invasively image the current pathways in the root-soil system. Methods A current flow is applied from the plant stem to the soil, the proposed geoelectrical approach images the resulting distribution and intensity of the electric current in the root-soil system. The numerical inversion procedure underlying the approach was tested in numerical simulations and laboratory experiments with artificial metallic roots. We validated the method using rhizotron laboratory experiments on maize and cotton plants. Results Results from numerical and laboratory tests showed that our inversion approach was capable of imaging root-like distributions of the current source. In maize and cotton, roots acted as “leaky conductors”, resulting in successful imaging of the root crowns and negligible contribution of distal roots to the current flow. In contrast, the electrical insulating behavior of the cotton stems in dry soil supports the hypothesis that suberin layers can affect the mobility of ions and water. Conclusions The proposed approach with rhizotrons studies provides the first direct and concurrent characterization of the root-soil current pathways and their relationship with root functioning and architecture. This approach fills a major gap toward non-destructive imaging of roots in their natural soil environment.
Why it matches plant phenotyping methods根圏の電流経路を非侵襲的に画像化し、根の構造・機能を推定する新規手法を開発・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 developed a current source density approach with the goal of non-invasively image the current pathways in the root-soil system.
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2件を確認しました。Code · publicERT and iCSD codes, with data from this study, are maintained at https://github.com/Peruz/ERTpm and https://github.com/Peruz/icsd .Open asset ↗Peruz/ERTpmlines:342-431Code · publicERT and iCSD codes, with data from this study, are maintained at https://github.com/Peruz/ERTpm and https://github.com/Peruz/icsd .Open asset ↗Peruz/icsdlines:342-431Code / dataset availability confirmedCrossref · Europe PMC · checked 9 Sept 2026
Abstract Background Plant population density is an important factor for agricultural production systems due to its substantial influence on crop yield and quality. Traditionally, plant population density is estimated by using either field assessment or a germination-test-based approach. These approaches can be laborious and inaccurate. Recent advances in deep learning provide new tools to solve challenging computer vision tasks such as object detection, which can be used for detecting and counting plant seedlings in the field. The goal of this study was to develop a deep-learning-based approach to count plant seedlings in the field. Results Overall, the final detection model achieved F1 scores of 0.727 (at $$IOU_{all}$$ I O U all ) and 0.969 (at $$IOU_{0.5}$$ I O U 0.5 ) on the $$Seedling_{All}$$ S e e d l i n g All testing set in which images had large variations, indicating the efficacy of the Faster RCNN model with the Inception ResNet v2 feature extractor for seedling detection. Ablation experiments showed that training data complexity substantially affected model generalizability, transfer learning efficiency, and detection performance improvements due to increased training sample size. Generally, the seedling counts by the developed method were highly correlated ( $$R^2$$ R 2 = 0.98) with that found through human field assessment for 75 test videos collected in multiple locations during multiple years, indicating the accuracy of the developed approach. Further experiments showed that the counting accuracy was largely affected by the detection accuracy: the developed approach provided good counting performance for unknown datasets as long as detection models were well generalized to those datasets. Conclusion The developed deep-learning-based approach can accurately count plant seedlings in the field. Seedling detection models trained in this study and the annotated images can be used by the research community and the cotton industry to further the development of solutions for seedling detection and counting.
Why it matches plant phenotyping methods圃場画像から植物個体数（苗立ち密度）を検出・計数する深層学習手法を開発し、人手評価および複数年・地点のデータで検証しており、植物表現型取得が中心である。
abstractThe goal of this study was to develop a deep-learning-based approach to count plant seedlings in the field.
Reproduction assets foundThe paper's availability statement explicitly archives original images and annotations, source code, and testing videos in a public GitHub repository, along with pretrained models for seedling detection and counting — directly reproducing this paper's phenotyping measurements and analysis.Code · publicOriginal images and annotations, source code, and testing videos are archived in a GitHub repository ( https://github.com/UGA-BSAIL/deepseedling ) along with the instruction to run pretrained models for seedling detection and counting.Open asset ↗UGA-BSAIL/deepseedlinglines:190-267Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Plant height is a morphological characteristic of plant growth that is a useful indicator of plant stress resulting from water and nutrient deficit. While height is a relatively simple trait, it can be difficult to measure accurately, especially in crops with complex canopy architectures like cotton. This paper describes the deployment of four nadir view ultrasonic transducers (UTs), two light detection and ranging (LiDAR) systems, and an unmanned aerial system (UAS) with a digital color camera to characterize plant height in an upland cotton breeding trial. The comparison of the UTs with manual measurements demonstrated that the Honeywell and Pepperl+Fuchs sensors provided more precise estimates of plant height than the MaxSonar and db3 Pulsar sensors. Performance of the multi-angle view LiDAR and UAS technologies demonstrated that the UAS derived 3-D point clouds had stronger correlations (0.980) with the UTs than the proximal LiDAR sensors. As manual measurements require increased time and labor in large breeding trials and are prone to human error reducing repeatability, UT and UAS technologies are an efficient and effective means of characterizing cotton plant height.
Why it matches plant phenotyping methods綿花の草丈という植物形質を対象に、複数のセンサーとUASを比較・検証しており、取得手法の技術性能評価が研究の中心である。
abstractThis paper describes the deployment of four nadir view ultrasonic transducers (UTs), two light detection and ranging (LiDAR) systems, and an unmanned aerial system (UAS) with a digital color camera to characterize plant height in an upland cotton breeding trial.
Reproduction assets foundThe paper's supplementary materials, hosted publicly on MDPI, contain paper-specific plant-phenotyping results: growth curves for 2016 and 2017, plant height means and standard deviations, and repeatability estimates with standard errors. No author analysis code, raw sensor data, or UAS imagery is stated to be publiclySupplement · publicAS-based images were also found to be an effective way
to measure plant height. The LiDAR sensors explored in this study were found to be less effective
and efficient overall but may have more intrinsic value for more complex traits such as leaf and
branching angle.
Supplementary Materials: The following are available online at http://www.mdpi.com/2072-4292/11/6/700/s1,
Figure S1: Growth curves 2016, Figure S2: Growth curves 2017, Table S1: Plant height means and standard
deviations, Table S2: Repeatability and standard error.
Author Contributions: A.T., K.T., D.P., and P.A.-S. conceived of the project and its components. A.T., M.C., and
D.M.E. performed data collections along with acknowledOpen asset ↗pdf-raw-page:17 lines:1-49