Crop nutrition deficiency poses a major challenge to achieving optimal yield, particularly in smallholder farming systems where timely expert diagnosis is limited. Early detection is crucial to minimize losses and reduce unnecessary fertilizer or pesticide usage. While deep learning offers potential for automated visual diagnosis, most existing approaches operate as black boxes and lack interpretability, explainability, or actionable recommendations. In this work, we present a neurosymbolic framework for early nutrient deficiency detection in ash gourd leaves using the EarlyNSD dataset. Our approach integrates a ResNet-50 backbone with a dual-head design: a classification head for deficiency prediction and a concept-prediction head that quantifies physiologically meaningful visual patterns such as yellowing, edge discoloration, spots, and vein greenness. These concept scores are combined with predefined domain rules to guide the learning of the neural component and to generate transparent, human-aligned explanations for each diagnosis. Building on the model outputs, we incorporate a Retrieval Augmented Generation (RAG)-based pipeline along with an agricultural knowledge base to generate targeted recommendations. This approach overcomes key shortcomings of pure neural models by incorporating domain knowledge in the form of differentiable fuzzy logic rules. The study demonstrates that the proposed framework improves both classification performance and interpretability compared to standard ResNet baselines. Grad-CAM analysis demonstrates that concept-guided attention aligns with symptom-specific regions, such as yellowed areas for Nitrogen deficiency or marginal discoloration for Potassium deficiency, providing visual validation of the reasoning process. Since EarlyNSD is limited in scale and visual diversity, the results are not directly comparable to large open-field datasets. Overall, our results establish a proof of concept for integrating neural detection with symbolic reasoning, enabling interpretable, actionable, and domain-informed nutrient management for practical applications.
Why it matches plant phenotyping methods葉画像から栄養欠乏状態と症状形質を推定する解釈可能な画像解析手法が研究の中心であり、植物表現型の取得・抽出に該当する。
abstractwe present a neurosymbolic framework for early nutrient deficiency detection in ash gourd leaves using the EarlyNSD dataset.
Plant diseases affect crop quality and quantity, which significantly reduces crop production. Grapes are a popular fruit, and pumpkin is a widely consumed vegetable in the world. Early detection of grape and pumpkin leaf disease is crucial to avoiding large losses in crop quality and yield. Even though the use of unmanned aerial vehicle (UAV) remote sensing can assist in reaching a broader scale for plant leaf disease identification, the work of detection is significantly hampered by the overlapping leaves, and blurring of UAV images. This paper suggests combining object detection techniques with Real-World Enhanced Super-Resolution Generative Adversarial Network (Real-ESRGAN) for enhancing image quality to detect grape and pumpkin leaf diseases in UAV-captured images. The Atrous Spatial Pyramid Pooling (ASPP) technique is incorporated into YOLOv11 architecture to enhance the network’s feature representation capabilities by efficiently incorporating depth-wise separable convolutions and dilated convolutions with different rates. The proposed approach has shown remarkable effectiveness. In the validation, the ASPP-YOLOv11 outperforms the baseline YOLOv11 architecture in precision achieving 84% with 5.2% increase, 94.7% with 3.3% increase in mAP50, and 74% with 0.9% increase in mAP50-95. Additionally, in the testing set the ASPP architecture significantly enhances mAP50 to 91.6%, attaining 1.6% gain and obtaining 1.6% increase in mAP50-95.
Why it matches plant phenotyping methodsUAV画像から植物葉の病害を推定する画像解析アーキテクチャを開発し、既存モデルと性能比較しており、植物病害状態の取得方法が中心である。
abstractThis paper suggests combining object detection techniques with Real-World Enhanced Super-Resolution Generative Adversarial Network (Real-ESRGAN) for enhancing image quality to detect grape and pumpkin leaf diseases in UAV-captured images.
Plant 3D reconstruction using optical imaging often suffers from incomplete point clouds due to viewpoint occlusion and sensor limitations. This incompleteness hinders accurate structural representation and subsequent feature extraction for plant analysis. To address these challenges, we propose a Multi-Scale Dual-Discriminator Generative Adversarial Network (MSDDG) for plant point cloud completion. A multi-scale point cloud generator (MSPG) that integrates local and global features from raw incomplete point clouds is used for MSDDG to reconstruct complete shapes. The dual-discriminators-a multi-view projected silhouette discriminator and a spatial distance discriminator-are designed to ensure geometric realism and spatial plausibility from multiple perspectives. To train MSDDG, we created the Plant4L dataset containing four plant species (sunflower, pumpkin, luffa, and eggplant) with high-quality 3D models augmented via 3D thin plate spline transformations and virtual occlusion simulation to generate incomplete point clouds and multi-view silhouettes. Experimental results on Plant4L demonstrate that MSDDG achieves superior completion performance, with Chamfer Distance (CD), Hausdorff Distance (HD), and Uniformity Chamfer Distance (UCD) all below 0.41. Comparative evaluations confirm MSDDG's superiority over previous point cloud completion methods. The application of MSDDG for 3D reconstruction from single view further validate its effectiveness in restoring occluded plant structures.
Why it matches plant phenotyping methods植物の不完全点群を補完して3D構造を再構成する手法を開発し、植物データセット上で比較評価・検証しており、表現型取得ワークフローが中心です。
abstractwe propose a Multi-Scale Dual-Discriminator Generative Adversarial Network (MSDDG) for plant point cloud completion.
Reproduction assets foundThe paper's data availability statement explicitly states that the source code and datasets (including the Plant4L point cloud completion dataset) are publicly available at the authors' GitHub repository.Code · publicThe source code and datasets used in this study are publicly available at https://github.com/Amuro-Aznable/MSCGPCN.git .Open asset ↗https://github.com/Amuro-Aznable/MSCGPCN.git · MSCGPCNlines:415-421Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
Plant diseases pose a significant threat to global agriculture, impacting crop yields and quality. Early and accurate detection is essential for effective health management but remains challenging due to visual similarity among diseases and complex field backgrounds. This study introduces AgriMM, a novel multi-modal detection framework that integrates visual images with expert-validated textual descriptions to improve diagnostic precision. The framework features three key innovations: a Hybrid Convolutional-Attention Collaborative Backbone (HCACB) to capture both fine-grained lesions and global context; a Context-enhanced Visual-Language Path Aggregation Network (CVL-PAN) for multi-scale feature fusion; and an Adaptive Region-Text Contrastive Learning (AR-TCL) module to enforce precise semantic alignment. We constructed a comprehensive dataset comprising 30,000 images and detailed symptom descriptions across five major crops (tomato, cucumber, pepper, eggplant, and squash). Experimental results demonstrate that AgriMM achieves a mean Average Precision (mAP) of 95.2%, significantly outperforming state-of-the-art unimodal baselines by 11.6%. These findings confirm that integrating linguistic semantic priors effectively resolves visual ambiguity, providing a robust tool for precision agriculture and sustainable crop protection.
Why it matches plant phenotyping methods植物病害の症状を画像から検出・診断するマルチモーダル手法を開発し、データセットと性能比較で検証しているため、植物フェノタイピング手法が中心である。
abstractThis study introduces AgriMM, a novel multi-modal detection framework that integrates visual images with expert-validated textual descriptions to improve diagnostic precision.
Introduction. Chlorophyll plays a crucial role in absorbing and transforming light energy into a chemical form that provides organic matter production in plants. Monitoring of chlorophyll content helps to assess plant-environment interactions and the degree of influence of stress factors that are essential for yield management. Traditional laboratory methods of analyzing are time-consuming, destroying samples and unsuitable for rapid field evaluations. A more reasonable solution is to use lowcost, portable devices. Aim of the Study. The study is aimed at developing and training an ANN architecture to predict the chlorophyll content in plant leaves based on their optical density within specific visible spectrum ranges. Materials and Methods. The artificial neural network dataset was compiled from experi- mental measurements using the DP-1M densitometer and the CCM-200 chlorophyll meter. Data were collected from lettuce, pepper, tomato and zucchini leaves of different ages, which were grown in different light environments. The artificial neural network training was carried out in the Google Colab environment with subsequent adaptation of the model for using in a microcontroller device – a photocolorimeter for leaves. Results. The dataset with 1,000 entries showed that the leaf optical density range isfrom 0.57 to 2.54 relative units (red), from 0.9 to 1.66 relative units (green), and from 1.09 to 3.53 relative units (blue). According to these data, the chlorophyll content variations are from 3.1 to 156.5 relative units. In the study, there were compared six artificial neural network architectures that differed by hidden-layer neurons. The structure “32:32” had the highest accuracy (MAE = 6.64 rel. units, MAPE = 16.34%, R² = 0.8886). A simplified structure “4:4” was selected to simplify the model and improve the microcontroller efficiency. This structure maintained the performance (MAE = 6.83 rel. units, MAPE = 16.86%, R² = 0.8808) with much smaller amount of resources used – 41 weight parameters and 164 bytes of memory. A comparative evaluation with classical machine learning algorithms demonstrated the superiority of the developed model across all metrics. Discussion and Conclusion. The trained artificial neural network was implemented on a microcontroller-based photocolorimeter for leaves that enabled the non-destroying optical density measurements. The developed model allows implementing non-destroying and operational monitoring of the condition of plants, which is especially important in precision farming systems. This approach has significant potential for ecological monitoring and precision agriculture. The study results demonstrate the viability of machine learning for improving plant status assessment and developing digital agrotechnology solutions.
Why it matches plant phenotyping methods葉の光学密度からクロロフィル含量を推定するANNとマイコン実装型フォトカラリメータを開発・比較評価しており、植物形質取得が研究の中心です。
abstractThe study is aimed at developing and training an ANN architecture to predict the chlorophyll content in plant leaves based on their optical density within specific visible spectrum ranges.
Breeding salt-tolerant pumpkin cultivars is crucial for improving crop quality and yield. In this study, a high-resolution imaging-based phenotyping platform was developed to capture true leaf images of pumpkin seedlings subjected to salt stress, and plant experts conducted field assessments to determine the severity of salt damage. After image preprocessing, binary mask images were generated, and the maximum likelihood values of normalized intensity were extracted in the red, green, and blue channels to establish a salt stress status index (β) for characterizing stress levels. The β value shows a strong correlation with SPAD value,which indicates that it can effectively reflect the chlorophyll content in leaves, thereby reflecting the physiological changes in leaves affected by salt stress. A leaf texture factor (α) was employed to investigate the directional characteristics of the leaf texture, it can facilitate the effective differentiation of the clusters identified in the clustering analysis and enhance model precision by incorporating detailed leaf structural features. The performance of machine learning, deep learning, and statistical modeling approaches was compared. Statistical model integrating β and α exhibited superior predictive accuracy, with a coefficient of determination, root mean square error, and mean absolute error of 0.901, 0.057, and 0.046, respectively, in the validation dataset. Accuracy assessment among 49 germplasm accessions achieved 95.65 %, demonstrating the model’s reliability. Compared to conventional salt injury assessment, this approach offers higher efficiency and greater objectivity, enabling rapid and accurate identification of salt stress levels in pumpkin seedlings. This study provides a rapid and efficient method for assessing salt stress in pumpkin seedlings, contributing to a deeper understanding of stress response mechanisms and facilitating the selection of salt-tolerant cultivars. Moreover, these findings offer a valuable reference for salt stress identification in other plant species.
Why it matches plant phenotyping methodsカボチャ葉画像から塩ストレス状態を推定する画像ベース表現型取得・解析手法を開発し、専門家評価および49系統で性能検証しており、手法が研究の中心である。
abstracta high-resolution imaging-based phenotyping platform was developed to capture true leaf images of pumpkin seedlings subjected to salt stress
Seed germination is a critical phase strongly affected by abiotic stresses including drought and artificial seed ageing. Traditional indices like Germination Percentage (GP) and Mean Germination Time (MGT) often fail to capture complex stress responses and priming efficacy. This study introduces eight novel indices that quantitatively measure distinct physiological mechanisms: The Priming Efficiency Index (SPEI), Stress Performance Stability Index (SPSI), Germination Recovery Ratio (SGRR), and Combined Vigor Index (SCVI), among others. Tested on wheat under drought stress and priming treatments, the indices demonstrated 34.2% improvement in germination recovery with gibberellin priming compared to 25.8% with hydro-priming. The SCVI showed a 20.7% enhancement in integrated seedling performance, while SGRR achieved complete stress recovery (1.004) with gibberellin treatment. Validation across triticale and pumpkin revealed consistent performance, with cross-species correlations exceeding 0.89. Statistical analyses confirmed the novel indices' superior discriminatory power, requiring 37.6% smaller sample sizes than traditional metrics while maintaining 94% rank stability under data perturbations. These indices provide robust, mechanistically informed tools for precision phenotyping in breeding programs and seed technology research.
Why it matches plant phenotyping methods発芽・幼植物性能を定量化する新規指標を開発し、複数作物で性能検証しており、表現型測定法が研究の中心である。
abstractThis study introduces eight novel indices that quantitatively measure distinct physiological mechanisms
The Ash Gourd dataset is valuable since it was collected from the diverse regions within the district of Dhaka in Bangladesh. This dataset represents one of the first attempts to document, elicit, and categorize the health conditions of Ash Gourd (Benincasa hispida) plants in Bangladesh based on healthy samples, aphid plurality, downy mildew, leaf curl, and leaf miner-infested categories. Ash Gourd is one of the region's most important vegetables because of its nutritional and economic value; thus, it is essential to know diseases' manifestation in the improvement of agricultural productivity. The Ash Gourd dataset contains 2676 images, structured into the five categories of Healthy, Aphid, Downy Mildew, Leaf Curl, and Leaf Miner. All images in all categories are raw which can be used flexibly according to the needs of analysis and model training. Concretely, the Healthy class consists of 803 images, while the four other classes contain 1,873 images. This structured way of collecting data will, in turn, enable deeper analysis and help construct machine learning models for disease classification, hence providing worthy insights into Ash Gourd plant health.
Why it matches plant phenotyping methodsアッシュゴード葉の画像データセットを構築し、植物の健康状態・病徴カテゴリを分類するための再利用可能なデータ資源を提供しており、植物病害状態の画像ベース表現型解析が中心です。
abstractThe Ash Gourd dataset contains 2676 images, structured into the five categories of Healthy, Aphid, Downy Mildew, Leaf Curl, and Leaf Miner.
Reproduction assets foundThe paper's own ash gourd leaf image dataset (2676 images, five classes) is publicly deposited on Mendeley Data with an explicit direct URL and DOI, matching the paper's phenotyping measurements.Dataset · publicRepository name: Mendeley Data
Data identification number: 10.17632/zj4th6xvdp.2
Direct URL to data:https://data.mendeley.com/datasets/zj4th6xvdp/2Open asset ↗Mendeley Data · 10.17632/zj4th6xvdp.2html-lines:1-98Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Apr 2025Computers and Electronics in Agriculture.
Pumpkin / squashLiDAR / point cloudLeafStem / branchSegmentation
Accurately segmenting plant organs in pumpkin seedling point clouds is crucial for automated plant phenotyping and is essential for improving cultivation efficiency and optimizing breeding strategies. The segmentation of pumpkin seedling organs in point clouds presents challenges such as leaf overlap, indistinct boundaries between stems and leaves, and the morphological diversity of leaves and stems. To address these issues, we propose a high-precision pumpkin seedling point cloud organ segmentation network and construct, for the first time, a labeled dataset for pumpkin seedling point cloud organ segmentation. Firstly, to tackle the difficulties caused by leaf overlap and ambiguous stem-leaf boundaries, we present the Fused Bilinear Feature Extractor (FBFE). This method utilizes bilinear operations to combine local and global features, precisely capturing subtle feature differences at overlapping leaves and stem-leaf junctions. Secondly, to address the challenge of morphological diversity between leaves and stems, we introduce the Adaptive Multi-Scale Feature Fusion Module (AMSF). This module automatically adjusts feature fusion strategies across different scales, effectively integrating information from various levels and enhancing the model’s ability to handle morphological diversity and capture fine details. Finally, we propose the Chebyshev Particle Snow Ablation Optimizer (CPSAO) to optimize the learning rate, improving the model’s convergence speed and segmentation accuracy. Experimental results show that FACNet achieves 95.06 % mIoU, 96.87 % mPrec, 98.02 % mRec, and 97.44 % mF1 on the pumpkin seedling point cloud segmentation dataset. Compared to popular point cloud segmentation models, FACNet offers superior precision and stability in segmenting organs from pumpkin seedling point clouds.
Why it matches plant phenotyping methodsカボチャ幼苗の点群から器官を抽出するセグメンテーション手法を開発し、ラベル付きデータセットも構築しており、植物表現型取得の方法が中心的です。
abstractAccurately segmenting plant organs in pumpkin seedling point clouds is crucial for automated plant phenotyping
Yield maps and in-season forecasts help optimize agricultural practices. The traditional approaches to predicting yield during the growing season often rely on ground-based observations, which are time-consuming and labor-intensive. Remote sensing offers a promising alternative by providing frequent and spatially extensive information on crop development. In this study, we evaluated the feasibility of high-resolution satellite imagery for the early yield prediction of an under-investigated crop, Japanese squash ( Cucurbita maxima ), in a small farm in Hollister, California, over the growing seasons of 2022 and 2023 using vegetation indices, including the Normalized Difference Vegetation Index (NDVI) and the Soil-Adjusted Vegetation Index (SAVI). We identified the optimal time for yield prediction and compared the performances across satellite platforms (Sentinel-2: 10 m; PlanetScope: 3 m; SkySat: 0.5 m). Pearson's correlation coefficient ( r ) was employed to determine the dependencies between the yield and vegetation indices measured at various stages throughout the squash growing season. The results showed that SkySat-derived vegetation indices outperformed those of Sentinel-2 and PlanetScope in explaining the squash yields (R 2 = 0.75-0.76; RMSE = 0.8-1.9 tons/ha). Remote sensing showed very strong correlations with yield as early as 29 days after planting in 2022 and 37 and 76 days in 2023 for the NDVI and the SAVI, respectively. These early dates corresponded with the vegetative stages when the crop canopy became denser before fruit development. These findings highlight the utility of high-resolution imagery for in-season yield estimation and within-field variability detection. Detecting yield variability early enables timely management interventions to optimize crop productivity and resource efficiency, a critical advantage for small-scale farms, where marginal yield changes impact economic outcomes.
Why it matches plant phenotyping methods高解像度衛星画像と植生指数を用いて作物収量を生育中に推定し、複数衛星プラットフォームの性能比較と精度評価を行っており、収量フェノタイピング手法が中心である。
abstractIn this study, we evaluated the feasibility of high-resolution satellite imagery for the early yield prediction of an under-investigated crop, Japanese squash
Pumpkin / squashLeafGrowth / time-series analysisGrowth / development / phenology
The emerging wearable plant sensors demonstrate the capability of in-situ measurement of physiological and micro-environmental information of plants. However, the stretchability and breathability of current wearable plant sensors are restricted mainly due to their 2D planar structures, which interfere with plant growth and development. Here, origami-inspired 3D wearable sensors have been developed for plant growth and microclimate monitoring. Unlike 2D counterparts, the 3D sensors demonstrate theoretically infinitely high stretchability and breathability derived from the structure rather than the material. They are adjusted to 100% and 111.55 mg cm -2 ·h -1 in the optimized design. In addition to stretchability and breathability, the structural parameters are also used to control the strain distribution of the 3D sensors to enhance sensitivity and minimize interference. After integrating with corresponding sensing materials, electrodes, data acquisition and transmission circuits, and a mobile App, a miniaturized sensing system is produced with the capability of in-situ and online monitoring of plant elongation and microclimate. As a demonstration, the 3D sensors are worn on pumpkin leaves, which can accurately monitor the leaf elongation and microclimate with negligible hindrance to plant growth. Finally, the effects of the microclimate on the plant growth is resolved by analyzing the monitored data. This study would significantly promote the development of wearable plant sensors and their applications in the fields of plant phenomics, plant-environment interface, and smart agriculture.
Why it matches plant phenotyping methods植物の成長をその場・オンラインで監視するウェアラブルセンサーの開発であり、植物形質の取得手法が中心と判断できる。
titleOrigami-inspired highly stretchable and breathable 3D wearable sensors for in-situ and online monitoring of plant growth and microclimate
Determining the seed type is very important for the correct indentification of genetic material. Some plant seeds can not be classified based on their visual diversity or small size by experts. Therefore, in this study was to develop a simple, accurate and rapid using different soft computing tecniques that estimates physical parameters for pumpkin seeds. The current investigation was devoted to determining some properties, such as physical dimensions, surface area, sphericity, density, rupture energy of pumpkin seeds. The methods using in this study are; (1) Multilayer perceptron (MLP); (2) Adaptive Neuro-Fuzzy Inference Systems (ANFIS). Different statistic parameters such as coffecient of determination (R2), root mean square error (RMSE), mean absolute error (MAE) are used to evaluate performance of the methods. These selected the best models predicted for plant seeds which can be used in the soft computing tecniques determined alternative approach to estimating the physical properties of estimation and clasification pumpkin seeds.
Why it matches plant phenotyping methodsカボチャ種子の物理形質を推定・分類するソフトコンピューティング手法を開発し、誤差指標で性能評価しており、形質取得・推定法が中心である。
abstractin this study was to develop a simple, accurate and rapid using different soft computing tecniques that estimates physical parameters for pumpkin seeds.
Considering the rising concern over climate change and the need for local food security, productive blue-green roofs (PBGR) can be an effective solution to mitigate many relevant environmental issues. However, their cost of operation is high because they are intensive, and an economical operation and maintenance approach will render them as more viable alternative. Low-cost sensors with the Internet of Things can provide reliable solutions to the real-time management and distributed monitoring of such roofs through monitoring the plant as well soil conditions. This research assesses the extent to which a low-cost image sensor can be deployed to perform continuous, automated monitoring of a urban rooftop farm as a PBGR and evaluates the thermal performance of the roof for additional crops. An RGB-depth image sensor was used in this study to monitor crop growth. Images collected from weekly scans were processed by segmentation to estimate the plant heights of three crops species. The devised technique performed well for leafy and tall stem plants like okra, and the correlation between the estimated and observed growth characteristics was acceptable. For smaller plants, bright light and shadow considerably influenced the image quality, decreasing the precision. Six other crop species were monitored using a wireless sensor network to investigate how different crop varieties respond in terms of thermal performance. Celery, snow peas, and potato were measured with maximum daily cooling records, while beet and zucchini showed sound cooling effects in terms of mean daily cooling.
Why it matches plant phenotyping methodsRGB-D画像センサーとセグメンテーションにより作物の草丈を自動推定し、観測値と相関検証しており、植物表現型取得手法が中心的である。
abstractThis research assesses the extent to which a low-cost image sensor can be deployed to perform continuous, automated monitoring of a urban rooftop farm as a PBGR
Introduction Recently, plant disease detection and diagnosis procedures have become a primary agricultural concern. Early detection of plant diseases enables farmers to take preventative action, stopping the disease's transmission to other plant sections. Plant diseases are a severe hazard to food safety, but because the essential infrastructure is missing in various places around the globe, quick disease diagnosis is still difficult. The plant may experience a variety of attacks, from minor damage to total devastation, depending on how severe the infections are. Thus, early detection of plant diseases is necessary to optimize output to prevent such destruction. The physical examination of plant diseases produced low accuracy, required a lot of time, and could not accurately anticipate the plant disease. Creating an automated method capable of accurately classifying to deal with these issues is vital. Method This research proposes an efficient, novel, and lightweight DeepPlantNet deep learning (DL)-based architecture for predicting and categorizing plant leaf diseases. The proposed DeepPlantNet model comprises 28 learned layers, i.e., 25 convolutional layers (ConV) and three fully connected (FC) layers. The framework employed Leaky RelU (LReLU), batch normalization (BN), fire modules, and a mix of 3×3 and 1×1 filters, making it a novel plant disease classification framework. The Proposed DeepPlantNet model can categorize plant disease images into many classifications. Results The proposed approach categorizes the plant diseases into the following ten groups: Apple_Black_rot (ABR), Cherry_(including_sour)_Powdery_mildew (CPM), Grape_Leaf_blight_(Isariopsis_Leaf_Spot) (GLB), Peach_Bacterial_spot (PBS), Pepper_bell_Bacterial_spot (PBBS), Potato_Early_blight (PEB), Squash_Powdery_mildew (SPM), Strawberry_Leaf_scorch (SLS), bacterial tomato spot (TBS), and maize common rust (MCR). The proposed framework achieved an average accuracy of 98.49 and 99.85in the case of eight-class and three-class classification schemes, respectively. Discussion The experimental findings demonstrated the DeepPlantNet model's superiority to the alternatives. The proposed technique can reduce financial and agricultural output losses by quickly and effectively assisting professionals and farmers in identifying plant leaf diseases.
Why it matches plant phenotyping methods葉画像から植物病害状態を分類する深層学習手法を開発しており、植物の病徴・病害状態の取得と推定が研究の中心であるため。
abstractThis research proposes an efficient, novel, and lightweight DeepPlantNet deep learning (DL)-based architecture for predicting and categorizing plant leaf diseases.
Reproduction assets foundThe paper's plant leaf disease classification experiments are built entirely on two public Kaggle image datasets explicitly cited by the authors: the PlantVillage Dataset (eight-class experiment) and the Plant Disease Prediction Dataset (three-class experiment). No author code, trained model, or supplementary deposit (Dataset · publicWe verified the effectiveness and robustness of the DeepPlantNet model by using images from the publicly available Kaggle “PlantVillage Dataset” dataset ( Dataset : https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset ).Open asset ↗Kaggle · abdallahalidev/plantvillage-datasetlines:355-366Dataset · publicWe validated our model using another common, publicly accessible Kaggle dataset, “Plant Disease Prediction Dataset,” to assess and estimate the generalizability and performance of the DeepPlantNet model ( Dataset : https://www.kaggle.com/datasets/shuvranshu/plant-disease-prediction-dataset ).Open asset ↗Kaggle · shuvranshu/plant-disease-prediction-datasetlines:729-756Plant phenotyping relevance match · UnverifiedCrossref · OpenAlex · checked 14 Sept 2026
Two and three-dimensional (2D and 3D) imaging of plant samples with the scanning and transmission electron microscope (SEM, TEM) can reveal important information regarding physiological and anatomical adaptations of plants to environmental (stress) situations [1-3]. This study provides an overview of SEM and TEM techniques for the rapid evaluation of 2D ultrastructural changes in plants. Additionally, methods for the 3D reconstruction and volume extraction of plant cells based on serial section TEM (ssTEM), focused ion beam SEM (FIB-SEM) are demonstrated. Leaves of Nicotiana tabacum and Cucurbita pepo were prepared conventionally and with the help of microwave irradiation [1]. Additionally, for SEM investigations leaf replicas were made by applying dental putty [2]. SEM investigations were performed with a Versa 3D SEM (FEI, Hillsboro, OR, USA). For TEM investigations ultrathin sections (80 nm) were imaged with a JEOL 1010 TEM (JEOL, Akishima, Japan). For ssTEM, 71 sections of tobacco cells were imaged with a Zeiss EM 902 TEM (Zeiss, Oberkochen, Germany) while FIB-SEM was used to image 126 slices of pumpkin cells. Track EM (Image J) was used for 3D reconstructions and volume extractions. The use of microwave irradiation strongly reduced sample preparation time from 6 to 2h for SEM and from 3d to 5h for TEM investigations. SEM revealed that the surface of samples prepared with microwave irradiation was well preserved and comparable to those prepared conventionally (Figure 1 a & b). Stomatal and epidermal cells could be clearly distinguished (Figure 1 a & b). Samples showed signs of shrinkage (Figure 1 b) which was not observed on leaf replicas which showed a smooth surface (Figure 1 c). TEM revealed that the ultrastructure of samples prepared with microwave irradiation was well preserved and similar to those prepared conventionally (Figure 2 a & b). The cytoplasm contained chloroplasts with thylakoids and starch grains, nuclei with eu- and hetero-chromatin, mitochondria, peroxisomes, vacuoles and cell walls (Figure 2a & b). 3D reconstruction by FIB-SEM was faster and less sophisticated than 3D reconstruction by ssTEM [3]. Nevertheless, both methods delivered adequate results (Figure 2 c & d). Volume extraction revealed that tobacco cells were larger (31410 μm3) than pumpkin cells (20697 μm3) and contained more chloroplasts (175 vs. 124), mitochondria (1317 vs. 291) and peroxisomes (745 vs. 79). While individual chloroplasts, mitochondria, peroxisomes were larger in pumpkin plants (25, 53, and 50%) they covered more total volume in tobacco plants (5390, 395, 374 μm3) when compared to pumpkin plants (4762, 134, 59 μm3). Summing up, microwave-assisted sample preparation and the production of leaf replicas enabled the rapid evaluation of 2D ultrastructure of plant cells for TEM and SEM investigations. 3D reconstructions based on FIB-SEM and TEM were well suited to extract volume data of whole plant cells. These techniques are well suited to study the effects of environmental stress situations on plant ultrastructure. SEM micrographs of the surface of tobacco leaves showing stomatal (arrows) and epidermal cells. While samples prepared with the help of microwave irradiation (a) and conventionally (b) showed signs of shrinkage (arrowheads in b), leaf surface replicas with dental putty (c) showed a smooth surface. Bars = 50 μm. TEM micrographs of the 2D ultrastructure of pumpkin (a) and tobacco (b) plant leaf cells with chloroplasts (C), mitochondria (M), nuclei (N), and vacuoles (V). 3D reconstructions [modified according to 4] of pumpkin (c) and tobacco (d) plant cells using FIB-SEM (c) and ssTEM (d). Cell wall (gray), chloroplasts (green), mitochondria (red), nucleus (brown), peroxisomes (purple), and vacuole (blue). Bars=1 μm. Cubes = 3 and 4 μm3.
Why it matches plant phenotyping methods植物細胞の2D/3D画像取得、再構成、体積抽出を中心に、SEM/TEMおよびFIB-SEM手法と試料調製法を比較・評価しているため、植物フェノタイピング手法として適格。
abstractThis study provides an overview of SEM and TEM techniques for the rapid evaluation of 2D ultrastructural changes in plants.
Stereo matching is a depth perception method for plant phenotyping with high throughput. In recent years, the accuracy and real-time performance of the stereo matching models have been greatly improved. While the training process relies on specialized large-scale datasets, in this research, we aim to address the issue in building stereo matching datasets. A semi-automatic method was proposed to acquire the ground truth, including camera calibration, image registration, and disparity image generation. On the basis of this method, spinach, tomato, pepper, and pumpkin were considered for experiment, and a dataset named PlantStereo was built for reconstruction. Taking data size, disparity accuracy, disparity density, and data type into consideration, PlantStereo outperforms other representative stereo matching datasets. Experimental results showed that, compared with the disparity accuracy at pixel level, the disparity accuracy at sub-pixel level can remarkably improve the matching accuracy. More specifically, for PSMNet, the EPE and bad−3 error decreased 0.30 pixels and 2.13%, respectively. For GwcNet, the EPE and bad−3 error decreased 0.08 pixels and 0.42%, respectively. In addition, the proposed workflow based on stereo matching can achieve competitive results compared with other depth perception methods, such as Time-of-Flight (ToF) and structured light, when considering depth error (2.5 mm at 0.7 m), real-time performance (50 fps at 1046 × 606), and cost. The proposed method can be adopted to build stereo matching datasets, and the workflow can be used for depth perception in plant phenotyping.
Why it matches plant phenotyping methods植物フェノタイピング用のステレオマッチングデータセット構築法を開発し、精度・性能を検証した研究であり、表現型取得手法が中心である。
abstractStereo matching is a depth perception method for plant phenotyping with high throughput.
Abstract Canopy temperature is generally accepted as an indirect but rapid, accurate, and large-scale indicator of crop water status and is, therefore, proposed to monitor irrigation needs. Crop Water Stress Index (CWSI) is the most widely used among the existing thermal-based indicators, and its links with water stress have been demonstrated. When calculating CWSI using the empirical approach, the differential between canopy and air temperature is normalized by two thresholds, also known as baselines. The Non-water stress baseline (NWSB) in the empirical approach is calculated as the relationship between T c – T a (°C) and the vapor pressured deficit (VPD, kPa) for well-irrigated crops. The baselines display different slopes depending on the species, which have a significant impact on the computed CWSI. This study analyzed the resulting errors on CWSI due to the measurement errors of critical inputs needed for its calculation. Six crop species were selected according to their NWSB with slopes that range from − 0.5 to − 3 °C·kPa −1 and used for this analysis, assuming measurement errors ranging 0.2–1 °C for T a , 0.25–2 °C for T c , and 5–10% for relative humidity (RH). It was concluded that the effects observed on CWSI are heavily dependent on the slope of the NWSB and therefore vary across species. The calculation was very sensitive to the bias in air and canopy temperature. These errors were maximal as the slope of the NWSB was less steep. When the VPD ranged from 2 to 6.6 kPa, an error of 1 °C in measuring the air temperature affected CWSI between 28 and 83% in orange, which is the species displaying the minimum slope (− 0.5 °C kPa −1 ). On the contrary, crops with steeper baseline slopes such as squash (− 3 °C kPa −1 ) showed errors ranging between 2 and 8% for the same VPD interval. This differences among the different crops species considered in this study may be related to the contrasting coupling of the species to the atmosphere, that determines the influence of vapor pressure on the transpiration rate. This study highlights the importance of reliable climatic data and the need for accurate calibrated thermal sensors to calculate CWSI accurately.
Why it matches plant phenotyping methods作物の水分状態を表すCWSIについて、温度・湿度測定誤差の影響を解析し、熱センサーの校正精度を評価する方法検証研究である。
titleAssessing the impact of measurement errors in the calculation of CWSI for characterizing the water status of several crop species
Time activity curve (TAC) signal processing in plant positron emission tomography (PET) is a frontier nuclear science technique to bring out the quantitative fluid dynamic (FD) flow parameters of the plant vascular system and generate knowledge on crops and their sustainable management, facing the accelerating global climate change. The sparse space-time sampling of the TAC signal impairs the extraction of the FD variables, which can be determined only as averaged values with existing techniques. A data-driven approach based on a reliable FD model has never been formulated. A novel sparse data assimilation digital signal processing method is proposed, with the unique capability of a direct computation of the dynamic evolution of noise correlations between estimated and measured variables, by taking into explicit account the numerical diffusion due to the sparse sampling. The sequential time-stepping procedure estimates the spatial profile of the velocity, the diffusion coefficient and the compartmental exchange rates along the plant stem from the TAC signals. To illustrate the performance of the method, we report an example of the measurement of transport mechanisms in zucchini sprouts.
Why it matches plant phenotyping methods植物PETの疎なTAC信号から、茎内の流速・拡散係数・交換速度を推定するデータ同化型信号処理法の開発が中心であり、植物の生理状態・輸送特性を定量化するフェノタイピング手法に該当する。
abstractA novel sparse data assimilation digital signal processing method is proposed
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · OpenAlex · checked 15 Sept 2026
Main conclusion Focused ion beam scanning electron microscopy is well suited for volumetric extractions and 3D reconstructions of plant cells and its organelles. The three-dimensional (3D) reconstruction of individual plant cells is an important tool to extract volumetric data of organelles and is necessary to fully understand ultrastructural changes and adaptations of plants to their environment. Methods such as the 3D reconstruction of cells based on light microscopical images often lack the resolution necessary to clearly reconstruct all cell compartments within a cell. The 3D reconstruction of cells through serial sectioning transmission electron microscopy (ssTEM) and focused ion beam scanning electron microscopy (FIB-SEM) are powerful alternatives but not widely used in plant sciences. Here, we present a method for the 3D reconstruction and volumetric extraction of plant cells based on FIB milling and compare the results with 3D reconstructions obtained with ssTEM. When compared to 3D reconstruction based on ssTEM, FIB-SEM delivered similar results. The data extracted in this study demonstrated that tobacco cells were larger (31410 µm 3 ) than pumpkin cells (20697 µm 3 ) and contained more chloroplasts (175 vs. 124), mitochondria (1317 vs. 291) and peroxisomes (745 vs. 79). While individual chloroplasts, mitochondria, peroxisomes were larger in pumpkin plants (25, 53, and 50%, respectively) they covered more total volume in tobacco plants (5390, 395, 374 µm 3 , respectively) due to their higher number per cell when compared to pumpkin plants (4762, 134, 59 µm 3 , respectively). While image acquisition with FIB-SEM was automated, software controlled, and less difficult than ssTEM, FIB milling was slower and sections could not be revised or re-imaged as they were destroyed by the ion beam. Nevertheless, the results in this study demonstrated that both, FIB-SEM and ssTEM, are powerful tools for the 3D reconstruction of and volumetric extraction from plant cells and that there were large differences in size, number, and organelle composition between pumpkin and tobacco cells.
Why it matches plant phenotyping methods植物細胞の3D画像再構成と体積・オルガネラ形質抽出法を開発し、FIB-SEMをssTEMと比較検証しており、表現型取得法が研究の中心である。
abstractHere, we present a method for the 3D reconstruction and volumetric extraction of plant cells based on FIB milling and compare the results with 3D reconstructions obtained with ssTEM.
Agriculture practices in monocropping need to become more sustainable and one of the ways to achieve this is to reintroduce intercropping. However, quantitative data to evaluate plant growth in intercropping systems are still lacking. Unmanned aerial vehicles (UAV) have the potential to become a state-of-the-art technique for the automatic estimation of plant growth. Individual plant height is an important trait attribute for field investigation as it can be used to derive information on crop growth throughout the growing season. This study aimed to investigate the applicability of UAV-based RGB imagery combined with the structure from motion (SfM) method for estimating the individual plants height of cabbage, pumpkin, barley, and wheat in an intercropping field during a complete growing season under varying conditions. Additionally, the effect of different percentiles and buffer sizes on the relationship between UAV-estimated plant height and ground truth plant height was examined. A crop height model (CHM) was calculated as the difference between the digital surface model (DSM) and the digital terrain model (DTM). The results showed that the overall correlation coefficient (R2) values of UAV-estimated and ground truth individual plant heights for cabbage, pumpkin, barley, and wheat were 0.86, 0.94, 0.36, and 0.49, respectively, with overall root mean square error (RMSE) values of 6.75 cm, 6.99 cm, 14.16 cm, and 22.04 cm, respectively. More detailed analysis was performed up to the individual plant level. This study suggests that UAV imagery can provide a reliable and automatic assessment of individual plant heights for cabbage and pumpkin plants in intercropping but cannot be considered yet as an alternative approach for barley and wheat.
Why it matches plant phenotyping methodsUAV画像とSfMによる個体植物高の自動推定を評価・検証しており、植物形質の取得手法が研究の中心です。
abstractThis study aimed to investigate the applicability of UAV-based RGB imagery combined with the structure from motion (SfM) method for estimating the individual plants height
To increase agriculture production, accurate and fast detection of plant disease is required. Expert advice is needed to detect disease in plants, nutrition deficiencies or any other abnormalities caused by extreme weather conditions. But this process is very tedious, costly, and takes more time. In this paper, hyperspectral imaging and machine learning were used to detect different stages (early, middle, and critical stage) of the powderly mildew disease (PMD) in squash plants. An unmanned aerial vehicle (UAV) was used to collect the data from the field and Locality Preserving Discriminative Broad Learning (LPDBL) was used to distinguish the diseased and healthy plants. In addition, the ability to detect the diseased plant by the proposed method was evaluated using 10 different spectral vegetation indices (VIs). The results show the proposed method detected the disease accurately in the early, middle, and critical stages of the squash plant. The proposed method's performance is compared with six different PMDs under indoor laboratory test and UAV-based field test conditions. The comparison's results show that the LPDBL provides better accuracy in detecting disease in the squash plant.
Why it matches plant phenotyping methodsスクワッシュ植物のうどんこ病の病期・病状を、ハイパースペクトル画像、UAV、機械学習で直接推定する手法を開発・比較評価しており、植物病害フェノタイピングが中心である。
abstracthyperspectral imaging and machine learning were used to detect different stages (early, middle, and critical stage) of the powderly mildew disease (PMD) in squash plants.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 15 Sept 2026
Abstract The finely tuned balance between sources and sinks determines plant resource partitioning and regulates growth and development. Understanding and measuring metabolic indicators of source or sink limitation forms a vital part of global efforts to increase crop yield for future food security. We measured metabolic profiles of Cucurbita pepo (zucchini) grown in the field under carbon sink limitation and control conditions. We demonstrate that these profiles can be measured non‐destructively using hyperspectral reflectance at both leaf and canopy scales. Total non‐structural carbohydrates (TNC) increased 82% in sink‐limited plants; leaf mass per unit area (LMA) increased 38% and free amino acids increased 22%. Partial least‐squares regression (PLSR) models link these measured functional traits with reflectance data, enabling high‐throughput estimation of traits comprising the sink limitation response. Leaf‐ and canopy‐scale models for TNC had R 2 values of 0.93 and 0.64 and %RMSE of 13 and 38%, respectively. For LMA, R 2 values were 0.91 and 0.60 and %RMSE 7 and 14%; for free amino acids, R 2 was 0.53 and 0.21 with %RMSE 20 and 26%. Remote sensing can enable accurate, rapid detection of sink limitation in the field at the leaf and canopy scale, greatly expanding our ability to understand and measure metabolic responses to stress.
Why it matches plant phenotyping methods葉・キャノピーのハイパースペクトル反射からTNC、LMA、遊離アミノ酸などの植物機能形質をPLSRで非破壊・高スループット推定する方法が中心であり、技術性能も定量評価している。
abstractWe demonstrate that these profiles can be measured non‐destructively using hyperspectral reflectance at both leaf and canopy scales.
A drought, which can be often accompanied by increased temperature, is a key adverse factor for agricultural plants. Remote sensing of early plant changes under water shortage is a prospective way to improve plant cultivation; in particular, the sensing can be based on measurement of difference reflectance indices (RIs). We complexly analyzed the efficiency of RIs based on 400–700 nm wavelengths for revealing the influences of water shortage and short-term heating on plant seedlings. We measured spectra of reflected light in leaves of pea, wheat, and pumpkin under control and stress conditions. All possible RIs in the 400–700 nm range were calculated, significances of differences between experimental and control indices were estimated, and heatmaps of the significances were constructed. It was shown that the water shortage (pea seedlings) changed absolute values of large quantity of calculated RIs. Absolute values of some RIs were significantly changed for 1–5 or 2–5 days of the water shortage; they were strongly correlated to the potential quantum yield of photosystem II and relative water content in leaves. In contrast, the short-term heating (pea, wheat, and pumpkin seedlings) mainly influenced light-induced changes in RIs. Our results show new RIs, which are potentially sensitive to the action of stressors.
Why it matches plant phenotyping methods植物葉の反射スペクトルからストレス状態を推定する反射指数を体系的に評価し、新規指数を提案しており、表現型取得手法が中心である。
abstractRemote sensing of early plant changes under water shortage is a prospective way to improve plant cultivation; in particular, the sensing can be based on measurement of difference reflectance indices (RIs).
Limited information is available regarding the physiology of squash plants grown under waterlogging stress. The objectives of this study were to investigate the growth and physiological performances of three cucurbit species, Cucurbita maxima cultivar (cv.) OK-101 (OK) and Cucurbita moschata cv. Early Price (EP) and Strong Man (SM), in response to waterlogging conditions, and to develop a precise, integrated, and quantitative non-destructive measurement of squash genotypes under stress. All tested plants were grown in a growth chamber under optimal irrigation and growth conditions for a month, and the pot plants were then subjected to non-waterlogging (control) and waterlogging treatments for periods of 1, 3, 7, and 13 days (d), followed by a 3-d post-waterlogging recovery period after water drainage. Plants with phenotypes, such as fresh weight (FW), dry weight (DW), and dry matter (DM) of shoots and roots, and various physiological systems, including relative water content (RWC), soil and plant analysis development (SPAD) chlorophyll meter, ratio of variable/maximal fluorescence ( Fv/Fm ), quantum photosynthetic yield (YII), normalized difference vegetation index (NDVI), and photochemical reflectance index (PRI) values, responded differently to waterlogging stress in accordance with the duration of the stress period and subsequent recovery period. When plants were treated with stress for 13 d, all plants exhibited harmful effects to their leaves compared with the control, but EP squash grew better than SM and OK squashes and exhibited stronger tolerance to waterlogging and showed less injury. Changes in the fresh weight, dry weight, and dry matter of shoots and roots indicated that OK plants suffered more severely than EP plants at the 3-d drainage period. The values of RWC, SPAD, Fv/Fm , YII, NDVI, and PRI in both SM and OK plants remarkably decreased at waterlogging at the 13-d time point compared with controls under identical time periods. However, the increased levels of SPAD, Fv/Fm , YII, NDVI, and PRI observed on 7 d or 13 d of waterlogging afforded the EP plant leaf with improved waterlogged tolerance. Significant and positive correlations were observed among NDVI and PRI with SPAD, Fv/Fm , and YII, indicating that these photosynthetic indices can be useful for developing non-destructive estimations of chlorophyll content in squashes when screening for waterlogging-tolerant plants, for establishing development practices for their cultivation in fields, and for enhanced cultivation during waterlogging in frequently flooded areas.
Why it matches plant phenotyping methods水logging耐性評価のため、NDVI、PRI、SPAD、蛍光指標などを統合した非破壊・定量的な植物表現型測定法の開発が目的として明示されており、方法が中心的です。
abstractand to develop a precise, integrated, and quantitative non-destructive measurement of squash genotypes under stress.
The culture of summer squash (Cucurbita pepo) has great socioeconomic importance worldwide.Characterization of C. pepo germplasm has been predominantly performed by field evaluations, which is very time consuming.Thus, the validation of new techniques capable of optimizing time for the field germplasm selection process would be useful.We evaluated agronomic potential and genetic dissimilarity of C. pepo germplasm and gathered data to determine whether aerial images obtained by drone imaging could assist in the selection of vegetative vigor; this is the first such analysis for this crop.Sixty-five genotypes belonging to the vegetable germplasm bank of the Federal University of Uberlândia were evaluated, with three replications I.F.Beloti et al. 2 in a randomized block design.The variables evaluated were: production per plant, number of fruits per plant, leaf temperature, precocity, and the indexes SPAD (Soil Plant Analysis Development), LAI (Leaf Area Index), NDVI (Normalized Difference Vegetation Index) and NDRE (Normalized Difference Red Edge Index) the last three variables were obtained using drone imaging.Genetic divergence analysis was performed with multivariate techniques using generalized Mahalanobis distance and UPGMA clustering.Hybrid performance was compared by the Scott-Knott test.UPGMA clustering showed considerable genetic diversity, with the formation of 12 distinct groups.The largest relative contribution was from the leaf area index in the discrimination of the genotypes, demonstrating high efficiency in the validation of the image phenotyping technique.Eight genotypes stood out for yield, fruit number, precocity and high leaf area index, NDVI and NDRE values.The use of image phenotyping using NDVI and NDRE sensors was efficient to identify C. pepo genotypes that differed in plant vigor.
Why it matches plant phenotyping methodsドローン画像によるNDVI・NDRE等の植物形質推定を検証し、遺伝資源選抜への有効性を評価しており、フェノタイピング手法が中心的である。
abstractdemonstrating high efficiency in the validation of the image phenotyping technique
Why it matches plant phenotyping methods携帯型NIRSによる無傷のカボチャ果実の形態・硝酸・品質パラメータ推定について、複数機器の評価と較正モデル開発を行っており、表現型取得手法が研究の中心である。
abstractThe aim of this research was to evaluate three handheld portable NIRS instruments for in situ and real-time analysis of intact summer squashes.
Bernd Zechmann, Günther Zellnig; 3D Reconstruction of Zucchini- and Tobacco Yellow Mosaic Virus Induced Ultrastructural Changes in Plants, Microscopy and M
Why it matches plant phenotyping methods植物のウイルス誘導性超微細構造変化を3D再構成する画像ベースの表現型取得・解析が題名上の中心であり、病害状態の形態計測に該当する。
title3D Reconstruction of Zucchini- and Tobacco Yellow Mosaic Virus Induced Ultrastructural Changes in Plants
Zucchini (Cucurbita pepo L.) is a cucurbitaceous plant ranking high in economic importance among vegetable crops worldwide. Pathogen infections cause alterations in plants primary and secondary metabolism that lead to a significant decrease in crop quality and yield. Such changes can be monitored by remote and proximal sensing, providing spatial and temporal information about the infection process. Remote sensing can also provide specific signatures of disease that could be used in phenotyping and to detect a pest, forecast its evolution and predict crop yield. In this work, metabolic changes triggered by soft rot (caused by Dickeya dadantii) and powdery mildew (caused by Podosphaera fusca) on zucchini leaves have been studied by multicolour fluorescence imaging and by thermography. The fluorescence parameter F520/F680 showed statistically significant differences between infected (with D. dadantii or P. fusca) and mock-control leaves during the whole period of study. Artificial neural networks, logistic regression analyses and support vector machines trained with a set of features characterising the histograms of F520/F680 images could be used as classifiers, discriminating between healthy and infected leaves. These results show the applicability of multicolour fluorescence imaging on plant phenotyping.
Why it matches plant phenotyping methods多色蛍光画像・熱画像から感染植物の生理状態を抽出し、画像特徴量と機械学習で健全葉と感染葉を識別する手法が中心であり、植物病害表現型の実質的な取得・解析に該当する。
abstractArtificial neural networks, logistic regression analyses and support vector machines trained with a set of features characterising the histograms of F520/F680 images could be used as classifiers, discriminating between healthy and infected leaves.
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 11 Sept 2026
The negative impact of conventional farming on environment and human health make improvements on farming management mandatory. Imaging techniques are implemented in remote sensing for monitoring crop fields and plant phenotyping programs. The increasingly large size and complexity of the data obtained by these techniques, makes the implementation of powerful mathematical tools necessary in order to identify informative parameters and to apply them in precision agriculture. Multicolor fluorescence imaging is a useful approach for the study of plant defense responses to stress factors at bench scale. However, it has not been fully applied to plant phenotyping. This work evaluates the possible application of multicolor fluorescence imaging in combination with thermography for the particular case of zucchini plants affected by soft rot, caused by Dickeya dadantii. Several statistical models -based on logistic regression analysis (LRA) and artificial neural networks (ANN)- were obtained for the experimental system zucchini-D. dadantii, which classify new samples as “healthy” or “infected”. The LRA worked best in identifying high dose-infiltrated leaves (in infiltrated and non-infiltrated areas) whereas ANN offered a higher accuracy at identifying low dose-infiltrated areas. To assess the applicability of these results to cucurbits in a more general way, these models were validated for melon infected by the same pathogen, achieving accurate predictions for the infiltrated areas. The values of accuracy achieved are comparable to those found in the literature for classifiers identifying other infections based on data obtained by different techniques. Thus, MCFI in combination with thermography prove useful at providing data at lab scale that can be analyzed by machine learning. This approach could be scaled up to be applied in plant phenotyping.
Why it matches plant phenotyping methodsマルチカラー蛍光画像と熱画像を用いて感染植物の健康・感染状態を分類し、統計モデルとANNを構築・検証しており、植物病害状態の取得・推定手法が中心である。
abstractThis work evaluates the possible application of multicolor fluorescence imaging in combination with thermography for the particular case of zucchini plants affected by soft rot, caused by Dickeya dadantii.