Abstract Purpose Agrivoltaic vineyards show strong spatio-temporal variability in canopy shading, but field methods to quantify panel-induced shading at canopy scale remain limited. Shading is a key factor because it affects plant physiological and morphological traits, with potential consequences for yield and production quality. This study developed a near-surface time-lapse RGB imaging approach to derive temporally explicit shading metrics in an agrivoltaic vineyard of Vitis vinifera cv. Falanghina in Southern Italy. Methods Two representative vine positions beneath the photovoltaic structure were monitored: Agrivoltaic Shade (AVS), with greater exposure to panel-induced shading, and Agrivoltaic Light (AVL), with lower exposure. Image-based canopy shading percentage was calculated through a dedicated processing workflow and integrated with radiometric and physiological measurements, including continuous photosynthetically active radiation (PAR), canopy-level spectral photon flux measurements, photosynthetic photon flux density (PPFD), band-specific photon flux densities, red:far-red ratio (R:FR), stomatal conductance (gₛ), and leaf temperature. PAR measurements beneath the panels were compared with a full-sun control area. Results AVS showed significantly higher shading than AVL (76.14% vs 39.45%, p Conclusion The proposed workflow offers a low-cost, non-destructive tool to quantify shading dynamics and support site-specific assessment of crop microenvironments in agrivoltaic systems. The approach provides crop-relevant information for precision monitoring and management of spatially heterogeneous light conditions across different crop species. Impact The data provided in this manuscript enable the quantification of in-season photovoltaic-induced canopy shading dynamics in an agrivoltaic vineyard using proximal RGB time-lapse imaging and crop-level radiometric measurements. These metrics reflect the spatial and temporal variability of light availability within the vineyard and support site-specific assessment of crop microenvironments and precision management of agrivoltaic systems.
Why it matches plant phenotyping methodsブドウ樹冠の遮光状態をRGBタイムラプス画像から定量化する手法を開発し、専用処理ワークフローと実測値で評価しており、植物フェノタイピング手法が中心である。
abstractThis study developed a near-surface time-lapse RGB imaging approach to derive temporally explicit shading metrics in an agrivoltaic vineyard of Vitis vinifera cv. Falanghina in Southern Italy.
The automation and standardization of hyperspectral imaging, particularly at high spatial resolution, are essential for advancing plant phenomics and supporting diverse plant science research. We developed HyperBird, a hyperspectral microscopic imaging robot, to automate the acquisition of up to 351 leaf-disc samples in a single tray within approximately 2.4 h and thereby support scalable plant experiments. System calibration established a spatial resolution of 24.3 μ m (full width at half maximum; FWHM), a spectral resolution of 1.78 nm (FWHM), and a depth of field of 4.53 mm, producing hyperspectral data cubes of 2195 × 2000 × 950 pixels across the 400–1000 nm spectral range. System-level and biological-sample consistency assessments confirmed stable spectral measurements during extended scanning sessions and across independently prepared trays. Thermal evaluation confirmed that the 150 W illumination design introduced minimal sample heating during scanning. HyperBird was first evaluated using grape downy and powdery mildews, where no consistent measurable effect of repeated imaging on pathosystem development was detected under the tested experimental conditions. Subsequently, HyperBird was used to characterize spatiotemporal spectral progression in grapevine leaves inoculated with Plasmopara viticola . An automated regional spectral tracing pipeline was developed to isolate spectra from retrospectively traced regions and compare them with whole-leaf averaged spectra across 0–9 days post-inoculation (DPI). Categorical mixed-effects modeling showed that these spatially resolved regional spectra exhibited significant disease-associated spectral change by 5 DPI, with a sharp transition at 6 DPI, whereas whole-leaf spectra showed delayed and weaker disease-aligned changes beginning at 7 DPI. These results demonstrate that HyperBird enables high-throughput, spatially resolved hyperspectral imaging for quantifying localized plant disease progression and provides a scalable platform for studying spectral biology across plant phenotyping applications.
Why it matches plant phenotyping methodsHyperBirdの高スループット・ハイパースペクトル画像取得ロボットを開発し、校正、再現性、加熱影響、病害進展の定量性能を検証しており、植物フェノタイピング手法が中心である。
abstractWe developed HyperBird, a hyperspectral microscopic imaging robot, to automate the acquisition of up to 351 leaf-disc samples in a single tray within approximately 2.4 h and thereby support scalable plant experiments.
Reproduction assets foundThe paper's Data Availability statement points to a public GitHub repository containing the authors' code and processed data supporting the phenotyping analyses; raw hyperspectral image data are request-only.Code · publicData Availability
857
The code and processed data supporting the findings of this study are available in the GitHub
858
repository at https://github.com/jy773Cornell/HyperBird-Robot. Raw hyperspectral image
859
data are available from the corresponding author upon reasonable request due to file size and storage
860
constraints.
861
Supplementary Materials
862
Supplementary materials accompany this article as a separate document (supplementary.pdf).
863
Supplementary Figure S1. Representative GSAM-based segOpen asset ↗HyperBird-Robotpdf-raw-page:39 lines:1-75Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · checked 14 Sept 2026
Abstract Grapevine leaves have dorsiventral anatomy with distinct adaxial (upper) and abaxial (lower) surfaces. Although morphological descriptor lists and ampelographic literature provide information on both the upper and lower side characteristics, in practice, the color traits of the upper side have become the focus of scientific publications. This study introduces the practical application of the recently developed LeafLaminaMap software and the use of trichromatic color indices in grapevine characterization. We aimed to compare colorimetric information on the adaxial and abaxial leaf surfaces as well as to explore the potential of machine learning models in classification. Five statistical descriptors (mean, standard deviation, contrast, energy, and entropy) were calculated for 25 RGB-based color indices on both leaf surfaces of 120 samples collected from four grapevine cultivars (‘Chardonnay’, ‘Pinot noir’, ‘Sauvignon blanc’, and ‘Syrah’). Data was subjected to multivariate statistical analysis and machine learning classifiers. Results showed that the abaxial leaf surface had stronger cultivar-specific color signatures, supporting its suitability for cultivar discrimination. These findings suggest that RGB-based analysis of both adaxial and abaxial leaf surfaces has potential for grapevine cultivar discrimination, offering a new perspective for cost-efficient plant phenotyping.
Why it matches plant phenotyping methodsRGB画像解析とLeafLaminaMap、色指数、機械学習を用いた葉面形質抽出・品種識別が研究の中心であり、植物フェノタイピング手法の実質的応用に該当する。
abstractThis study introduces the practical application of the recently developed LeafLaminaMap software and the use of trichromatic color indices in grapevine characterization.
Adventitious root formation is critical for the cost-effective propagation of grapevine rootstocks, and poor rooting limits the adoption of new rootstocks derived from underutilized Vitis L. species. We evaluated 308 accessions representing 18 Vitis species over three growing seasons, scoring rooting at two developmental stages, callus-stage and post-transplant, together with root biomass, cutting weight, and a derived transplant- response index. Phenotypic variation was extensive within and among species, and species rankings depended on the trait considered: V. riparia , V. rupestris and V. californica ranked among the top five species for all four rooting traits, whereas V. arizonica and V. acerifolia rooted well at the callus stage but only intermediately after transplanting. Repeatability was moderate to high for root weight and callusstage rooting and lower for post-transplant rooting. Between-species differences accounted for most of the genetic variance in callus-stage rooting but little of that in cutting weight. After removing differences among species, accessions originating from wild sites with lower dry-season precipitation rooted better and produced more root biomass. Genome-wide association analysis of 3.4 million single-nucleotide polymorphisms identified 54 significant markers resolving into 18 independent loci across four traits. Candidate genes implicate auxin-linked cell proliferation, cell wall and lignin remodeling, and solute transport. Genomic and phenomic prediction achieved moderate accuracies across traits and seasons, including for previously unevaluated accessions, and combining spectral with genotypic data improved performance; accuracy was essentially flat between 5,000 and 50,000 markers. These results provide a framework for broadening the germplasm base of grapevine rootstock breeding. Plain Language Summary Grapevines are almost always grown as two plants joined together: a fruiting variety grafted onto a rootstock that supplies the root system. Nurseries build these plants from dormant cuttings, so a rootstock is only useful in practice if its cuttings root easily. Almost all commercial rootstocks descend from just three wild North American grape species, in part because cuttings of other species are believed to root poorly. We grew cuttings from 308 wild and cultivated grapevines representing 18 species over three years and measured how well each one rooted, first in the callusing room and again after the young plants were transplanted. Rooting ability varied a great deal, and several species outside the usual three rooted as well as the standards. Vines originally collected from places with drier summers tended to root best. We also located regions of the grape genome linked to rooting, and showed that the rooting ability of a vine can be predicted from its DNA or from light reflected by its leaves. Together, these results give breeders a way to screen a much wider range of wild grapevines for rootstock development before testing them in a nursery. Core Ideas Rooting ability varies widely across 18 Vitis species, well beyond the three used in rootstock breeding. Callus-stage and post-transplant rooting behave as genetically distinct stages of root formation. Accessions from collection sites with drier summers rooted better and produced more root biomass. GWAS resolved 18 loci implicating auxin signaling, cell wall remodeling, and solute transport. Genomic and phenomic prediction reached moderate accuracy and was insensitive to marker density.
Why it matches plant phenotyping methods発根形質を対象に、スペクトル情報を用いたフェノミック予測を実施し、遺伝子型情報との統合や予測精度を評価しているため、形質取得・推定手法が研究の主要な技術的要素です。
abstractGenomic and phenomic prediction achieved moderate accuracies across traits and seasons, including for previously unevaluated accessions, and combining spectral with genotypic data improved performance
Plant disease and plant stress early warning systems have advanced through deep learning, remote sensing, digital phenotyping, disease forecasting, and sensor networks. Detection accuracy, precision, recall, F1-score, and area under the curve remain indispensable, but they are insufficient for judging whether a warning can support timely and proportionate phytoprotection under field variability. This Mini Review argues that intelligent plant health warning systems should be evaluated not only as prediction models, but also as safety-relevant decision-support systems embedded in biological, agronomic, and operational contexts. We first relate AI-based detection to established plant disease forecasting and decision-support traditions, including weather-based models, epidemiological forecasting, and integrated disease management. We then adapt selected safety-assurance concepts, including risk assessment, failure mode and effects analysis, Bow-tie reasoning, warning-threshold governance, reliability analysis, resilience thinking, and response closure, to host-pathogen-environment warning chains. The proposed framework links AI or sensor outputs with pathogen biology, host susceptibility, environmental conduciveness, inoculum pressure, uncertainty assessment, risk classification, threshold decisions, human or automated verification, intervention, and feedback learning. Illustrative crop-pathogen scenarios, including wheat rust, rice blast, potato late blight, grapevine downy mildew, and citrus greening, show how safety assurance can complement existing forecasting and decision-support systems rather than replace them. The framework remains conceptual, and whether these added assurance functions improve existing warning systems requires comparative evaluation under field conditions. Future systems should be evaluated through detection performance and response-oriented indicators such as lead time, calibration, false-alert burden, missed-warning rate, response completion, disease suppression, economic value, and learning after field action.
Why it matches plant phenotyping methods植物病害・ストレスの検出を含む知的警戒システムについて、AI・リモートセンシング・デジタルフェノタイピング・センサーネットワークの評価枠組みを体系的に論じる方法論レビューであり、方法論が中心です。
abstractPlant disease and plant stress early warning systems have advanced through deep learning, remote sensing, digital phenotyping, disease forecasting, and sensor networks.
Why it matches plant phenotyping methodsブドウ葉ディスク上のべと病症状を画像から自動分割し、病害重症度を推定する手法を開発・検証しており、植物表現型取得が中心である。
abstractWe developed SporaScan, an automated pipeline combining YOLO v8n for leaf disc localization, Mobile SAM for background removal, and UNet for sporulation segmentation.
Common beanGrapevineLeafClassificationObject detectionStress / disease detectionDisease symptoms / severity
Plant leaf disease is a grave risk to crop suitability and agricultural sustainability, and therefore, the ability to make early and accurate diagnosis is a mandatory need in the contemporary precision farming system.In recent years, deep learning has gained significant attention for image-based plant disease detection.Despite its effectiveness, model performance can be influenced by factors such as redundant feature representations and sensitivity to hyperparameter selection.To address these challenges, this study proposes a nested hybrid optimization framework that combines the Cuckoo Search Algorithm (CSA) for channel selection with the Beluga Whale Optimization Mechanism (BWOM) for tuning the hyperparameters of a Convolutional Neural Network (CNN).The proposed approach is independently evaluated on bean and grape leaf datasets under consistent experimental conditions to assess its disease classification performance.In addition to strong predictive performance, the framework incorporates explainable AI (XAI) techniques, namely Gradient-weighted Class Activation Mapping (Grad-CAM) and Gradientweighted Class Activation Mapping Plus Plus (Grad-CAM++), to enhance model interpretability.These approaches highlight the most significant visual features influencing predictions, thereby providing valuable insights for agronomists and fostering trust in AI-based systems.Experimental results show that the proposed CSA-BWOM optimized CNN achieves classification accuracies of 99.61% and 99.38% on the bean and grape datasets, respectively, outperforming baseline CNN models and exhibiting competitive performance when compared to a few existing approaches.
Why it matches plant phenotyping methods植物葉の病徴を画像から分類するCNN最適化・説明可能AI手法が研究の中心であり、植物病害状態の表現型推定に該当する。
titleExplainable AI-based CNN Optimization Model for Plant Leaf Disease Detection
Plant phenotyping plays a critical role in understanding plant health and improving agricultural productivity by enabling quantitative analysis of disease-related physiological characteristics.Among these, leaf diseases significantly impact crop yield and quality, necessitating accurate and automated phenotyping approaches.Traditional phenotyping methods rely on manual inspection or handcrafted feature extraction, which are time-consuming, prone to human error, and lack scalability under diverse environmental conditions.This study introduces a deep learning (DL)-based approach for image-based plant phenotyping, focusing on the classification of disease-affected traits.The developed method integrates segmentation-based region extraction, dual-branch feature learning, and attention-based feature fusion.Initially, input images are preprocessed and passed through a TransUNet-based segmentation module to isolate phenotypically relevant leaf regions while suppressing background interference.Both the original image and the segmented region are then processed using a RegNet-based feature extraction network to capture global structural information and localized disease-specific characteristics.The extracted features are fused using an attention-based mechanism, followed by fully connected layers for multiclass classification.Experimental results obtained on the controlled PlantVillage grape leaf dataset, which serves as a standardized benchmark for plant disease classification, demonstrate an overall classification accuracy of 97.8%, with precision, recall, and F1-score values of 97.7%, 97.9%, and 97.8%, respectively.In addition, the segmentation module achieves an Intersection over Union (IoU) of 94.1% and a Dice score of 96.8%, confirming its effectiveness in isolating relevant phenotypic regions.
Why it matches plant phenotyping methods植物病害形質の画像取得・領域抽出・分類を中核とする深層学習フェノタイピング手法の開発と性能評価であり、方法が中心的です。
abstractThis study introduces a deep learning (DL)-based approach for image-based plant phenotyping, focusing on the classification of disease-affected traits.
GrapevineLaboratory / benchtopStem / branchPhysiological trait estimationWater status / transpiration
Non-invasive, real-time monitoring of plant water status is critical for precision agriculture and plant physiology. However, existing methods often lack continuous in situ measurement capability or are limited by temporal resolution. This paper proposes a novel non-invasive method based on xylem electrical conductivity, inspired by industrial non-contact fluid measurement. As a ground-based complement to remote sensing, this approach demonstrates the feasibility of online, in situ, and non-invasive monitoring of water stress in grapevine stems under controlled laboratory conditions. The industrial C4D sensing system is adaptively modified into a specialized Plant-C4D sensor with an array-based design for batch signal acquisition. To validate the electrical response to water loss, a gravimetric natural dehydration experiment was conducted, demonstrating a clear correlation between electrical signals and water content changes in detached stem samples. Full-day dynamic experiments are conducted under three conditions: normal water supply, varying water stress, and plant inactivation. Sensitive characteristic parameters are extracted through signal analysis, and a pattern recognition framework is established to eliminate environmental interference and suppress individual differences. Experimental results on 24 plant samples (Shine Muscat) show that the method accurately discriminates viable from inactivated plants with an accuracy of 91.67% (22/24 correct). Furthermore, the Fuzzy C-Means (FCM) clustering algorithm successfully quantifies the severity of water stress in viable plants, yielding results consistent with actual water supply conditions. While these findings demonstrate the capability of Plant-C4D sensor to capture stem water status-related information, the current results do not establish full physiological validation, warranting further exploration with in vivo experiments.
Why it matches plant phenotyping methods植物の水分状態を直接推定する非侵襲センサーと解析手法の開発・検証が研究の中心であり、明確な植物フェノタイプ測定に該当する。
abstractThe industrial C4D sensing system is adaptively modified into a specialized Plant-C4D sensor with an array-based design for batch signal acquisition.
Grapevine yellows diseases, including Flavescence Dorée (FD) and Bois Noir (BN), severely affect vineyards, making reliable detection methods essential. Detecting symptoms, particularly in Chardonnay variety, remains challenging due to year-to-year vine variability and the presence of symptoms that can be misleading. We present a dataset of multispectral images of grapevine leaves collected over four years (2021–2024). The dataset includes healthy leaves, grapevine yellows - overwhelmingly BN, and three other diseases (Esca, Discoloration, and Leafroll) with visually similar symptoms particularly for Leafroll, providing a diverse benchmark for disease recognition. Ground-truth labels were assigned based on field inspections conducted by Comité Champagne expert viticulturists who selected the vines before acquisitions and analyzed the images after acquisition. Images were acquired under laboratory conditions using a DJI Phantom 4 Multispectral camera from leaves collected in the Plumecoq Comité Champagne experimental vineyards. Leaves were collected on designated vines and placed on polystyrene boards on a 2x2 grid. One RGB image and five multispectral images corresponding to the Blue (B - 450 nm ± 16 nm), Green (G - 560 nm ± 16 nm), Red (R - 650 nm ± 16 nm), Red-Edge (RE - 730 nm ± 16 nm), and Near-Infrared (NIR - 840 nm ± 26 nm) bands were captured using the same acquisition system (camera-to-board distance of approximately 90 cm) over the four years.. For the vast majority, ambient light was used; for some acquisitions, indirect halogen spotlights were used. The 512x512 leaf images were manually cropped from the acquired original images with the same offsets between the bands; offsets were already present due to the parallax phenomenon. The dataset is organized into six folders (B, G, R, RE, NIR, and RGB). Each leaf is represented by one RGB image and five corresponding multispectral band images, enabling multimodal analysis and machine learning for grapevine disease detection. Each image filename encodes acquisition and annotation information as follows: Characters 1–2: acquisition year (21 = 2021, 22 = 2022, 23 = 2023, 24 = 2024). Character 3: image type/band (0 = RGB, 1 = B, 2 = G, 3 = R, 4 = RE, 5 = NIR). Character 4: class (J = Grapevine Yellows, T = Healthy, S = Esca, E = Leafroll, D = Discoloration). Character 5: illumination condition (0 = ambient light, G = additional left halogen spotlight, D = additional right halogen spotlight, 2 = additional both left and right halogen spotlightsl). Characters 6–9: image identifier, numbered sequentially from 0000. Character 10: relabelling status. After image acquisition, all leaves were independently reviewed by expert viticulturists on the recorded RGB images. A value of 1 indicates that the expert confirmed the original field label, 0 indicates that the original label was considered incorrect based on the information visible in the RGB image, and D indicates that the sample was reclassified as Discoloration. This dataset complements the "Multi-annual spectral data of Chardonnay grapevine leaves" dataset published on Recherche Data Gouv. While the previous dataset contains spectral measurements, the present dataset provides multispectral images acquired from the same grapevine leaves, enabling multimodal analyses that combine spectral signatures with image-based information. Note, however, that there is no bijection between the respective files.
Why it matches plant phenotyping methodsブドウ葉の病徴をマルチスペクトル画像で取得し、専門家ラベル付きの疾病認識用データセット／ベンチマークとして提供しており、植物状態の画像ベース表現型計測が中心である。
abstractWe present a dataset of multispectral images of grapevine leaves collected over four years (2021–2024).
Accurate and reliable diagnosis of grape leaf diseases is essential for sustainable viticulture, enabling timely intervention, reducing economic losses, and supporting precision crop management. Although deep learning (DL) models have demonstrated remarkable classification performance, their reliability under real-world field conditions remains insufficiently explored. In particular, confidence estimates often fail to reflect true predictive correctness when models are exposed to distributional shifts, limiting their practical applicability. To address this challenge, this study proposes a confidence- and uncertainty-aware DL framework for grape leaf disease diagnosis that extends evaluation beyond conventional accuracy-based metrics. Two publicly available grape leaf datasets were employed for stratified five-fold cross-validation in binary and multiclass classification tasks, while a third independently collected dataset was reserved exclusively for leakage-free external validation. EfficientNet-B0 and MobileNetV3-Large were evaluated under four inference configurations: raw inference, temperature scaling (TempScaling), Monte Carlo dropout, and an ensemble strategy combining uncertainty estimation with calibration. External validation demonstrated that both architectures maintained discriminative capability under domain shift. Under ensemble inference, EfficientNet-B0 achieved an accuracy of 73.8%, a macro-F1 score of 73.3%, a Matthews correlation coefficient (MCC) of 0.468, and a receiver operating characteristic area under the curve (ROC-AUC) of 0.804, while MobileNetV3-Large achieved 71.8% accuracy, 71.2% macro-F1, an MCC of 0.429, and a ROC-AUC of 0.787. Despite these promising results, raw predictions exhibited substantial overconfidence, with Expected Calibration Error (ECE) values of 0.287 and 0.312 for EfficientNet-B0 and MobileNetV3-Large, respectively. TempScaling markedly improved calibration quality, reducing ECE to 0.038 and 0.042 without affecting classification performance. Ensemble inference further enhanced the balance between predictive discrimination and reliability. The results demonstrate that strong classification performance alone is insufficient for trustworthy deployment in agricultural environments. Confidence calibration and uncertainty quantification provide complementary information for identifying overconfident predictions and improving decision reliability under field variability. The proposed framework offers a reliability-oriented approach for developing trustworthy artificial intelligence systems for grape leaf disease diagnosis in precision agriculture.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像から診断する深層学習手法を開発し、信頼度校正・不確実性推定と外部検証を中心に評価しているため、植物フェノタイピング手法として適格。
abstractthis study proposes a confidence- and uncertainty-aware DL framework for grape leaf disease diagnosis
Grapevines are economically vital crops but are highly susceptible to fungal, bacterial, and viral diseases that threaten yield and quality. Traditional detection methods rely on manual inspection, are time-consuming, prone to human error, and often delay intervention. This study introduces a lightweight convolutional neural network (CNN) architecture specifically designed for accurately and efficiently detecting grapevine leaf diseases—including Black Rot, ESCA, and Leaf Blight—based on image classification. The proposed model integrates optimized residual blocks, batch normalization, dropout layers, and global average pooling to maximize accuracy while minimizing computational complexity. This lightweight design makes it well-suited for deployment on edge devices such as drones and mobile systems used in precision agriculture. A comprehensive data augmentation strategy was applied during training to simulate real-world variability and enhance generalization. The model was trained using 9,027 labeled grape leaf images from a publicly available grape disease image dataset, and it achieved 99.8% overall accuracy with near perfict precision, recall, F1-score, and area under the ROC curve (AUC) across all classes. These findings highlight the practical potential or real-time, scalable and sustainable disease monitoring in smart vineyard management systems.
Why it matches plant phenotyping methodsブドウ葉の病害状態を画像から分類するCNN手法を開発・評価しており、植物病害表現型の取得・推定が研究の中心です。
abstractThis study introduces a lightweight convolutional neural network (CNN) architecture specifically designed for accurately and efficiently detecting grapevine leaf diseases—including Black Rot, ESCA, and Leaf Blight—based on image classification.
Crop disease identification is still a big problem in agriculture, which results in large yield losses and food lacks, especially in regions dependent on manual monitoring. Traditional methods of identifying plant diseases are often labor intensive, error-prone, and ineffective in early-stage diagnosis. To overcome these limitations, this study proposes a hybrid machine learning model for accurate crop type identification, disease classification, and severity prediction using image data. The methodology utilizes the PlantVillage dataset, encompassing over 50,000 annotated leaf images across 14 crops. After rigorous preprocessing involving image resizing, normalization, and cleaning, Improved Deep Joint Segmentation is applied to localize disease-affected regions. Feature extraction incorporates color, texture (GLCM, LBP), and shape attributes to enhance classification accuracy. A hybrid approach integrating XGBoost for feature selection and Support Vector Machine (SVM) for classification is proposed to capture both overarching trends and intricate details. Experimental results across four major crops—potato, tomato, corn, and grape—demonstrate superior performance, with the hybrid model achieving 98.6% accuracy, 98.3% precision, 99.0% recall, and 99.1% F1-score. The model outperforms existing approaches, offering a robust, scalable, and accurate solution for early crop disease detection in precision agriculture.
Why it matches plant phenotyping methods画像から病変領域を抽出し、植物病害の分類と重症度を推定する機械学習ワークフローが研究の中心であり、植物状態の表現型計測に該当する。
abstractthis study proposes a hybrid machine learning model for accurate crop type identification, disease classification, and severity prediction using image data.
Grapevine downy mildew (GDM), caused by Plasmopara viticola, is managed largely through repeated fungicide applications, yet evaluating spray-program performance is difficult because field infections are spatially heterogeneous. However, host physiological changes can precede visible symptom development. We tested whether standardized, high-throughput VNIR hyperspectral imaging of field-grown grapevine leaf discs can capture early optical changes associated with infection and discriminate program-linked mitigation. Leaves were collected from a 2025 vineyard trial in Geneva, NY (cv. La Crescent) managed under three spray programs (conventional fungicides, biofungicides, and untreated control). Leaf discs (≈87 per program) were excised, inoculated with P. viticola, and imaged at 0.5, 1.5, and 3.5 days post inoculation (dpi) using a custom built, automated, hyperspectral imaging microscopy platform (400-980 nm). No visible sporulation occurred at 0.5 or 1.5 dpi; sporulation was first observed at 3.5 dpi and occurred only in the untreated control, whereas no discs from either treated program sporulated. Multivariate spectral analyses showed significant separation by spray program and dpi, and disease-related spectral indices exhibited program-dependent trajectories, with treated discs showing attenuated optical change relative to the control. A random-forest classifier trained on pre-sporulation spectra (0.5 and 1.5 dpi; N = 169) predicted subsequent sporulation with 0.78 accuracy and 0.77 ROC-AUC (95% CI 0.68-0.85), highlighting informative bands across visible, red-edge, and near-infrared regions. Together, these results support outcome-linked, replicate-rich hyperspectral phenotyping of vineyard spray programs using field-derived material under standardized acquisition conditions.
Why it matches plant phenotyping methods自動化VNIRハイパースペクトル撮像で、発病前のブドウ葉の光学的変化と将来の胞子形成を推定する手法を中心に検証しており、植物病害状態の表現型取得・分類に該当する。
abstractWe tested whether standardized, high-throughput VNIR hyperspectral imaging of field-grown grapevine leaf discs can capture early optical changes associated with infection and discriminate program-linked mitigation.
Abstract Climate change and increasing drought frequency are intensifying water stress risks in viticultural systems, particularly in semi-arid regions where water availability directly influences grapevine productivity and quality. This study investigated pre-harvest and post-harvest water stress dynamics in a commercial Ağın Beyazı vineyard located in Eastern Türkiye using UAV-based multispectral imagery. High-resolution orthomosaics (2.31 cm pixel⁻¹) were acquired with a DJI Mavic 3 Multispectral platform and used to generate Green Normalized Difference Vegetation Index (GNDVI) and Normalized Difference Red Edge Index (NDRE) maps for both phenological periods. Water stress responses were evaluated through pixel-based change detection, descriptive statistics, spatial heterogeneity metrics (Local Mean and Local Standard Deviation), and Global Moran’s I spatial autocorrelation analysis. Results revealed measurable spectral differences between pre-harvest and post-harvest periods, indicating changes in canopy physiological activity associated with water stress and post-harvest vine responses. GNDVI exhibited a higher relative change (+ 10.49%) and stronger standardized response ratio, suggesting greater sensitivity to overall canopy vigor and photosynthetic activity. In contrast, NDRE showed the highest heterogeneity response (+ 64.97%), demonstrating superior capability for identifying localized stress variability and fine-scale physiological differences within the vineyard. Spatial heterogeneity analyses indicated increasing local variability after harvest, while consistently positive Moran’s I values (≈ 0.257) revealed that stress patterns remained spatially organized rather than randomly distributed. These findings suggest that vineyard water stress is controlled not only by vine physiology but also by persistent spatial factors such as soil conditions, micro-topography, and water availability. Overall, the combined use of GNDVI and NDRE provided complementary information for vineyard water stress assessment. The proposed framework demonstrates the potential of UAV-based multispectral monitoring for precision irrigation management, early stress detection, and climate-resilient viticulture in water-limited environments.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像からGNDVI・NDREを抽出し、ブドウ樹の水ストレスと空間変動を評価する測定・解析フレームワークが研究の中心である。
abstractThis study investigated pre-harvest and post-harvest water stress dynamics in a commercial Ağın Beyazı vineyard located in Eastern Türkiye using UAV-based multispectral imagery.
In Japan, the quantity of domestically produced fruit has been gradually decreasing, while wholesale prices have continued to rise due to declining production volumes and a shift toward high-quality varieties. To address these trends, improving quality and reducing labor through automation have become urgent challenges. In precision viticulture, monitoring the growth of grape clusters plays a key role in yield estimation, disease management, and optimal harvest timing. Although recent advances in deep learning and 3D reconstruction have enabled accurate fruit detection and modeling in vineyards, tracking the same clusters on different days remains challenging because of branch movement, fruit growth, and varying imaging conditions. This study proposes a branch-based 3D alignment framework for the cross-day tracking of grape clusters. Stable vine structures, such as trunks and main branches, are reconstructed using Structure from Motion, and their spatial correspondences are estimated through SIFT-based matching and similarity transformation. Once the coordinate systems of different days are aligned, the grape clusters detected by CenterNet are associated based on spatial proximity in the unified 3D space. Experiments over multiple observation days demonstrated that the proposed method successfully maintained the consistent tracking of grape clusters throughout the growth period. These results indicate that branch-based alignment effectively stabilizes multi-day observations and facilitates the temporal monitoring of fruit growth, supporting automated phenotyping and future field robot applications in viticulture.
Why it matches plant phenotyping methodsブドウ房の経日追跡を目的とする3D画像アライメント手法を開発し、果実成長の時系列モニタリングと自動フェノタイピングへの利用を実験的に検証しているため、フェノタイピング手法が中心である。
abstractThis study proposes a branch-based 3D alignment framework for the cross-day tracking of grape clusters.
Continuous monitoring of vineyard dynamics is essential for optimizing viticultural practices and assessing plant health. While the seasonal behaviors of satellite-derived vegetation indices are widely studied, robust parametric modeling of these temporal trends remains underexplored. Building upon initial clues derived from Italian vineyards, this study proposes a novel analytical framework based on the consistent parabolic temporal signature of optical and Synthetic Aperture Radar (SAR) indices. Focusing on the elevated Trinity Canyon Vineyards in Armenia, we model the yearly evolution and temporal aggregations of these indices using a parabolic fitting approach. Our results suggest that the parabola vertex, which we hypothesize corresponds to the absolute maximum of vegetative activity, remains remarkably stable across diverse vine types, satellite orbits, and years. While this stable behavior suggests an underlying phenological or structural consistency, distinct exceptions to this trend have also been identified and considered. Furthermore, to bridge the gap between remote sensing observables and agronomic traits, we investigated the relationship between the fitted parabolic parameters and the Winkler index, which is used here as an estimator of above-ground biomass (AGB). By correlating the vegetation indices’ temporal dynamics with biomass growth and by isolating specific anomalies driven by environmental or anthropogenic factors, this work offers a basis for a predictive methodology that enables tracking vineyard structural development.
Why it matches plant phenotyping methods衛星画像の植生指数を放物線フィッティングし、ブドウ園の植生活動・地上部バイオマス・構造発達を推定する分析手法が研究の中心であるため。
abstractthis study proposes a novel analytical framework based on the consistent parabolic temporal signature of optical and Synthetic Aperture Radar (SAR) indices
Abstract Accurate crop disease detection supports precision agriculture, but field-deployable identification remains hindered by complex backgrounds, varying illumination, and heavy deep learning models. This work presents a lightweight visual detection approach for grape leaf diseases under unconstrained field conditions. Built on the YOLO11n backbone, the method integrates three customized modules: C3k2-UltraLightBlock for efficient feature representation, LeafRepFusionStem for low-level feature enhancement, and RCSA-HSFPN for refined multi-scale fusion with residual channel-spatial attention. A dedicated dataset with complex backgrounds is constructed via augmentation and background replacement. Experiments show the model achieves 92.0% precision, 92.9% recall, and 93.0% mAP@0.5, with only 2.9 GFLOPs and 1.73 M parameters, representing 54.7% and 33.2% reductions over the baseline. Heatmap visualization confirms improved lesion focusing and background suppression, while cross-crop tests validate strong generalization. This framework provides an efficient solution for real-time, edge-deployable plant disease monitoring, balancing accuracy and computational efficiency for practical agricultural visual computing applications.The implementation code for this study is available at:https://github.com/aitizc/Lightweight-Visual-Detection-Framework-for-Complex-Background-Grape-Leaf-Disease-Identification.git
Why it matches plant phenotyping methodsブドウ葉の病斑・病害状態を画像から推定する軽量な視覚検出手法を開発・評価しており、植物病害表現型の取得が中心である。
abstractThis work presents a lightweight visual detection approach for grape leaf diseases under unconstrained field conditions.
Reproduction assets foundThe authors explicitly state that the implementation code for this study is publicly available on their GitHub repository. The paper's grape leaf disease dataset itself is not stated as deposited (only the public PlantVillage source is cited), so only the authors' code qualifies as a paper-specific public asset.Code · publicThe implementation code for this study is avail-
able at:https://github.com/aitizc/Lightweight-Visual-Detection-Framework-for-
Complex-Background-Grape-Leaf-Disease-Identification.gitOpen asset ↗https://github.com/aitizc/Lightweight-Visual-Detection-Framework-for-pdf-page:2 lines:1-43Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
Plant tissues consist of diverse cell populations that collectively contribute to development, metabolism, environmental responses, and phenotype formation. Although single-cell and single-nucleus RNA sequencing have greatly advanced the study of plant cellular heterogeneity, their application to large sample cohorts remains limited by cost, technical complexity, tissue dissociation constraints, and throughput. In contrast, bulk RNA-seq datasets have accumulated extensively across plant species, tissues, developmental stages, and environmental conditions, yet the celltype-level information embedded in these datasets remains difficult to resolve because plant-oriented deconvolution frameworks are still lacking. Existing deconvolution methods have largely been developed in mammalian systems and have not been systematically optimized for plant transcriptomic features, leaving their applicability under plant-specific constraints unclear. Here, we present SEED, an adaptive deconvolution framework optimized for plant transcriptomic data. SEED integrates candidate reference-template construction with seven deconvolution strategies and automatically identifies an optimal combination for a given dataset. In grapevine simulated benchmarking, SEED showed its clearest advantage under low-replication conditions and remained broadly competitive, rather than uniformly dominant, when larger pseudo-bulk sample sizes were evaluated. SEED further performed robustly in public Arabidopsis thaliana and Nicotiana tabacum datasets. Finally, we applied SEED to bulk RNA-seq data generated in this study from Vitis vinifera cv . Cabernet Sauvignon berries collected from Yinchuan and Yantai, identifying terroir-associated cell subtypes and coordinated celltype interaction patterns. Together, these results establish SEED as a practical framework for plant transcriptome deconvolution and provide a new tool for dissecting cellular heterogeneity associated with environmental adaptation and phenotype formation in plants.
Why it matches plant phenotyping methods植物トランスクリプトームから細胞サブタイプと相互作用パターンを推定するSEEDを開発し、複数データセットでベンチマーク・検証している。分子データ解析だが、植物の細胞状態・表現型形成に結び付く再利用可能な推定手法が研究の中心である。
abstractHere, we present SEED, an adaptive deconvolution framework optimized for plant transcriptomic data.
Cold hardiness is a critical trait for grapevine survival and productivity in cold climates. This study examined the relationships among cane morphological characteristics, shoot color parameters, and cold hardiness in two grapevine cultivars ('Prairie Star' and 'Frontenac') across four dormant-season sampling times (ST 1-ST 4) and three internode diameter classes (small, normal, and large). Morphological traits, including internode length, shoot diameter, and cross-sectional area, did not show a consistent temporal trend across sampling periods, suggesting that the observed variation was primarily associated with sampling time and cane class rather than progressive structural change during dormancy. In contrast, colorimetric traits showed a clear seasonal pattern, with shoots becoming darker and redder from ST 1 to ST 4, consistent with advancing lignification and cane maturation. Cold hardiness, assessed using low-temperature exotherms of bud, phloem, and xylem tissues, increased substantially from early to mid-dormancy, with xylem tissues reaching the greatest freezing tolerance by ST 3-ST 4. 'Prairie Star' showed slightly greater xylem cold hardiness than 'Frontenac', while bud survival remained consistently high across all treatments. Strong associations between shoot color and LTE values indicate that color traits, particularly at the fifth internode, may serve as reliable non-destructive indicators of cold hardiness status. Sampling time was the primary source of multivariate variation, with cultivar and internode class contributing secondary effects. These findings demonstrate that observable cane traits, especially shoot color, reflect the progression of seasonal cold acclimation and may support the evaluation and selection of cold-hardy grapevine germplasm.
Why it matches plant phenotyping methods枝の色・形態を用いてブドウの耐寒性を非破壊推定する指標として評価しており、単なる生物学的測定ではなく表現型取得法の妥当性評価が中心です。
abstractStrong associations between shoot color and LTE values indicate that color traits, particularly at the fifth internode, may serve as reliable non-destructive indicators of cold hardiness status.
Abstract Background This study evaluates the performance of two UAV (Unmanned Aerial Vehicle) platforms for vineyard monitoring, focusing on geometric accuracy, effective resolution, and vegetation index consistency. Methods Two UAV (Unmanned Aerial Vehicle) platforms were compared over a vineyard using multiple flights with a consumer-grade drone and a single RTK (Real-Time Kinematic) enabled flight at similar altitude over a single vineyard block under comparable acquisition conditions, followed by photogrammetric reconstruction, DSM (Digital Surface Model) co-registration, effective resolution analysis, and RGB (Red-Green-Blue) based vegetation index comparison. Results This study demonstrates that the integration of RTK (Real-Time Kinematic) technology in the Mavic 3E improves the absolute accuracy of DSMs (Digital Surface Models), eliminating systematic vertical offsets of ~ 35 m observed in the Mavic 2E products. After applying a robust Z-shift (vertical) correction, DSMs (Digital Surface Models) from both platforms became directly comparable, with RMSE (Root Mean Square Error) values reduced to ~ 1.3 m while NMAD (Normalized Median Absolute Deviation) remained stable. RTK (Real-Time Kinematic) positioning in the Mavic 3E ensured reliable absolute georeferencing, whereas DSMs (Digital Surface Models) derived from the Mavic 2 Enterprise Zoom remained internally consistent after post-processing, though with lower absolute positional accuracy. Effective resolution analysis further showed that the Mavic 3E imagery preserves higher spatial detail than the Mavic 2E, underscoring the importance of sensor optics and stability for vineyard monitoring. Comparisons of vegetation indices revealed that normalized indices such as NGRDI (Normalized Green Red Difference Index) and VARI (Visible Atmospherically Resistant Index) provide consistent results across platforms, while ExG (Excess Green Index) exhibited strong biases and wide limits of agreement, reflecting its sensitivity to radiometric differences. Conclusions For cross-platform or multi-temporal monitoring, NGRDI (Normalized Green Red Difference Index) and VARI (Visible Atmospherically Resistant Index) proved to be more robust, whereas ExG (Excess Green Index) should only be applied after radiometric harmonization. This study provides a replicable workflow for UAV (Unmanned Aerial Vehicle) based vineyard monitoring that integrates geometric alignment, DSM (Digital Surface Model) correction, effective resolution assessment, and index comparison, offering practical recommendations for researchers and practitioners aiming to ensure reliable and comparable UAV (Unmanned Aerial Vehicle) derived vineyard metrics.
Why it matches plant phenotyping methodsUAV画像・写真測量・DSM補正・植生指数を比較検証し、ブドウ園の植物状態を再現可能に測定するワークフローを中心に扱っているため、植物フェノタイピング手法研究に該当する。
abstractThis study evaluates the performance of two UAV (Unmanned Aerial Vehicle) platforms for vineyard monitoring, focusing on geometric accuracy, effective resolution, and vegetation index consistency.
We introduce the Grapevine (Vitis vinifera) Leaf Image Dataset (GVLiD), a carefully curated set of 3,477 annotated images of Grapevine (Vitis vinifera) leaves to catalyze research in computer vision and plant pathology. Whereas the PlantVillage and Hermos datasets, for instance, contain mainly scanned or laboratory-acquired leaves, GVLiD features vineyard in situ images along with detailed metadata (GPS, lighting, weather, and device model) and expert-verified annotations. To measure the reliability of the annotation, label consistency was very high (κ = 0.86-0.92; 95% CI) as assessed by inter- and intra-rater agreement. Besides Indian viticulture, the dataset also aims to support the field of foliar disease detection in precision agriculture and ML benchmarking, which face significant challenges due to variable illumination and natural leaf backgrounds under field conditions. GVLiD is intended to enable worldwide, reproducible, real-world testing of AI systems for crop disease monitoring.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像で注釈化したデータセットであり、植物病害状態の画像ベース表現型測定と再現可能なベンチマークが中心です。
abstractWe introduce the Grapevine (Vitis vinifera) Leaf Image Dataset (GVLiD), a carefully curated set of 3,477 annotated images of Grapevine (Vitis vinifera) leaves to catalyze research in computer vision and plant pathology.
Reproduction assets foundThe paper's grapevine leaf image dataset (GVLiD) is deposited on Mendeley Data, but that URL is not among the allowed URLs. The authors' validation/analysis code (image-quality metrics, metadata-completeness checks, annotation-reliability calculations) is publicly available on GitHub at the allowed URL, with explicit 'Code · publicAll validation scripts (image-quality metrics, metadata-completeness checks, and annotation-reliability calculations) are publicly available in the GVLiD GitHub repository. This ensures full reproducibility of all validation results reported here.
git clone https://github.com/MilindGayakwad/DNNOpen asset ↗https://github.com/MilindGayakwad/DNNhtml-lines:255-349Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Crop disease detection is crucial for maintaining high agricultural productivity and minimizing the financial losses of farmers.•This study compares the performance of a hybrid method based on integrating the Long Short-Term Memory algorithm (LSTM) and Convolutional Neural Network (CNN) with a standalone CNN method for grape crop disease detection and classification.•The proposed methods are trained and tested on a comprehensive balanced and imbalanced grape leaf dataset with 4000 and 4062 images, respectively. The imbalanced dataset has 423 images of the healthy class, 1383 of ESCA, 1076 of leaf blight, and 1180 images of the black rot class, while the balanced dataset has 1000 images in each class•The LSTM-CNN method achieved 100%, and the CNN method achieved 99.47% accuracy on the balanced dataset. Also, in the case of the imbalanced dataset, 100% and 97.89%, respectively.Among these two methods implemented, the proposed LSTM-CNN method has achieved excellent results in terms of performance metrics. This study presents a deep learning-based system for automated leaf disease classification. The model is designed for end-user deployment, where users can upload a leaf image and receive an immediate prediction of the disease without requiring model retraining or technical expertise.
Why it matches plant phenotyping methodsブドウ葉画像から病害状態を推定する深層学習手法を比較・評価しており、植物病害表現型の取得・分類が研究の中心である。
abstractThis study compares the performance of a hybrid method based on integrating the Long Short-Term Memory algorithm (LSTM) and Convolutional Neural Network (CNN) with a standalone CNN method for grape crop disease detection and classification.
Abstract Context Downy mildew, caused by Plasmopara viticola , remains one of the most damaging diseases affecting grapevines, especially in humid viticultural regions such as the “Vinhos Verdes” in northern Portugal. Traditional detection relies on visual inspection and laboratory techniques, which are subjective and reactive, often delaying effective intervention. Aims This study aims to evaluate the potential of field spectroscopy combined with machine learning to detect downy mildew in Vitis vinifera cv. Loureiro field conditions. The focus is on providing an early, non-destructive detection method that can be used in precision viticulture, reducing the need for costly, widespread pesticide applications. Methods and Key Results Leaf spectral reflectance data were collected along the 2023 and 2024 growing seasons using a portable spectroradiometer. Measurements were obtained from both untreated and fungicide-treated grapevines, covering different infection stages. Spectral signatures from 600 grapevine leaves were used to train and validate classification models using Partial Least Squares Linear Discriminant Analysis (PLS-LDA) and Random Forest (RF) models. Both RF and PLS-LDA models showed an overall accuracy of 95.1% when trained with all spectral features from the dataset. Red edge (700–750 nm) and visible (400–700 nm) wavelengths demonstrated the highest classification contribution. Moreover, the twenty most informative wavelengths for infection discrimination were identified for each model. Conclusion The results confirm the effectiveness of field spectroscopy, making it possible to detect downy mildew symptoms in different stages of infection. This method offers a rapid, cost-effective, and sustainable tool for early disease detection, which can greatly benefit winegrowers by enabling more timely and specific interventions and reducing the reliance on chemical treatments. Implications and Impacts This study demonstrates a novel approach to precision viticulture, offering winegrowers a rapid, cost-effective, and sustainable tool for early disease detection. The methodology not only promotes more disease management strategies but also aligns with environmental and regulatory goals.
Why it matches plant phenotyping methodsブドウ葉の病害症状をフィールド分光と機械学習で直接検出する方法を開発・評価しており、植物病害状態の取得手法が中心である。
abstractThis study aims to evaluate the potential of field spectroscopy combined with machine learning to detect downy mildew in Vitis vinifera cv. Loureiro field conditions.
Abstract Artificial Intelligence and Machine Learning rely on large, high-quality datasets for accurate and robust models, yet data scarcity remains a major challenge, especially in smart farming. Phenology modeling, a key application, studies how plant biological events relate to climate and seasons. Accurate phenology models improve crop quality, support climate adaptation, and guide decisions such as pesticide use and harvesting, enhancing environmental and economic sustainability. However, agricultural data are highly diverse and heterogeneous, complicating model development. This study presents a proposed georeferenced dataset for Machine Learning-based grapevine phenology prediction across 3 Protected Designations of Origin in Aragón, Spain. Developed by a multidisciplinary team, the dataset combines 9 datasets from 8 sources (including meteorological time series, field phenology observations, and Copernicus Sentinel-2 multispectral imagery) covering the period 2016–2022. It supports both physical and Machine Learning-based phenology modeling and facilitates knowledge extraction in agronomy and plant biology. Its relevance lies in its comprehensive scope, the inclusion of 9 phenological stages, and a rigorous methodology ensuring reproducibility. This framework enables the creation of similar datasets for other regions or crops, advancing smart farming through scalable, data-driven solutions. The open publication of the code further supports this objective. We further anticipate its potential contribution to developing foundation models as well as to the creation of new knowledge in biology and agronomy.
Why it matches plant phenotyping methodsブドウの9段階のフェノロジーを対象に、圃場観測と衛星画像などを統合した再利用可能な地理参照データセットを構築しており、植物形質取得・モデル化の方法論が中心である。
abstractThis study presents a proposed georeferenced dataset for Machine Learning-based grapevine phenology prediction across 3 Protected Designations of Origin in Aragón, Spain.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 15 Sept 2026
Hyperspectral sensing has emerged in recent years as a powerful tool for discovery in plant pathology. However, bottlenecks in reflectance data collection, including throughput, data handling, and volume, hinder the speed at which this technology can transition from niche to widespread use, and its downstream applications from proof-of-concept to proof-of-practice. We developed an automated high-throughput hyperspectral imaging (AHHI) platform to address this challenge. Our system includes a push broom hyperspectral camera (MSV500, Middleton Spectral Vision; 400-1000 nm, 8nm spatial and 0.65 nm spectral resolution) and a sample positioning automatic system inherited from “Blackbird,” a microscopic imaging robotic platform. The system acquires line images at 100 frames per sec, equivalent to about 40 sec per 10-mm leaf disc sample (4 hours per 351 samples). The output of the system is 2.5 GB *.raw and *.hdr files, which are processed by an in-house Python script. Our system can collect 9 TB of high-quality hyperspectral images in a single day without external variation in imagery. We demonstrate three use cases for scaling established hyperspectral reflectance applications to the imaging scale via the AHHI: pre-symptomatic disease detection, fungicide detection, and grapevine breeding family lineage discrimination. PERMANOVA and Random Forest analysis all yielded significant p-values and AUC ranging from 77.8 to 99.9%, mirroring prior accuracies from handheld spectrometers when accounting for a loss of spectral resolution in going from VSWIR to VNIR sensing. Access to large amounts of high-resolution hyperspectral data with systems such as AHHI can ease and accelerate translation of plant pathology discoveries from niche to mainstream use.
Why it matches plant phenotyping methods植物病理学における症状・病態の高スループット取得を目的とした自動ハイパースペクトル画像プラットフォームを開発し、複数用途で性能を実証しており、表現型取得法が中心である。
abstractWe developed an automated high-throughput hyperspectral imaging (AHHI) platform to address this challenge.
Early disease diagnosis plays a key role in grape production for minimizing crop risk and maximizing yield. Downy Mildew, Powdery Mildew, and Bacterial Leaf Spot are some of the major diseases that threaten productivity and require timely and accurate diagnosis. This research introduces a new multi-model framework that integrates AI-based image segmentation triggered by Environmental Susceptibility Conditions to inform precision grape farming. The proposed method combines a soft-voting ensemble of the DeepLabV3+, U-Net, and FCN-8’s models for segmentation of diseased and healthy leaf areas with high accuracy, by understanding environment data to evaluate the risk of disease propagation. Major contributions of the study are the understanding of environmental conditions for context-aware disease propagation, an efficient ensemble segmentation method for accurate leaf disease segmentation and severity analysis, performed on a self-collected dataset from a grape farm in Nashik, Maharashtra, India. The system enables early warning and decision support mechanisms to promote sustainable disease management in grape cultivation, with potential implications for reducing unnecessary pesticide usage. Experimental results show the efficacy of the proposed method, with segmentation accuracy of 96.81% and precision of 99.09%, with a Dice score of 0.95 and a mean Intersection over Union (mIoU) of 0.91, demonstrating excellent robustness under noise conditions. Unlike existing studies either image or sensor-approaches, this work introduces the integration of image data and knowledge of environmental insights offers a scalable, reliable, and real-time disease monitoring solution aligned with the goals of smart and sustainable farming.
Why it matches plant phenotyping methodsブドウ葉の病斑領域を画像分割し、病害の重症度を推定する手法を開発・評価しており、植物の病害状態の取得が研究の中心です。
abstractThe proposed method combines a soft-voting ensemble of the DeepLabV3+, U-Net, and FCN-8’s models for segmentation of diseased and healthy leaf areas with high accuracy
Reproduction assets foundThe paper's grape leaf disease image dataset (NGLDD/NGLD) used for segmentation phenotyping is publicly deposited on Mendeley Data by the authors. No code or model checkpoints are reported as publicly available.Dataset · publicThe dataset used in this study is publicly available in the Mendeley Data repository as the Niphad Grape Leaf Disease
Dataset (NGLD) (DOI: https://doi.org/10.17632/8nnd2ypcv3.5).Open asset ↗Mendeley Data · 10.17632/8nnd2ypcv3.5pdf-page:25 lines:1-65Plant phenotyping relevance match · UnverifiedOpenAlex · checked 15 Sept 2026
GrapevineNeRF / 3D Gaussian SplattingLiDAR / point cloudRGB / grayscaleFruitStem / branchPose / keypoint estimation2D/3D reconstructionSegmentation
• End-to-end pipeline from neural reconstruction to physical grape berry manipulation. • Efficient point clouds generation using NeRF and the metric scale derived directly from robot kinematics. • RANSAC sphere fitting achieves 92.1% berry detection precision without annotated training data. • Stem-aligned 6-DoF pose optimization improves end-to-end grip success by 17.2%. Table grape thinning requires selective removal of 20–40% of berries from dense clusters. In practice, workers decide which berries to remove by considering both the approximate berry count and local 3D spatial characteristics such as crowding and relative positioning. Automating this task is challenging because conventional 2D image-based approaches suffer from occlusion-related counting errors and lack explicit 3D spatial information necessary for reliable manipulation. We propose a robot-integrated vision pipeline that reconstructs grape bunch structure from posed multi-view RGB images. Neural Radiance Fields (NeRF) is used to learn a volumetric scene representation, from which a dense, low-noise point cloud is extracted via depth back-projection, and RANSAC-based geometric fitting models individual berries and stems, enabling berry-level segmentation and orientation estimation for manipulation planning. The perception pipeline uses an eye-in-hand RealSense D405 camera mounted on a Fanuc CRX-5iA collaborative robot. Camera poses are derived from the robot kinematic chain, allowing the reconstructed point cloud and detected berry centers to be expressed directly in the metric robot base frame without external scale recovery. In robot-mounted RealSense D405 experiments on 10 grape bunches, RANSAC sphere fitting achieved a counting MAE of 0.50 berries, RMSE of 0.71 berries, and mean center localization error of 2.71 mm. On a 52-bunch benchmark, RANSAC sphere fitting outperforms the learning-based SoftGroup++ method for 3D berry instance segmentation (92.1% vs 82.8% average precision) without requiring annotated training data. In 35 manipulation trials, the system achieved an 85.7% pre-grasp reachability rate and an 83.3% conditional target success rate demonstrating an end-to-end pipeline from neural reconstruction to manipulation-ready berry poses.
Why it matches plant phenotyping methodsブドウ房の3D再構成、ベリー分割・計数・位置推定を中核とするロボット統合型フェノタイピング手法であり、技術性能も定量評価している。
abstractWe propose a robot-integrated vision pipeline that reconstructs grape bunch structure from posed multi-view RGB images.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 5 Sept 2026
GrapevineRaman / spectroscopyLeafPhysiological trait estimationPhotosynthesis / fluorescenceWater status / transpiration
Estimating crop trait data is critical for predicting crop responses to environmental change, enabling more informed diagnoses of crop performance and the development of on-farm management strategies. Yet, many traditional methods for quantifying plant traits are time-consuming and resource-intensive, limiting sample sizes and study durations. In response, high-throughput phenotyping-specifically reflectance spectroscopy-has emerged as a key element of plant trait research, enabling rapid estimation of plant traits. However, little is known about whether reflectance spectroscopy can detect within-species variation in resource acquisition and plant-water traits, especially variation that exists among different cultivars or genotypes of the same crop. Using wine grapes (V. vinifera subsp. vinifera) as a focal crop, this study aimed to assess the ability of reflectance spectroscopy to quantify intraspecific variation in 12 leaf traits across 12 different cultivars from seven different varieties. We find significant variability in traits across and within cultivars, especially in gas-exchange and hydraulic traits, with cultivars varying along a resource-conservative-to-resource-acquisitive trait axis. Models based on spectral reflectance data were able to differentiate and predict this fine-scale trait variation among cultivars for seven plant traits, with a predictive power range of R2 = 0.12-0.57. Models predicting leaf chemical (i.e., carbon and nitrogen concentrations), physiological (i.e., maximum rate of light-saturated photosynthesis), and morphological traits (i.e., leaf dry matter content) were more accurate in their predictions, while models predicting leaf water status were less accurate. Our results indicate that reflectance spectroscopy can capture certain dimensions of the fine-scale trait variation that exists within genetically diverse agroecosystems, though spectroscopic estimates of intraspecific variation in leaf water status are less accurate.
Why it matches plant phenotyping methods反射分光法を用いてブドウ葉の複数形質を推定し、品種内変異に対する予測性能を評価しており、植物表現型取得・推定手法が研究の中心である。
abstracthigh-throughput phenotyping-specifically reflectance spectroscopy-has emerged as a key element of plant trait research, enabling rapid estimation of plant traits.
GrapevineField / plotStereoWhole plant / canopy / plot / field
• A stereovision-controlled variable-rate sprayer was tested over three growth stages • Spray volume, canopy deposit, coverage, airborne and ground drift were assessed • Spray volume was up to 79.6% less than constant-rate spray application • Higher canopy deposit and coverage was achieved at middle and late growth stages • Growth stage strongly influenced spray volume, canopy deposit, and spray losses Optimizing spray application efficiency in 3D crops remains a major challenge due to the spatial and temporal variability of canopy structure. Variable-rate spray (VRS) technologies have emerged as a promising solution to address this limitation by adapting spray output to canopy characteristics. Stereo vision systems have recently gained attention as a cost-effective real-time canopy detection sensor. Despite encouraging results, previous research has been conducted with prototype sprayers. In this study, a commercial airblast sprayer retrofitted with a stereo vision controlled spray system was evaluated and compared with a constant-rate spray (CRS) application across three grapevine growth stages (BBCH 53, 57, and 77). Spray volume, canopy deposit, spray coverage, airborne drift, and ground drift were assessed. The VRS application reduced spray volume by 79.6% and 43.0% at the early and middle growth stages, respectively, whereas a 22.0% increase was observed at the late growth stage compared with CRS application. Despite the spray volume reduction at the early growth stage, canopy deposit and spray coverage decreased to a lesser extent. At the middle and late growth stages, canopy deposit and spray coverage were higher with VRS application than with CRS application. Airborne and ground drift were reduced by 29.3% and 48.9%, respectively, at the early growth stage. At the middle growth stage, airborne drift increased by 44.0% while ground losses decreased by 23.0%. At the late growth stage, airborne and ground drift increased by 30.9% and 55.9%, respectively. Despite promising results, further optimization of spray dose according to canopy development is required.
Why it matches plant phenotyping methodsステレオビジョンでブドウ樹冠の構造を検出し、その情報に基づく可変散布システムを技術評価している。樹冠への散布量・付着・被覆など植物状態に関する取得性能が中心で、単なる農薬試験ではない。
abstractStereo vision systems have recently gained attention as a cost-effective real-time canopy detection sensor.
Phenomic prediction (PP) is a genetic value prediction method based on near infrared spectroscopy (NIRS). Spectra pre-processing is a key step in the analysis pipeline of PP and generally involves chemometrics methods. However, the choice of pre-processing is usually done either arbitrarily or through a search of the optimal set of methods and associated parameters. In this study, we propose to implement a singular value decomposition (SVD) step in the pre-processing pipeline where genetic values of spectra are estimated on a set of principal components instead of individual wavelengths. This way, estimations are based on a few informative, orthogonal and interpretable features of spectra instead of many correlated, uninformative wavelengths. We tested this pre-processing method on five datasets representing four plant species (maize, rice, sorghum and grapevine). Results show that estimating genetic values on components of raw spectra, that are not weighted by their eigenvalues, performs as well as doing it on spectra pre-processed with the best classical chemometrics methods in most cases, while requiring less parameter optimization. Moreover, this SVD step opens up possibilities for better understanding and selecting parts of the spectral information that are relevant for PP. Plain language summary Cultivated plants are the result of a breeding process during which their genetic values are used to select those to breed. Estimating these values requires heavy experimental means and is time consuming. Phenomic prediction is a low cost and high throughput method that is increasingly being used for this purpose. It often uses, as predictors, near infrared spectroscopy measurements that are easy to collect and thus routinely used in many species. However, near infrared spectra generally require pre-processing before being used in prediction. Currently used pre-processing methods arise from the chemometrics community, and still deserve a better in-depth appropriation by geneticists. In this study, we propose a pre-processing approach that performs as well as the best chemometrics pre-processing generally used, reduces computation time, and allows for a better understanding of what parts of spectral information are relevant for prediction. Core Ideas The SVD-based pre-processing performs as well as the best performing classical chemometrics pre-processing in most cases Using the SVD-based pre-processing reduces computing time of genetic value estimation and requires less parameter optimization than using classical chemometrics pre-processing Spectra are composed of chemical and physical information and classical pre-processing methods remove the physical part of the signal It is likely that chemical information is the most important for phenomic prediction even though physical information remains valuable Performance of the SVD-based pre-processing is likely due to a good estimation of the genetic part of spectra and the conservation of physical information of spectra
Why it matches plant phenotyping methods植物のNIRSスペクトルから遺伝的価値を推定するフェノミック予測について、SVDベースの前処理法を提案し、複数植物種のデータセットで既存法と比較検証しているため、フェノタイピング手法が中心である。
abstractIn this study, we propose to implement a singular value decomposition (SVD) step in the pre-processing pipeline where genetic values of spectra are estimated on a set of principal components instead of individual wavelengths.
Crop diseases significantly reduce agricultural output and are a serious problem, especially in the parts of the world where diagnostic experts are not readily available. Deep learning has recently shown us that it is possible for a computer to identify plant diseases directly from images of the leaves. Nevertheless, to make such solutions available on the web or mobile devices one has to really think about how heavy the calculations will be, how easy the user interface should be, and also the limit on the data used. Here is a paper on a web-based applied deep learning system for disease detection in multiple crops. The system detects disease in eight crops Apple, Banana, Grape, Mango, Cauliflower, Tomato, Potato, and Corn with each crop having several disease classes and healthy samples. Three transfer-learning-based CNN architectures MobileNetV3, EfficientNetB4, and ResNet50 were compared for classification performance on the public datasets collected from PlantVillage, Kaggle, and Mendeley. Considering class-wise accuracy, prediction time, and deployment scenarios, MobileNetV3 was picked as the main model to be integrated into the system. To compensate for the differences in image quality often found in pictures taken by users, an optional super-resolution preprocessing step with Real-ESRGAN is added and quantitatively assessed. Disease prediction with spectral activation maps (Grad-CAM) enhances the model's interpretability by highlighting image areas where the disease is detected. The resulting model is embedded in a multilingual Progressive Web Application (PWA). The platform enables users to submit their crop images and receive predicted disease names and treatment options, which are generated by a Large Language Model (LLM) using structured disease metadata. The research acknowledges dataset bias and limitations in extrapolating from curated datasets to the general real-world setting although it reports very good performance of the method on the test sets. In summary, the system proposed here is intended as a practical digital agriculture decision-support tool that demonstrates deployment feasibility and raises a few issues for future validation at the field level and improvement.
Why it matches plant phenotyping methods葉画像から植物病害を推定するCNN手法の比較・前処理評価・実装を中心とした研究であり、植物の病害状態を直接評価するため、植物フェノタイピング手法として中心的です。
abstractDeep learning has recently shown us that it is possible for a computer to identify plant diseases directly from images of the leaves.
Abstract Grapevine crops often suffer from various diseases, posing significant challenges to agricultural production. The diverse morphology and dense distribution of grape leaf diseases complicate identification efforts. As demand for intelligent devices grows, mobile platforms require more efficient neural network algorithms. To address these challenges, we propose a lightweight grape leaf disease detection model, YOLOv10-RLF, based on the YOLOv10n architecture. We introduce the C2f-RVB module, which separates token and channel mixing to enhance feature extraction and reduce redundancy, achieving a 21.74% reduction in parameters and a 19.54% reduction in Giga Floating-point Operations Per Second(GFLOPs). Additionally, we propose a lightweight detection head, v10_LSCD, utilizing shared convolution to further decrease the model's parameters and operational load. To improve detection accuracy, especially for small targets, we replace the SPPF module with a Feature Pyramid Shared Conv (FPSC) module, enhancing multi-scale feature fusion. We also optimize the loss function using SIOU to boost convergence speed and computational accuracy. Experimental results demonstrate that our model reduces parameters and operations by 26.2% and 30.4%, respectively, while compressing model size by 36.4%, all without sacrificing detection accuracy. This research provides a technical foundation for mobile detection of grapevine diseases and supports precise variable applications in agriculture.
Why it matches plant phenotyping methodsブドウ葉の病害状態を画像から検出する軽量なYOLOv10改良手法を開発しており、病害表現型の取得・推定が研究の中心である。
abstractwe propose a lightweight grape leaf disease detection model, YOLOv10-RLF, based on the YOLOv10n architecture.
This study addresses the early detection of water stress in grapevines ( Vitis vinifera L. cv. Monastrell), a key challenge for precision irrigation. The main objective is to assess the feasibility of VIS-NIR hyperspectral imaging (400-1000 nm) to anticipate water stress, relating the spectral signal to stem water potential. This study was developed over two campaigns, in 2024 and 2025, using 18 potted plants. In 2024, eight vines were irrigated, and the remaining 10 were subjected to water-deprivation treatments, whilst in 2025, all plants were irrigated, but half at a control dose and the rest at a reduced dose equivalent to 33% of the control. Images were acquired over five dates in June 2024 and over seven in June 2025 using a Specim IQ camera; stem potential was also measured to provide a physiological reference. Individual time series were developed, calculating the Mahalanoubis distance in a PCA space. Results revealed a change window between 10 and 13 June, consistent with the divergence in water potential from 17 to 24 June. PCA highlighted spectral regions related to changes in pigments, nitrogen and water content as main indicators of water stress. We conclude that HSI is a promising tool for early water stress detection.
Why it matches plant phenotyping methodsブドウの水ストレス状態を近接ハイパースペクトル画像から推定・早期検出する方法が研究の中心であり、生理学的基準との比較も行っている。
abstractThe main objective is to assess the feasibility of VIS-NIR hyperspectral imaging (400-1000 nm) to anticipate water stress, relating the spectral signal to stem water potential.
Summary (1) Rationale Quantifying and predicting plant morphology is central to understanding development and evolution, yet many plant forms lack homologous features required for traditional morphometrics. We apply the Euler Characteristic Transform (ECT), an injective descriptor from topological data analysis, to encode 2D plant shapes. The ECT converts contours into image-like representations that preserve shape information while enabling deep learning. (2) Methods We computed ECTs for large datasets of leaf and pavement cell shapes and used convolutional neural networks (CNNs) for classification. We also trained CNNs to approximate the inverse mapping, predicting leaf shape masks from radial ECTs. (3) Key results ECT-based models achieved high classification accuracy, surpassing previous approaches on millions of herbarium-derived leaves. Notably, grapevine leaf venation was predicted from blade geometry alone, demonstrating that vascular structure is encoded in the outline. (4) Main conclusion The ECT provides a compact, information-preserving representation of biological shape that integrates naturally with deep learning. It enables both accurate classification and predictive reconstruction, revealing latent morphological information and offering new opportunities to study plant form across scales.
Why it matches plant phenotyping methods植物形状を定量化・分類し、葉形状や葉脈を推定するECTベースの計算手法を中心に開発・評価しているため、植物フェノタイピング手法研究に該当する。
abstractQuantifying and predicting plant morphology is central to understanding development and evolution
Plant diseases can damage crops and reduce their growth, which often results in economic problems in agriculture. Detecting these diseases at an early stage is essential to protect crop health and improve productivity. In this work, we developed an automated plant disease identification system using deep learning and leaf images from crops such as corn, grapes, and apples. Our introduced model is based on EfficientNetB0, which showed outstanding performance in classifying multiple disease categories. To make the system more reliable, data augmentation techniques were used to manage different lighting condition in lighting, angles, and backgrounds. The model obtained an accuracy of 97.92%, outperforming other architectures like InceptionV3, MobileNetV2, DenseNet121, Xception, and VGG16 with a lesser accuracy of 83.33%, 78.33%, 77.08%, 75%, 73.33% respectively. Performance measures like precision, recall, and F1-score confirmed its strong performance. The confusion matrix further showed that the model effectively distinguishes between healthy and diseased leaves. This approach provides a fast and accurate solution for real-time disease detection. Overall, the proposed system can support farmers in taking timely action and promoting sustainable agricultural practices.
Why it matches plant phenotyping methods葉画像から植物の病害状態を推定する深層学習手法を開発・比較評価しており、植物表現型の取得・分類が研究の中心であるため。
abstractwe developed an automated plant disease identification system using deep learning and leaf images
Sugar and organic acid content are crucial factors determining grape quality. Non-destructive testing of these components aids in accurately determining optimal harvest timing and wine-making potential. However, few studies have addressed how varietal differences impact the universality of predictive models. This study combines near-infrared spectroscopy with machine learning, specifically partial least squares regression (PLSR) and convolutional neural networks (CNN). It compares single-variety and mixed-variety modeling strategies for grape sugars (glucose, fructose) and organic acids (malic acid, tartaric acid, shikimic acid). The results indicate that PLSR models constructed based on single varieties demonstrate superior performance in predicting malic acid, glucose, and fructose, with model R 2 P ranging from 0.835 to 0.923, notably outperforming PLSR and CNN models based on mixed varieties. The competitive adaptive reweighted sampling (CARS) and successive projections algorithm (SPA) algorithms successfully compressed the full-spectrum variables to 6-29 key wavelengths. The simplified models maintained high accuracy (R 2 P = 0.777-0.927) while substantially improving model efficiency. Mechanistically, SHapley Additive exPlanations (SHAP) analysis revealed the significance of key variables. The critical variables for glucose and fructose models occur around 1150 nm and 1450 nm, respectively. In contrast, the feature variables for the malic acid model exhibit broader distribution, spanning multiple bands including 1150 nm, 1200 nm, 1600 nm, and 1650 nm. This study provides a solid foundation and mechanistic explanation for non-destructive grape quality assessment, while also offering theoretical support for developing specialized spectral sensors.
Why it matches plant phenotyping methodsNIR分光と機械学習によりブドウ果実の糖・有機酸を非破壊推定し、品種別モデルの比較、波長選択、精度評価を行う方法中心の研究である。
abstractThis study combines near-infrared spectroscopy with machine learning, specifically partial least squares regression (PLSR) and convolutional neural networks (CNN).
Double-cropping grape systems offer enhanced land productivity but face significant challenges from climate variability, particularly rain stress and pest outbreaks during critical phenological stages. Accurate phenological prediction is essential to synchronize management practices with crop development and improve ecological resilience. This study presents a novel deep learning framework that integrates MobileNet with an augmented version of Dream Optimizer [Augmented Dream Optimizer (ADO)] to model grape phenology using satellite-derived time series of Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and rainfall from the CropHarvest dataset. The model transforms temporal data into pseudo-images for efficient spatiotemporal feature extraction, achieving 93% classification accuracy and a 6.1-day mean absolute error in stage prediction. ADO enhances convergence and generalization by optimizing key hyperparameters through a hybrid metaheuristic search. The system further identifies high-risk periods for rain damage and pest infestation, enabling proactive interventions. The model statistically significantly outperforms machine learning and deep learning baselines (p< 0.01) across three different agroecological zones [Mediterranean (California, USA, and southern Europe), subtropical (South Australia), and temperate (central Europe)] through spatially stratified fivefold cross-validation on 3,000 held-out test samples of the CropHarvest dataset. This work demonstrates the potential of optimized lightweight neural networks in sustainable viticulture, providing a scalable tool for precision management in climate-resilient double-cropping systems.
Why it matches plant phenotyping methods衛星時系列からブドウの生育ステージを推定する深層学習手法を開発し、交差検証とベースライン比較で性能を検証しており、植物フェノタイピング手法が中心である。
abstractThis study presents a novel deep learning framework that integrates MobileNet with an augmented version of Dream Optimizer [Augmented Dream Optimizer (ADO)] to model grape phenology using satellite-derived time series of Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and rainfall from the CropHarvest dataset.
Climate change, particularly increasing frequency and intensity of spring frost events, poses a serious threat to viticulture by reducing yield and product quality. This study proposes an image processing and machine learning-based framework for early, rapid, and accurate segmentation of frost damage in vineyards using YOLOv11s enhanced with Atrous Spatial Pyramid Pooling (ASPP). A unique dataset called FGVL dataset from Sultana seedless grape vineyards in Manisa, Türkiye, following a severe frost event in April 2025. FGVL includes 418 frost-damaged grapes, 510 frost-damaged leaves, 395 healthy grapes, and 698 healthy leaves, all manually annotated by experts under natural field conditions. By integrating ASPP into YOLOv11s, proposed model improved multi-scale contextual feature extraction and achieved mAP@50 of 0.7686, demonstrating stronger performance in instance segmentation of small, overlapping, and visually similar grapevine organs. In addition, Dynamic Confidence Thresholding (DCT) strategy was introduced to improve prediction reliability in dense and visually complex vineyard scenes. Despite challenges such as background clutter, object overlap, and small target structures, model maintained stable performance with low computational demand, requiring only 6.45 GB of GPU memory. Proposed framework offers an accurate, efficient, and practically deployable early recognition system for frost damage assessment in viticulture.
Why it matches plant phenotyping methodsブドウの器官における霜害状態を画像からセグメンテーションする手法を開発・評価しており、植物の病害・障害状態の取得が研究の中心である。
abstractThis study proposes an image processing and machine learning-based framework for early, rapid, and accurate segmentation of frost damage in vineyards using YOLOv11s enhanced with Atrous Spatial Pyramid Pooling (ASPP).
Reproduction assets foundThe paper's Data availability statement explicitly shares the FGVL frost-damage dataset and source code in the corresponding author's public GitHub repository, matching an allowed URL.Code · publicSource code and dataset are publicly shared in GitHub repository of corresponding author. GitHub repo: https://github.com/kaanarikk/Grape-Instance-Segmentation-For-ViticultureOpen asset ↗https://github.com/kaanarikk/Grape-Instance-Segmentation-For-Viticulturelines:230-236Plant phenotyping relevance match · UnverifiedEurope PMC · checked 14 Sept 2026
Published10 Apr 2026Spectrochimica acta. Part A, Molecular and biomolecular spectroscopyCited by 0 · OpenAlex ↗
Near- and short-wave infrared spectroscopy has become increasingly relevant for the assessment of plant composition, yet the reproducibility of chemometric models across different instruments remains a major limitation. This work proposes a methodological approach for calibration transfer from a benchtop hyperspectral imaging (HSI) system to a portable SWIR spectrometer, using powdered grapevine leaves as a controlled case study. The strategy relies on Direct Standardization (DS), a technique that transforms spectra acquired under different conditions into a common space. By applying DS to a representative subset of samples, models trained under laboratory conditions can be adapted to spectra acquired with portable devices. Two complementary modelling strategies were adopted: an Error-Correcting Output Codes Support Vector Machine (ECOC-SVM) classifier, used to assess qualitative improvements in class discrimination before and after DS, and an eXtreme Gradient Boosting (XGB) regression model, developed for quantitative prediction of elemental concentrations measured by micro-XRF. Validation on an independent set of homogenized leaf powders confirmed that DS markedly reduced inter-instrument spectral divergence, improving class separability and enabling accurate regression of macro- and micronutrients (Fe, Mn, P, S, Zn, Ca, K, Si). Although limited to laboratory-scale samples, the study demonstrates that calibration transfer is effective in harmonizing spectral domains. The proposed workflow provides a reproducible and scalable methodology for cross-instrument adaptation, with potential applicability to diverse agricultural products and portable spectroscopy platforms.
Why it matches plant phenotyping methods植物葉のスペクトルから元素濃度を推定する測定ワークフローを対象に、装置間キャリブレーション転移を開発・検証しており、植物形質取得法が研究の中心である。
abstractThis work proposes a methodological approach for calibration transfer from a benchtop hyperspectral imaging (HSI) system to a portable SWIR spectrometer, using powdered grapevine leaves as a controlled case study.
ABSTRACT A multi-stage approach for plant leaf disease classification using Convolutional Neural Networks (CNN) and image segmentation is presented. The system identifies plant types and classifies diseases from leaf images. Preprocessing includes noise removal and image enhancement, followed by K-means clustering for segmentation. A hybrid CNN model performs accurate classification of plant species and disease type, and suitable fertilizer recommendations are provided. The method achieves accuracies of 99.8%, 96.5%, and 98.3% on apple, tomato, and grape datasets. The results show improved accuracy and robustness, making the system effective for practical agricultural applications. INDEX TERMS Convolutional Neural Network (CNN), Image Segmentation, K-means clustering, Plant Leaf Diseases.
Why it matches plant phenotyping methods葉画像から病害状態を直接推定するCNN・画像セグメンテーション手法が研究の中心であり、植物病害フェノタイピング手法として該当する。
abstractA multi-stage approach for plant leaf disease classification using Convolutional Neural Networks (CNN) and image segmentation is presented.
This dataset consists of a collection of high-resolution RGB images of grapevine leaves, designed to support research in plant pathology, precision viticulture, and computer vision. The images were collected in situ from experimental and commercial vineyards in the north of Portugal, covering different vineyard conditions and management practices. The dataset includes healthy leaves from three grapevine Portuguese cultivars Loureiro, Viosinho and Malvasia Fina, photographed under natural lighting conditions without artificial adjustments. It is organized into five categories: healthy leaves and leaves showing symptoms of downy mildew ( Plasmopara viticola ), powdery mildew ( Erysiphe necator ), Esca complex and Erineum Mite ( Colomerus vitis ). Images are provided in JPEG format with a resolution of 3000 × 3000 pixels and 1024 × 1024 pixels and arranged in folders by health status and disease type. This dataset can be used for machine learning and deep learning applications in disease detection/classification, cultivar identification, and can support other precision agriculture applications, as well as being used for agricultural robotics and educational purposes. An evaluation on three deep learning architectures demonstrated the suitability of the dataset into separating the five classes.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像化した再利用可能なデータセットで、植物の健康状態・病害状態の画像ベース推定を支えることが中心です。深層学習による5クラス分類評価も記載されています。
abstractThis dataset consists of a collection of high-resolution RGB images of grapevine leaves, designed to support research in plant pathology, precision viticulture, and computer vision.
Reproduction assets foundThe paper is a Data in Brief article describing a public Zenodo repository of RGB grapevine leaf images (healthy plus downy mildew, powdery mildew, Esca complex, erineum mite) collected for plant disease/phenotyping research, with explicit data accessibility details. No author analysis code or trained model checkpointsDataset · publicData accessibility
Repository name: Zenodo
Data identification number: https://doi.org/10.5281/zenodo.17343473
Direct URL to data: https://zenodo.org/records/17343473Open asset ↗Zenodo · 10.5281/zenodo.17343473html-lines:93-144Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 5 Sept 2026
Abstract Grapevine cluster architecture is a key selection target in breeding programs because it influences disease susceptibility, yield stability and juice quality. High-throughput phenotyping offers a rapid and non-destructive approach to capture biochemical and structural variation in these traits, yet the influence of plant organ reflectance and data partitioning strategies on trait prediction remains poorly understood. In this study, we evaluated how hyperspectral reflectance from different grapevine organs contributes to the prediction of cluster architecture and juice quality traits in two clonal populations of Riesling and Pinot. Using partial least squares regression (PLSR), we assessed the prediction accuracy of eight cluster architecture and six juice quality traits under two data partitioning strategies. Models based on cluster reflectance outperformed those using dry leaf reflectance for most traits, except for pH. Partitioning the dataset by cluster type increased trait variance and improved predictions for number of berries (R² = 0.53), berry diameter (R² = 0.79), and total acidity (R² = 0.48). Visible, red-edge and NIR spectra were most informative regions to predict the traits studied. Together, our results highlight the importance of organ-specific data and appropriate calibration strategies to improve phenomic models for the development of scalable proxies for grapevine improvement. Highlight Spectral phenomics reveals that prediction accuracy in grapevine depends on organ spectral signatures and traits, with cluster reflectance outperforming leaves, informing new phenotyping strategies for breeding improvement.
Why it matches plant phenotyping methodsブドウの器官反射スペクトルとPLSRを用いて、房構造および果汁品質形質の予測性能を評価することが中心であり、スペクトル表現型解析手法の検証・応用に該当する。
abstractHigh-throughput phenotyping offers a rapid and non-destructive approach to capture biochemical and structural variation in these traits
In perennial crops, inner wood degradation by pathogens often escapes detection until irreversible damage has occurred. Grapevine trunk diseases (GTD) are a well-known example in viticulture that alter plants from within, years before foliar symptoms arise, making early assessment difficult. To overcome this limitation, we present a novel non-destructive 3D + t pipeline for high-resolution Magnetic Resonance Imaging (MRI) spatial quantification and monitoring of early internal host tissue degradation resulting from fungal pathogen colonization. The pipeline integrates spatio-temporal anatomical alignment and rigid registration; a generalized cylindrical-coordinate transformation; supervised segmentation of water-depleted regions; and population-level statistical analyses, including population mean images, probabilistic atlases, and 3D lesion descriptors. Applied to multiple Vitis vinifera cultivars inoculated with a fungal wood pathogen, our approach enables in vivo time-lapse comparisons between cultivars and treatments. The results reveal reproducible early degradation signals across individuals and cultivar-dependent differences in lesion progression. Overall, this methodological innovation provides a new paradigm for internal plant phenotyping, enabling non-invasive quantification of disease development and comparative spatio-temporal assessment of host responses in woody plants, with strong potential to advance early diagnosis and management of GTDs and internal diseases.
Why it matches plant phenotyping methodsMRI画像と計算パイプラインにより、ブドウ樹内部の病変・組織劣化を非破壊かつ時空間的に定量化する手法を開発・適用しており、植物表現型取得が中心である。
abstractwe present a novel non-destructive 3D + t pipeline for high-resolution Magnetic Resonance Imaging (MRI) spatial quantification and monitoring of early internal host tissue degradation resulting from fungal pathogen colonization.
Reproduction assets foundThe paper's MRI phenotyping data (~160 GB raw, 1.4 TB processed) is only available upon request, but the authors' processing pipeline (scripts and parameters) is publicly deposited on Zenodo with an explicit URL.Code · publicThe processing pipeline (including scripts and parameters required to reproduce the processed outputs from the raw data) is available at https://doi.org/10.5281/zenodo.17944369 .Open asset ↗Zenodo · 10.5281/zenodo.17944369lines:370-484Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
This study addresses the challenge of detecting white grape clusters (Vitis vinifera L) in high-density vineyard canopies, a critical task for precision viticulture and yield estimation. Traditional statistical and image-processing methods have struggled with occlusion issues. In this work, over 100 field RGB images were collected at La Bergonza (Toledo, Spain) and expanded through data augmentation, with various preprocessing strategies tested to enhance cluster visibility. Convolutional Neural Network (CNN) architectures were compared, highlighting YOLOv8 as superior to Mask R-CNN in both accuracy and efficiency. YOLOv8, trained for up to 100 epochs on equalized and augmented datasets, achieved outstanding performance: 84.9% precision, 72.6% recall, and mAP@0.5 of 83%, far surpassing Mask R-CNN (17% precision, 26% recall). The model successfully detected partially hidden clusters, including those invisible to human experts, better than previous studies that required controlled backgrounds or artificial lighting. Results confirm that combining RGB equalization with data augmentation optimizes detection. These findings underscore the potential of deep learning and low-cost RGB imaging systems to enable automated, scalable solutions for yield estimation and canopy analysis. In conclusion, YOLOv8 emerges as a promising tool for accurate grape bunch detection under field conditions, overcoming previous limitations.
Why it matches plant phenotyping methodsブドウ房を対象としたRGB画像とCNNによる検出手法を開発・比較し、精度を定量評価しているため、植物器官の表現型取得が中心である。
abstractIn this work, over 100 field RGB images were collected at La Bergonza (Toledo, Spain) and expanded through data augmentation, with various preprocessing strategies tested to enhance cluster visibility.
Reproduction assets foundThe paper's Data Availability Statement points to the authors' public GitHub repository containing the original grape-cluster image dataset and annotations used in this study. The ultralytics repository is a generic third-party library, not a paper-specific asset.Dataset · publicData Availability Statement: The original data presented in the study are openly available at
[https://github.com/upmValeriano/racimosUva.git.]Open asset ↗upmValeriano/racimosUvapdf-page:13 lines:1-66Code / dataset availability confirmedEurope PMC · bioRxiv · checked 5 Sept 2026
Grapevine Trunk diseases (GTDs) represent a major threat for the wine industry. Despite several break-through, their etiology remains unclear and no curative treatment is currently available. Wood anatomy and water transport contribute to the symptoms of young plant decline. This study investigates wood anatomical alterations in two Alsatian grapevine cultivars presenting different susceptibility to GTDs, focusing on wood structure over six months of vegetative growth and in response to infection. Using a validated FasGa staining protocol, wood sections from transverse, tangential, and radial directions were stained to differentiate lignified and cellulosic tissues. Microscopic analysis was performed at x4, x10, and x40 magnifications, yielding a dataset of 4771 images. To support this high-throughput quantitative analysis of microscopy images, a computational model was developed, enabling reliable and efficient assessment of anatomical traits. Pre-established woody tissues presented higher xylem vessels diameter in Gewurztraminer than Riesling, with a dorsoventral arrangement whereas the number of vessels remained the same all over the cross section. No significant anatomical changes were observed in established woody tissues, whereas newly formed xylem anatomy showed a possible rearrangement during infection, especially in Gewurztraminer cultivar. Furthermore, colorimetric analysis quantified the lignification of woody tissues in response to wounding damage compared to un-treated plants. While definitive conclusions remain limited due to the experimental timeframe and sample variability, the findings highlight the need for longer-term studies and broader cultivar evaluation. Code and microscopy images have been made publicly available, providing a scalable digital tool for future research in plant vascular systems.
Why it matches plant phenotyping methods植物組織画像から木部解剖形質と木化を定量する計算モデルを開発・検証し、大規模画像データセットと公開コードを提供しており、表現型取得手法が研究の中心である。
abstractTo support this high-throughput quantitative analysis of microscopy images, a computational model was developed, enabling reliable and efficient assessment of anatomical traits.
Reproduction assets foundThe paper's microscopy image dataset (4771 grapevine wood images) is publicly deposited on Zenodo with an explicit DOI matching an allowed URL. The authors also state their Python analysis pipeline is available at github.com/courbot/vineside, but that URL is not among the allowed URLs, so only the Zenodo image dataset,Dataset · publicThis database can benefit the research community, and is publicly available
online at https://doi.org/10.5281/zenodo.18850060 [35].Open asset ↗Zenodo · 10.5281/zenodo.18850060pdf-page:4 lines:1-56Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
Accurate and non-destructive evaluation of grape quality is crucial for intelligent viticulture, yet most existing approaches address cultivar classification and soluble solid content (SSC) prediction as independent tasks based on single-modality data, limiting robustness and practical applicability. This study proposes DualStream-RTNet, a unified multimodal deep learning framework that simultaneously performs grape cultivar classification and SSC prediction by integrating RGB-HSV fused images and PCA-compressed hyperspectral spectra. The dual-stream architecture enables the complementary learning of external chromatic-textural cues and internal physicochemical information, while a Transformer-enhanced fusion module strengthens global representation and cross-modal correlation. A dataset of 864 berries from five grape cultivars was used to validate the model. DualStream-RTNet achieved 93.64% classification accuracy, outperforming ResNet18 and other CNN baselines, and produced more compact and consistent confusion-matrix patterns. For SSC prediction, it consistently yielded the highest performance across cultivars, with R2p values up to 0.9693 and RMSE as low as 0.2567, surpassing the PLSR, SVR, LSTM, and Transformer regression models. These results demonstrate the superiority of the proposed framework in capturing both visual and spectral characteristics. DualStream-RTNet provides an efficient and scalable solution for comprehensive grape quality assessment, offering strong potential for real-time sorting, precision grading, and smart agricultural applications.
Why it matches plant phenotyping methodsRGB-HSV画像とハイパースペクトルを統合し、ブドウ果実のSSCを非破壊推定する新規モデルを開発・検証しており、果実形質の取得・推定法が研究の中心である。
abstractThis study proposes DualStream-RTNet, a unified multimodal deep learning framework that simultaneously performs grape cultivar classification and SSC prediction by integrating RGB-HSV fused images and PCA-compressed hyperspectral spectra.
Purpose The study aims to develop a method for detecting and discriminating grapevine yellows (GY) diseases, including Flavescence dorée (FD), Bois noir (BN), and Palatinate grapevine yellows (PGY), using hyperspectral imaging (HSI). Specifically, it seeks to address the challenge of symptom misclassification among visually similar diseases, such as distinguishing between GY diseases and other biotic and abiotic stresses. Methods Hyperspectral images of detached leaves from field with GY diseases, Grapevine Leafroll-associated Virus (V), leafhopper (LH) infestation, iron deficiency (Fe), or magnesium deficiency (Mg) were acquired using a mobile platform in the laboratory or in the field. The images were taken in the spectral range of 400-1,000 nm. Classification models were trained on the mean spectra of the leaves to distinguish between different classes. Results The model achieved F1-scores ranging from 54.2% to 96.8% for white cultivars and exceeded 95% for all six classes for black cultivars. When limited to GY classes and nonsymptomatic leaves, the F1-scores were 89.8%, 78.7%, 63.4%, and 53.4% for nonsymptomatic, FD, PGY, and BN, respectively. Conclusion The study demonstrates the feasibilityof using HSI to detect and discriminate different biotic and abiotic stresses on grapevine leaves, including distinguishing between symptom-similar GY diseases. The developed mobile platform and classification models show promise for largescale monitoring and diagnosis of GY diseases, potentially improving disease management and reducing the risk of symptom misclassification.
Why it matches plant phenotyping methodsブドウ葉のハイパースペクトル画像から病害・ストレス状態を識別する手法と移動型プラットフォーム、分類モデルを開発・評価しており、植物表現型取得が中心である。
abstractThe study aims to develop a method for detecting and discriminating grapevine yellows (GY) diseases, including Flavescence dorée (FD), Bois noir (BN), and Palatinate grapevine yellows (PGY), using hyperspectral imaging (HSI).
Advanced methods are necessary to improve the detection of flavescence dorée (FD), one of the most relevant grapevine ( Vitis vinifera ) diseases in Europe, caused by flavescence dorée phytoplasma (FDp). Detection is commonly carried out visually by agronomists/winegrowers, and is time-consuming and error-prone. The present study demonstrated that full-range hyperspectral data (i.e., 400 to 2,400 nm) collected at leaf level can be used as a tool to rapidly and nondestructively detect FD infection directly in the field. Focusing on a Sangiovese (red grape) vineyard of Tuscany (Central Italy), heavily affected by FD (incidence higher than 75%), we showed that the proposed hyperspectral approach is capable of (i) detecting FDp infection, even before the occurrence of leaf symptoms (accuracy >70%); (ii) discriminating FD-induced leaf symptoms, even between asymptomatic and lightly symptomatic leaves (accuracy 80%); and (iii) elucidating the complex physiological responses of grapevines to FDp infection, with changes in leaf parameters estimated from spectra suggesting that the disease not only impaired early-season photosynthetic efficiency but also accelerated leaf senescence, potentially impacting grapevine productivity and grape quality. Although the hyperspectral approach proposed here is not intended to replace traditional diagnostic methods (molecular analyses), it could serve as a valuable tool to support the monitoring of plants affected by FD and may represent a crucial advancement in FD management. Further and broader studies including vineyards less challenged by FD and with other grape varieties (e.g., white ones, showing leaf yellowing instead of reddening) are encouraged.
Why it matches plant phenotyping methods葉のハイパースペクトル計測により、ブドウの病徴・感染状態・生理状態を推定する手法が研究の中心であり、植物病害フェノタイピングに該当する。
abstractfull-range hyperspectral data (i.e., 400 to 2,400 nm) collected at leaf level can be used as a tool to rapidly and nondestructively detect FD infection directly in the field.
The objective of this study is to create a highly accurate and interpretable deep learning (DL) model for the multi-class classification of fruit using convolutional and transformer architectures. The classification performance can be enhanced by making sure that the used technique is explainable and interpretable. This research data was obtained from Kaggle which contains images of banana, grape, lemon, mango, and strawberry fruit classes. The total data was divided into 70:15:15 for training, validating and testing. To ensure consistent size and quality, all images were pre-processed before use. This study considered four pretrained models namely RegNetY-B3-GE, DarkNet53-SCSE, BEiT, and PVTv2 for performance assessment. We proposed a lightweight hybrid (convolution plus attention-based) CoAT-AgriLite model for fruit disease classification which extracts local lesion features and global context. Transferring training and data augmentation technique was utilized during training for better performance. To ensure interpretability of model decisions, Gradient-weighted Class Activation Mapping (Grad-CAM) which captures the discriminative regions from the input images for model predictions. Among all evaluated models, the proposed model achieved the highest classification accuracy of 99.37% on the testing dataset. Comparative results demonstrated that the proposed model outperformed other pretrained models in terms of precision, recall, and F1-score, confirming its robustness and effectiveness in real-world agricultural classification tasks. The experimental findings validate that the proposed model not only achieves superior classification accuracy but also provides interpretability through Grad-CAM visualizations. This hybrid framework offers a promising solution for intelligent and transparent fruit classification systems, with potential applications in precision agriculture and automated sorting systems.
Why it matches plant phenotyping methods果実病害を画像から分類する深層学習フレームワークを開発・比較し、病斑特徴の抽出とGrad-CAMによる説明性を評価しており、植物の病害状態の画像計測が中心である。
titleA hybrid convolution and attention-based framework with visual explanation for fruit disease identification.
Service crops are grown to provide ecosystem services in viticulture, but their adoption remains limited due to their competition with grapevine for soil resources. To identify trade-offs between services, the effect of service crops management strategies on grapevine performances still need further research. This dataset presents data from two experiments conducted to study the effect of service crops management on soil resources and grapevine performances. The inter-row vegetation was sampled in two Mediterranean vineyards using quadrats for biomass estimation. In addition, an unmanned aerial vehicle (UAV) was regularly flown over the vineyards for a period spanning more than four years in total over the two vineyards. The dataset presented here includes both raw data acquired during fieldwork and processed data derived from this raw inputs. The raw data consists of image series captured by two UAVs during each flight campaign, including RGB and multispectral imagery. Images were acquired between 2021-06-10 and 2022-07-29 for the first vineyard, and between 2023-06-08 and 2025-03-12 for the second vineyard. Based on these raw data, the processed data comprises spatial vectors, raster layers, and dense point clouds generated from UAV images using a Structure from Motion (SfM) photogrammetry workflow, at a 5 cm spatial resolution. The raster layers and dense point clouds provide specific information on vineyard characteristics for each UAV flight date, including elevation, vegetation indices, visible and near-infrared reflectance, and canopy height. In addition, the processed data include measurements of vegetation dry biomass, as well as separate measurements of dry biomass and leaf area measured for selected service crops species. This dataset can be reused for the calibration and/or evaluation of classification algorithms aimed at discriminating vines from the inter-row vegetation, or as part of a larger dataset to explore relationships between remotely-sensed vegetation indices and field-measured vegetation biomass or surface.
Why it matches plant phenotyping methodsUAV画像とSfM処理により、植生指数・樹冠高・バイオマス等の植物形質を取得した再利用可能なデータセットで、分類アルゴリズムの校正・評価用途も明示されており、植物フェノタイピング手法・データ基盤が中心です。
abstractThe dataset presented here includes both raw data acquired during fieldwork and processed data derived from this raw inputs.
Reproduction assets foundThe paper is a Data in Brief article describing a public dataset on Research Data Gouv (doi: 10.57745/MXM55R) containing UAV RGB/multispectral imagery, SfM-derived rasters and point clouds, and field-measured vegetation biomass/leaf-area data from two Mediterranean vineyards — directly the paper's phenotyping inputs. ADataset · publicollected in vineyards located in southern France near Montpellier (43°32.5243′N, 3°50.8240′E). Data are stored on Research Data Gouv, a remote storage solution curated by the French Department of Research.
Data accessibility
Repository name: Research Data Gouv
Data identification number: doi: 10.57745/MXM55R
Direct URL to data: https://doi.org/10.57745/MXM55R
Related research article
None
1.
Value of the Data
•
The fine scale imaging of vineyards (i.e., 5 cm resolution) allows for classification of the vegetation in the vineyard inter-rows, and subsequent exploration of its respective dynamics.
•Open asset ↗Research Data Gouv · 10.57745/MXM55Rlines:1-47Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Grapevines ( Vitis vinifera L.) undergo structural and physiological changes throughout the growing season, progressing through distinct phenological stages that require regular monitoring. This dataset consists of high-resolution point cloud data acquired with a stationary terrestrial laser scanner (TLS) to document grapevine development from early leaf development to dormancy. Georeferenced point clouds were generated from 15 TLS scans along two vineyard rows at nine phenological stages. The dataset also includes multispectral and RGB photogrammetric point clouds and orthorectified raster products from an unmanned aerial vehicle survey conducted before harvest. Ground-truth measurements leaf area index, grape production, and pruning wood biomass were collected for each monitored grapevine. As a result, the dataset provides multi-temporal TLS observations that support grapevine structural analysis and development, phenological monitoring, and can be used for the development of AI-based models for precision viticulture.
Why it matches plant phenotyping methodsブドウの生育・構造・フェノロジーを対象とするTLS点群および関連画像データセットであり、植物フェノタイピング用の再利用可能なデータ基盤として中心的です。
titleTLS-grapevine2024: A terrestrial laser scanner point cloud dataset of grapevines at different phenological stages.
Reproduction assets foundThe paper is a Data in Brief article describing the TLS-grapevine2024 dataset itself, publicly deposited on Zenodo with DOI 10.5281/zenodo.16751663. This is a paper-specific, openly available asset containing the TLS point clouds, UAV imagery/rasters, and ground-truth agronomic measurements (LAI, grape production, prunDataset · publicditions: clear sky.
Data source location
Institution: University of Trás-os-Montes e Alto Douro
City/Town/Region: Arroios, Vila Real, Norte
Country: Portugal
Coordinates: 41°17′28.83″N 7°43′17.90″W,
Altitude: 435 m
Data accessibility
Repository name: Zenodo
Data identification number: 10.5281/zenodo.16751663
Direct URL to data: https://doi.org/10.5281/zenodo.16751663
Related research article
None
1.
Value of the Data
•
This dataset covers nine phenological stages of grapevine growth from April 2024 to January 2025, providing multi-temporal terrestrial laser scanner (TLS) observations for structural and phenological analysis.
•
It includes TLS point clouds collected at multiple stages and muOpen asset ↗Zenodo · 10.5281/zenodo.16751663lines:1-50Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Published14 Feb 2026International Journal of Engineering Trends and TechnologyCited by 0 · OpenAlex ↗
The study is a design and development of a strong disease detection system of cucumber and grape leaves with noisy image data, focusing on the ability to withstand salt-and-pepper and Gaussian noises. The image datasets used in agriculture are usually affected by noise because of changes in light, sensor defects, and environmental conditions, which may lead to lower diagnostic accuracy. In order to address this, the proposed system incorporates high noise reduction methods whereby a median filter and a Gaussian filter are used to restore the image quality without compromising on the important leaf texture information. After processing, colour, texture, and shape are used to extract features, which are effective in extracting disease-specific visual representations. These fine features are then trained on various optimized machine learning models, such as Light Gradient Boosted Machine (LGBM), Quantum Support Vector Machine (QSVM), a Modified Random Forest (MRF) with adaptive weighted features, and a Multi-SVM classifier with a custom kernel to map nonlinear features. Through experimental analyses, the proposed ensemble framework is shown to be highly accurate, robust, and noise-tolerant as opposed to the traditional frameworks. The hybrid method is effective in recognizing the significant cucumber diseases and grapes, including powdery mildew, downy mildew, and anthracnose, which will be utilized in the noisy agricultural conditions in the real world. In general, this system offers a noise-resistant, reliable, and computationally efficient system to detect early signs of plant diseases, which can be used in sustainable crop monitoring and precision farming.
Why it matches plant phenotyping methods植物葉の画像から病徴を推定する画像処理・特徴抽出・機械学習システムの設計開発が中心であり、植物病害状態のフェノタイピング手法に該当する。
abstractThe study is a design and development of a strong disease detection system of cucumber and grape leaves with noisy image data
Reproduction assets foundThe paper uses two public Kaggle leaf-image datasets (cucumber and grape) as its phenotyping inputs; both are publicly accessible with URLs given in the references. No author code or model checkpoints are reported.Dataset · publicGrape Disease Dataset, which was collected on Kaggle
[17], is an extensive collection of images created for the
classification and analysis of different diseases in grape
leaves.Open asset ↗Kagglepdf-raw-page:7 lines:1-64Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 5 Sept 2026
In the recent era, the growth of deep learning is inevitable. Various models such as convolutional neural networks (CNNs) and transformers are used widely in images for high classification accuracy. Since the invention of transformers, researchers have widely used novel approaches using transformers to achieve an impressive accuracy. In spite of this, this paper proposes a novel custom lightweight CNN model called Attentive and Lightweight Network (ALNet). ALNet consists of three major blocks: stem, core, and head. The core part is the novel classifier built as an inspiration from various pre-trained models such as ResNet, SENet (Squeeze and Excitation Network), EfficientNet, SqueezeNet, and ShuffleNet. The main objective is to build a model that has a high classification accuracy while reducing the number of parameters. This reduces the size of the model and hence makes it easy to deploy on cloud platforms and use in edge devices. The model was evaluated using 5-fold cross-validation on three different datasets. The primary dataset was a grapevine dataset with an accuracy of 99.78 percent and 100 percent in multi-class and binary classification respectively. To test the robustness of the model, a multi-class classification using the apple dataset achieved an accuracy of 99.95 percent and a binary classification with the cherry dataset achieved an accuracy of 100 percent. ALNet uses only 0.17 million parameters which is 18 times less parameters than the lightest model (SqueezeNet) and it takes only 14 seconds to train each epoch while pretrained models take 17–31 seconds. ALNet requires only 151.98 MFLOPs with a model size of 677.20 KB, making it approximately 18 times smaller than SqueezeNet. On the whole, ALNet is a highly accurate, lightweight model for plant leaf diseases prediction.
Why it matches plant phenotyping methods葉画像から植物病害状態を分類する軽量CNNを開発・検証しており、病害表現型の画像ベース抽出手法が研究の中心である。
abstractthis paper proposes a novel custom lightweight CNN model called Attentive and Lightweight Network (ALNet).
The study of annual growth rings in Vitis vinifera has recently emerged as a promising framework to explore long-term vine responses to climate and management. This review synthesizes current knowledge on grapevine wood anatomy, xylem functionality, and isotopic signatures, highlighting their role as archives of environmental and agronomic information. Grapevine growth rings provide high-resolution records of climate signals, including drought and heat stress, and reveal cultivar-specific hydraulic strategies shaped by soil conditions and management practices. Stable isotope analyses further complement anatomical chronologies by integrating physiological responses to water availability. Together, these approaches offer valuable insights into the structural memory and plasticity of grapevine xylem, with direct implications for vineyard sustainability and climate adaptation. We also examine the impact of viticultural practices such as irrigation, pruning, grafting, and rootstock choice on xylem architecture, emphasizing how agronomic decisions leave long-lasting anatomical imprints in wood. Finally, we outline future research perspectives, including the integration of dendrochronological, isotopic, and high-resolution imaging techniques, to fully exploit grapevine chronologies as tools for understanding vine resilience and for guiding varietal selection, breeding, and precision viticulture under changing environmental conditions.
Why it matches plant phenotyping methodsブドウの年輪解剖、安定同位体、画像解析を用いて環境応答や木部機能などの植物形質・状態を評価する方法群をレビューしており、測定・解析手法が中心的である。
abstractGrapevine growth rings provide high-resolution records of climate signals, including drought and heat stress, and reveal cultivar-specific hydraulic strategies shaped by soil conditions and management practices.
The leaf blade and vasculature develop together within a shared morphological space. Despite shared molecular patterning pathways, it is unknown if developmental and evolutionary variation affect these tissues separately or together in a coordinated way. Grapevine leaves have a morphometric history and abundant data measuring the shape of the blade and vasculature together. Using a combination of topological data analysis and deep learning, we perform reciprocal semantic segmentation of leaf blade and vasculature. Each tissue contains sufficient information to predict the other. We hypothesize that this is due to a one-to-one relationship between blade and vein. Using thin plate splines to swap and warp different combinations of blade and vein shapes, we show that a set of leaves with a many-to-one relationship of blade and vein are distinguishable from true leaves. We also swap blade and vein across the developmental series and between species and show that only reversing the developmental series disrupts the relationship between blade and vasculature. We end by discussing the evolutionary and developmental implications that there is a unique, one-to-one mapping between blade and vein that allows each to be predicted from the other. Author summary Leaves are made of two closely connected parts: the flat blade that captures light and the network of veins that transports water, nutrients, and developmental signals. Although these tissues grow together and share common molecular patterning pathways, it has remained unclear whether a particular blade shape is uniquely linked to a specific vein pattern. In this study, we use grapevine leaves as a model system and combine mathematical shape analysis with deep learning to examine this relationship. We show that the shape of the blade alone can accurately predict the vein network, and that the vein network can likewise predict the blade. This finding suggests a near one-to-one relationship between these two tissues. To test this idea, we created artificial leaves in which blade and vein shapes were deliberately mismatched. Although these synthetic leaves appeared realistic at a global level, a neural network was able to distinguish them from real leaves based on subtle differences. We further show that this tight coupling is maintained by the developmental sequence of leaf growth rather than by species identity, revealing a conserved constraint linking leaf form and internal structure.
Why it matches plant phenotyping methods葉身と葉脈の形状を深層学習による相互セグメンテーションと形状解析で抽出・予測する方法が研究の中心であり、植物形態表現型の方法開発に該当する。
abstractUsing a combination of topological data analysis and deep learning, we perform reciprocal semantic segmentation of leaf blade and vasculature.
Reproduction assets foundThe Data Availability Statement explicitly lists three public Zenodo deposits containing data and code to reproduce the paper's analyses: reciprocal U-Net blade/vein prediction, the two-tower CNN 1:1 vein:blade relationship, and interspecies/intraspecies developmental series swaps. All URLs are in allowed_urls and the Code · public393
which the relationship between blade and vasculature is conserved or diversified across the
394
spectacular variety of leaf shapes remains to be seen.
395
Data Availability Statement
396
Data and code to reproduce this work can be found for the following analyses: Reciprocal
397
prediction of vein and blade from the other, https://zenodo.org/records/16920155; Two tower CNN
398
1:1 vein:blade relationship, https://zenodo.org/records/17014105; Interspecies and Intraspecies
399
developmental series swaps, https://zenodo.org/records/17013783
400
Conflict of Interest Statement
401
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the aOpen asset ↗zenodo · 16920155pdf-raw-page:22 lines:1-53Code · publicd across the
394
spectacular variety of leaf shapes remains to be seen.
395
Data Availability Statement
396
Data and code to reproduce this work can be found for the following analyses: Reciprocal
397
prediction of vein and blade from the other, https://zenodo.org/records/16920155; Two tower CNN
398
1:1 vein:blade relationship, https://zenodo.org/records/17014105; Interspecies and Intraspecies
399
developmental series swaps, https://zenodo.org/records/17013783
400
Conflict of Interest Statement
401
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuitOpen asset ↗zenodo · 17014105pdf-raw-page:22 lines:1-53Code · publicment
396
Data and code to reproduce this work can be found for the following analyses: Reciprocal
397
prediction of vein and blade from the other, https://zenodo.org/records/16920155; Two tower CNN
398
1:1 vein:blade relationship, https://zenodo.org/records/17014105; Interspecies and Intraspecies
399
developmental series swaps, https://zenodo.org/records/17013783
400
Conflict of Interest Statement
401
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted January 29, 2026.
;
https:Open asset ↗zenodo · 17013783pdf-raw-page:22 lines:1-53Plant phenotyping relevance match · UnverifiedCrossref · checked 5 Sept 2026
Abstract Field experiments are complex to interpret due to interactions between genotypes, environment, plant development and cultivation practices. This complexity challenges the accurate phenotyping of individual plant traits over the season. Here, we quantified the primary sources of seasonal variation in stomatal conductance (gs) across 15 grapevine cultivar–rootstock combinations within a large-scale phenotyping platform, comprising over 6000 observations. Environment-related traits and date of measurement accounted for up to 76% of the variance, potentially obscuring cultivar–rootstock effects. Therefore, we integrated machine learning, spatiotemporal normalization of the gs response, and the use of mixed models to disentangle the influences of environmental factors, plant material and crop performance related traits. After spatio-temporal normalization, cultivar and cultivar–rootstock interactions explained over 25% of the variation in gs, and Grenache exhibited the most conservative water-use behavior resulting in high water-use efficiency. Specific rootstock–scion combinations also exhibited smaller, but still significant, differences in gs and water-use efficiency, highlighting the specificity arising from the interaction within each rootstock–scion combination. The high variability in gs indicates that accurate quantification of rootstock–scion contributions to key traits in field studies is complex and requires accounting for spatial heterogeneity driven by the environment.
Why it matches plant phenotyping methods大規模な圃場フェノタイピングで測定した気孔コンダクタンスを対象に、機械学習、時空間正規化、混合モデルを統合して環境変動と遺伝的要因を分離する手法が中心である。
abstractwe integrated machine learning, spatiotemporal normalization of the gs response, and the use of mixed models to disentangle the influences of environmental factors, plant material and crop performance related traits.
Vegetation volume is a useful indicator for assessing canopy structure and supporting vineyard management tasks such as foliar applications and canopy management. The photogrammetric processing of imagery acquired using unmanned aerial vehicles (UAVs) enables the generation of dense point clouds suitable for estimating canopy volume, although point cloud quality depends on spatial resolution, which is influenced by flight height. This study evaluates the effect of three flight heights (30 m, 60 m, and 100 m) on grapevine canopy volume estimation using convex hull, alpha shape, and voxel-based models. UAV-based RGB imagery and field measurements were collected during three periods at different phenological stages in an experimental vineyard. The strongest agreement with field-measured volume occurred at 30 m, where point density was highest. Envelope-based methods showed reduced performance at higher flight heights, while voxel-based grids remained more stable when voxel size was adapted to point density. Estimator behavior also varied with canopy architecture and development. The results indicate appropriate parameter choices for different flight heights and confirm that UAV-based RGB imagery can provide reliable grapevine canopy volume estimates.
Why it matches plant phenotyping methodsUAV画像からブドウ樹のキャノピー体積を推定する手法を、飛行高度・推定モデル間で評価し、実測値と比較しているため、植物形質取得法の技術的検証が中心です。
abstractThis study evaluates the effect of three flight heights (30 m, 60 m, and 100 m) on grapevine canopy volume estimation using convex hull, alpha shape, and voxel-based models.
Grapevine trunk diseases in subtropical climates show complex patterns of multi-pathogen co-infection and spatial clustering, while current diagnosis still relies mainly on expert judgement with limited quantification and functional testing. This study investigated an 18-acre vineyard in south-eastern Queensland and used 7,440 vine records from 744 plots to build quantitative indices for symptoms and cross-section necrosis, followed by comprehensive characterisation of 46 fungal isolates through isolation, microscopy, physiological assays and greenhouse pathogenicity tests. Analyses identified three spatial disease regions, with wedge- and semi-ring-shaped necrosis strongly enriched in high-disease plots, and showed that Botryosphaeriaceae and Phomopsis groups dominated the pathogen community and had much higher composite pathogenicity indices than other fungi. Even without molecular data, the integrated pipeline of disease quantification, microscopic and physiological traits, pathogenicity testing and computational analysis allowed robust identification of dominant pathogen combinations in a subtropical vineyard and provided a methodological basis for regional risk assessment and targeted management.
Why it matches plant phenotyping methodsブドウ樹の症状と壊死を定量化する指標および統合解析パイプラインが研究の中心で、植物の病害状態を直接測定・評価しているため。
abstractused 7,440 vine records from 744 plots to build quantitative indices for symptoms and cross-section necrosis
• Very-high-resolution UAVs dominate row/plant analyses; Sentinel-2 underpins regional monitoring. • Multi-sensor fusion (UAV/satellite) and 3D bring robustness to vine identification. • Deep Learning enables detection of plots and accurate row delineation in complex terrains. • Deep Learning models requires vast annotated data and high computational cost limits routine use. • Validation and model portability remain the weakest methodological areas. Sustainable vineyard management and planning require reliable methods for identification and monitoring. This systematic review synthesises and appraises the literature on automatic vineyard identification using remote sensing (RS), from classical techniques to artificial intelligence (AI), describing the state of the art, patterns, challenges, and gaps. Guided by PRISMA and informed by selected SWiM reporting items, we conducted a systematic search across multiple databases, gathering all relevant records up to 13 July 2025, and included 108 sources, of which 80 empirical studies contributed to the synthesis. The risk of bias was assessed by adapting the principles of PROBAST-AI and QUADAS-2 to the agricultural context, covering data representativeness, sensors/pre-processing, ground-truth, validation, and portability; its application also guided the selection and organisation of the synthesis. The analysis was narrative and structured by scale and application objective (regional, parcel, row, and plant). The most common tasks were classification (28%), detection (26%), and segmentation (24%), with multitask pipelines being frequent. We observe a clear transition from pixel-based approaches using satellite imagery to methodologies that integrate very-high - resolution UAV imagery, 3D reconstruction, and Deep Learning (DL). UAVs dominate row and plant-level analyses, whereas Sentinel-2 has become the main tool for multitemporal regional monitoring. DL models, such as CNNs and Vision Transformers (ViTs), tend to deliver superior performance in canopy segmentation and parcel classification. The assessment identified model validation as the weakest methodological domain across studies. The main limitations lie in weak spatiotemporal portability of models and high computational costs, aggravated by reliance on large volumes of annotated data. Promising directions include multisensory fusion (UAV + satellite) and the integration of 3D information into DL pipelines, which increase robustness and operational applicability. These advances are enabling high-value, specialised objectives such as mapping in complex terrain, detecting abandoned vineyards, and identifying missing plants.
Why it matches plant phenotyping methodsブドウ園・列・植物レベルの自動識別とモニタリングに用いるリモートセンシング手法を体系的にレビューし、センサー、3D再構成、深層学習、検証性、可搬性を評価しており、植物状態・構造の取得手法が中心である。
abstractThis systematic review synthesises and appraises the literature on automatic vineyard identification using remote sensing (RS), from classical techniques to artificial intelligence (AI), describing the state of the art, patterns, challenges, and gaps.
This study develops a voxel-based leaf area estimation framework and validates it using a three-year multi-temporal dataset (2022–2024) of pergola-trained grapevines. The workflow integrates 2D image analysis, ExGR-based leaf segmentation, and 3D reconstruction using Structure-from-Motion (SfM). Multi-angle canopy images were collected repeatedly during the growing seasons, and destructive leaf sampling was conducted to quantify true leaf area across multiple vines and years. After removing non-leaf structures with ExGR filtering, the point clouds were voxelized at a 1 cm3 resolution to derive structural occupancy metrics. Voxel-based leaf area showed strong within-vine correlations with destructively measured values (R2 = 0.77–0.95), while cross-vine variability was influenced by canopy complexity, illumination, and point-cloud density. In contrast, optical LAI tools (DHP and LAI–2000) exhibited negligible correspondence with true leaf area due to multilayer occlusion and lateral light contamination typical of pergola systems. This expanded, multi-year analysis demonstrates that voxel occupancy provides a robust and scalable indicator of canopy structural density and leaf area, offering a practical foundation for remote-sensing-based phenotyping, yield estimation, and data-driven management in perennial fruit crops.
Why it matches plant phenotyping methodsブドウ樹の葉面積・樹冠構造を推定する画像解析、SfM、ボクセル化ワークフローを開発し、破壊測定および既存LAI手法と比較検証しており、植物表現型取得法が研究の中心である。
abstractThis study develops a voxel-based leaf area estimation framework and validates it using a three-year multi-temporal dataset (2022–2024) of pergola-trained grapevines.
The transition toward Agriculture 5.0 requires intelligent and autonomous monitoring systems capable of providing early, accurate, and scalable crop health assessment. This study presents the design and field evaluation of an artificial intelligence (AI)-based unmanned aerial vehicle (UAV) system for the detection of Botrytis cinerea in vineyards using multispectral imagery and deep learning. The proposed system integrates calibrated multispectral data with vegetation indices and a YOLOv8 object detection model to enable automated, geolocated disease detection. Experimental results obtained under real vineyard conditions show that training the model using the Chlorophyll Absorption Ratio Index (CARI) significantly improves detection performance compared to RGB imagery, achieving a precision of 92.6%, a recall of 89.6%, an F1-score of 91.1%, and a mean Average Precision (mAP@50) of 93.9%. In contrast, the RGB-based configuration yielded an F1-score of 68.1% and an mAP@50 of 68.5%. The system achieved an average inference time below 50 ms per image, supporting near real-time UAV operation. These results demonstrate that physiologically informed spectral feature selection substantially enhances early Botrytis cinerea detection and confirm the suitability of the proposed UAV-AI framework for precision viticulture within the Agriculture 5.0 paradigm.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像とYOLOモデルにより、ブドウ樹の病害状態を直接検出する手法を設計・実地評価しており、植物フェノタイピング手法が中心である。
abstractThis study presents the design and field evaluation of an artificial intelligence (AI)-based unmanned aerial vehicle (UAV) system for the detection of Botrytis cinerea in vineyards using multispectral imagery and deep learning.
Abstract In perennial crops, inner wood degradation by pathogens often escapes detection until irreversible damage has occurred. Grapevine trunk disease (GTD) is a well-known example in viticulture that alters plants from within, years before foliar symptoms arise, making early assessment difficult. To overcome this limitation, we present a novel non-destructive 3D + t pipeline for Magnetic Resonance Imaging (MRI) spatial quantification and monitoring of early internal tissue degradation resulting from fungal colonization. This pipeline integrates (i) anatomical alignment and rigid time-series registration of volumetric MRI scans, (ii) a generalized cylindrical coordinate transformation for cross-sectional trunk anatomy normalization, (iii) supervised classification to segment water-depleted (diseased/non-functional) regions, and (iv) population-level statistical analyses including construction of population mean images, probabilistic atlases of lesions, and 3D lesion descriptors. Applied to multiple Vitis vinifera cultivars inoculated with a fungal trunk pathogen, our approach enables time-lapse comparisons between cultivar and treatment in vivo. The results reveal consistent early degradation signals across individuals and cultivar-dependent lesion differences. By combining high-resolution MRI with advanced image processing and statistical atlas tools, this method provides a new paradigm for 3D plant phenotyping of internal disease progression. This methodological innovation allows non-invasive quantification of disease development and comparative assessment of host responses in woody plants, demonstrating its potential to advance understanding and management of GTDs.
Why it matches plant phenotyping methodsMRI画像と画像処理・統計アトラスを統合し、ブドウ樹内部の病変・組織劣化を3Dで定量化する植物フェノタイピング手法の開発が中心である。
abstractwe present a novel non-destructive 3D + t pipeline for Magnetic Resonance Imaging (MRI) spatial quantification and monitoring of early internal tissue degradation resulting from fungal colonization.
Reproduction assets foundThe paper's raw/processed MRI datasets are only available from the corresponding author upon reasonable request, but the authors' processing pipeline (scripts and parameters to reproduce processed outputs from raw data) is publicly deposited on Zenodo with an explicit DOI.Code · publiceer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
available under a CC-BY 4.0 International license.
1 reasonable request. The processing pipeline (including scripts and parameters required to reproduce the
2 processed outputs from the raw data) is available at https://doi.org/10.5281/zenodo.17944369.
3
Plant Phenomics Page 26 of 29Open asset ↗zenodo · 10.5281/zenodo.17944369pdf-layout-page:26 lines:1-14Plant phenotyping relevance match · UnverifiedOpenAlex · checked 15 Sept 2026
In recent years, the development of automated methods for phenotypic assessments of observable plant traits is gaining interest because they can provide advantages over standard ones. This study presents a novel methodology to estimate the canopy volume of grapevine plants during their growth cycle using a low-cost red, green, blue and depth (RGB-D) sensor mounted on a ground-based platform. The aim is to address the need for cost-effective and easy-to-use systems that can operate under variable environmental conditions without requiring highly specific data acquisition constraints and advanced technical skills. Using a Microsoft Azure Kinect RGB-D camera, detailed 3-D images of the plants are captured. The iterative closest point (ICP) algorithm is then applied to reconstruct a full view of the plants useful to estimate the canopy volumes. The effectiveness of this approach is validated by comparing the volumetric estimates with the leaf area index (LAI) measures obtained with traditional agronomic techniques during significant phenological periods of the plants’ lifecycle. The results demonstrate a correlation between the two approaches, indicating the reliability of the proposed method and highlighting advantages in terms of precision. These outcomes demonstrate the potential of automated vineyard monitoring systems, providing reliable data to assess plant growth conditions.
Why it matches plant phenotyping methodsRGB-D画像とICP再構成によりブドウ樹の樹冠体積を推定する手法を開発・検証しており、植物形質の取得が研究の中心です。
abstractThis study presents a novel methodology to estimate the canopy volume of grapevine plants during their growth cycle using a low-cost red, green, blue and depth (RGB-D) sensor mounted on a ground-based platform.
Predicting the key phenological stages in grapevine in response to increased temperatures due to climate change is essential to assess the potential risk of the growth cycle shifting to less suitable periods, such as frost periods or periods with unfavorable ripening conditions. This understanding is crucial for selecting grape varieties that are well-suited to specific production areas under current and future climatic conditions. Temperature and phenology at the local scale can be highly variable depending on the environmental context. Most studies have attempted to represent phenology at large scale, with few studies considering the local scale, critical for winegrower’s adaptation strategies. This research aimed to explore phenology modeling at the local scale by developing temperature-based models for budburst, flowering, veraison, and sugar concentration of 200 g/L for Vitis vinifera L. cv. Merlot in Saint-Émilion and surrounding appellations (Bordeaux, France). Fixed start dates versus phenophase based models, as well as simple versus more complex models, were tested and compared to current models in the literature. Selected phenological models were also compared under different warmer temperature scenarios. High-performance models were parameterized for all stages, showing few differences between approaches. However, the easy-to-use linear growing degree day models were slightly less accurate than the more complex curvilinear models, which are considered closer to plant development. Phenophase models performed better in predicting phenology with external validation data. The developed models outperformed existing models in literature, especially for the budburst stage. Little differences were observed in phenology projections among models with a 1 °C increase. In contrast, a 4 °C increase showed significant differences between models, suggesting a need for deeper understanding of plant development under extreme temperatures. In addition to methodological findings on model selection, these results could help professionals optimize vineyard management and plant material adaptations to terroir and climate change.
Why it matches plant phenotyping methodsブドウの複数の生育フェノロジー段階を予測する温度ベースモデルを開発・比較し、外部検証および既存モデルとの性能比較を行っており、植物形質取得・予測手法が研究の中心である。
abstractThis research aimed to explore phenology modeling at the local scale by developing temperature-based models for budburst, flowering, veraison, and sugar concentration of 200 g/L for Vitis vinifera L. cv. Merlot
This study explores the conceptual framework and evaluation methods of grape berry uniformity, elucidating its multidimensional nature arising from the coordinated contributions of berry size, shape, and cluster structure. Quantitative evaluation approaches based on the coefficient of variation, composite multi-trait indices, and high-throughput phenotyping technologies are systematically summarized. On this basis, key factors influencing berry uniformity are further analyzed, including genetic background, pollination and fertilization processes, berry developmental dynamics, plant growth regulator treatments, and water-nutrient environmental conditions. Integrating breeding strategies with production practices, a framework for improving berry uniformity is proposed, centered on “multi-trait selection, marker-assisted selection, and cultivation regulation.” Meanwhile, with the advancement of machine vision, high-throughput phenotyping, and multi-source data integration technologies, the evaluation of berry uniformity is shifting toward automation, precision, and intelligence. However, challenges remain in the standardization of evaluation systems, elucidation of molecular mechanisms, and integration of multi-source data. Future research directions toward data-driven precision improvement are discussed. This study aims to provide theoretical foundations and technical support for enhancing the quality and standardized production of table grapes.
Why it matches plant phenotyping methodsブドウ果実の均一性を対象に、評価指標、高スループットフェノタイピング、機械ビジョンによる自動評価を体系的に扱うレビューであり、フェノタイピング手法が中心です。
abstractQuantitative evaluation approaches based on the coefficient of variation, composite multi-trait indices, and high-throughput phenotyping technologies are systematically summarized.
Today, the agricultural sector faces significant challenges due to population growth and limited resources. Enhancing productivity and minimizing losses is of great importance for the sustainability of agriculture. Therefore, leveraging technological advancements plays a critical role, particularly in the development of sustainable farming practices. Among these advancements, artificial intelligence (AI) stands out with its potential to contribute significantly to agricultural production. The primary objective of this study is to provide farmers with fast and accurate information regarding plant health, thereby preventing the spread of diseases and optimizing agricultural output. In line with this goal, AI-based image processing techniques were employed. Specifically, this study focuses on detecting grapevine leaf diseases namely powdery mildew ($Erysiphe$ $necator$), downy mildew ($Plasmopara$ $viticola$), and grapevine rust mite ($Eriophyes$ $vitis$) using AI. Disease detection was carried out using leaf images, which were then used for classification. A hybrid dataset was constructed using a combination of publicly available images and manually collected samples captured via smartphone cameras in vineyards, fields, and gardens. This diverse and balanced dataset was used to train several CNN-based transfer learning models, including AlexNet, DarkNet53, Inception-ResNet-V2, Inception-V3, MobileNet-V3, ResNet50, ResNet101, VGG16, and VGG19 architectures. Among these, Inception-ResNet-V2 achieved the best performance with an accuracy of 97.45%, a training loss of 8.19%, a test accuracy of 93.00%, and a test loss of 20.60%. These results demonstrate that the model performs well in detecting diseases from grapevine leaves during both training and testing phases.
Why it matches plant phenotyping methodsブドウ葉の画像から病害状態を推定する画像解析・転移学習手法が研究の中心であり、データセット構築と複数モデルの性能評価も行っているため含める。
abstractAI-based image processing techniques were employed.
Abstract. The demand for food production is increasing rapidly with a surge in the population. To cope with this increasing food demand, precise agricultural management is essential. The existing techniques involve extensive field surveys for agricultural land discrimination. To minimize the man-hour efforts and time required by these techniques, automated techniques for precise crop type mapping and monitoring have been used. These techniques utilize satellite imagery and advanced machine learning techniques for crop type mapping and monitoring. However, the performance of such techniques is affected by factors such as fragmented land parcels, seasonal variability, and inconsistent field-level observations. To overcome these issues, this study attempts to classify grape and non-grape crops and monitor their phenological stages in the study area in Pune district, India, using Sentinel-2 satellite imagery and deep learning (DL) segmentation techniques: U-Net and DeepLabV3. Further, Sentinel- 1C SAR imagery (VV and VH polarization) for the years 2016 to 2024 was utilized to train and evaluate a long short-term memory network (LSTM) model with an aim to analyze the temporal behavior of the grape crop from pruning to harvesting stage with emphasis on growth stages like leaf set, fruit set, and ripening. The experimental results demonstrate that U-Net outperforms DeepLabV3 (F1-score: 0.96; mAP: 0.95) in grape crop classification. The LSTM model showed performance (F1-score 0.82) for phenological stage identification. This study can help agricultural stakeholders in effective and large-scale crop discrimination with minimum human intervention. It has the potential to reveal grape distribution and development stages in a faster time.
Why it matches plant phenotyping methods衛星画像と深層学習を用いてブドウ作物の分類および生育（フェノロジー）段階を推定し、モデル性能も評価しているため、植物状態の取得・抽出手法が中心である。
abstractthis study attempts to classify grape and non-grape crops and monitor their phenological stages in the study area in Pune district, India, using Sentinel-2 satellite imagery and deep learning (DL) segmentation techniques: U-Net and DeepLabV3.
Grapevine water relations are increasingly influenced by drought under climate change, with significant implications for yield, fruit composition and wine quality. Stable isotopes of hydrogen, oxygen, carbon and nitrogen (δ 2 H, δ 18 O, δ 13 C and δ 15 N) provide sensitive tracers of plant water sources and physiological responses to stress. Here, we combined dual water isotopes (δ 2 H, δ 18 O), carbon and nitrogen isotopes (δ 13 C, δ 15 N), and high-resolution micrometeorological/soil observations to diagnose drought dynamics in Vitis vinifera cv. Sauvignon blanc (Orlești, Romania; 2023-2024). Dual-isotope relationships delineated progressive evaporative enrichment along the soil-plant-atmosphere continuum, with slopes LMWL ≈ 6.41 > stem ≈ 5.0 > leaf ≈ 2.2, consistent with kinetic fractionation during transpiration (leaf) superimposed on source-water signals (stem). Weekly leaf δ 18 O covaried strongly with relative humidity (RH; r = -0.69) and evapotranspiration (ET; r = +0.56), confirming atmospheric control of short-term enrichment, while stem isotopes showed buffered responses to soil water. We integrated Δ 18 O (leaf-stem), RH, ET, and soil matric potential at 60 cm (Soil 60 ) into an Isotopic Drought Index (IDI), which captured the onset, intensity, and persistence of the July-August 2024 drought (IDI 0-100 > 90; RH 40 mm wk -1 , Soil 60 > 100 cb). Carbon and nitrogen isotopes provided complementary, integrative diagnostics: δ 13 C increased (less negative) with drought (r = -0.52 with RH; +0.49 with IDI), reflecting higher intrinsic water-use efficiency, whereas δ 15 N rose with soil dryness and IDI (leaf: r ≈ +0.48 with Soil 60 ; +0.42 with IDI), indicating constraints on N acquisition and enhanced internal remobilization. Together, multi-isotope and environmental data yield a mechanistic, field-validated framework linking atmospheric demand and edaphic limitation to vine physiological and biogeochemical responses and demonstrate the operational value of an isotope-informed drought index for precision viticulture.
Why it matches plant phenotyping methods複数同位体と環境データからブドウの水分状態・干ばつ応答を推定するIsotopic Drought Indexを構築し、圃場で検証した研究であり、植物の生理状態取得手法が中心である。
abstractWe integrated Δ 18 O (leaf-stem), RH, ET, and soil matric potential at 60 cm (Soil 60 ) into an Isotopic Drought Index (IDI), which captured the onset, intensity, and persistence of the July-August 2024 drought
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Supplement · publicTable S1: Isotopic data of leaf and stem of Vitis vinifera cv. Sauvignon Blanc blanc from Orlești-Vâlcea (Romania), during 2023-2024 vintage; Table S2: Meteorological and soil measurements (Romania), during the sampling campaign (Orlești – Vâlcea, Romania; 2023-2024 vintage)Open asset ↗lines:149-204Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Abstract This study evaluated a non-invasive, integrated monitoring approach to characterize the soil-plant-atmosphere continuum (SPAC) in a commercial vineyard of Pignoletto (PG) and Trebbiano Romagnolo (TR). The approach is based on a cosmic-ray neutron sensor (CRNS) to continuously monitor soil water content (SWC), which was normalized into extractable soil water (ESW) to represent plant-available water. Moreover, vapor pressure deficit (VPD) was calculated based on weather data to characterize the atmospheric demand. Finally, remotely sensed NDVI data were used to detect canopy development and vine physiological responses. Over two growing seasons, measurements of midday stem water potential (Ψ stem ) and berry composition complemented the monitoring activities. In 2023, ripening was largely buffered from atmospheric demand, with Ψ stem values between − 0.66 and − 1.06 MPa, reflecting SWC as a non-limiting factor and uniform ripening. Conversely, the 2024 season showed more negative Ψ stem (-0.95 to -1.12 MPa) and an accelerated ripening process, particularly in TR. Principal Component Analysis (PCA) explained 65% of the variance in 2023 and 81.5% in 2024, revealing that environmental drivers (ESW, VPD) became more tightly linked to physiological and grape composition traits (Ψ stem , TSS, TA). Overall, the results showed the capability of the integrated approach to capture the main interactions within the SPAC offering a non-invasive and scalable tool for supporting precision and sustainability in Mediterranean viticulture.
Why it matches plant phenotyping methods土壌水分・大気需要・リモートセンシングNDVIを統合し、ブドウ樹の樹冠発達と生理応答を非侵襲的・スケーラブルにモニタリングする手法が研究の中心であるため。
abstractThe approach is based on a cosmic-ray neutron sensor (CRNS) to continuously monitor soil water content (SWC), which was normalized into extractable soil water (ESW) to represent plant-available water.
Owing to changing climatic and environmental conditions, plant diseases are becoming increasingly prevalent, posing a serious threat to global agriculture. Timely and accurate diagnosis remains challenging, especially where scouting still relies on manual inspection. We propose B2-GraftingNet, a deep learning framework for automated detection of grape leaf diseases. B2-GraftingNet is a streamlined variant of our earlier B4-GraftingNet, retaining its strengths while simplifying blocks for faster inference and deployment. The architecture combines a VGG16 backbone with Inception-style blocks inside a custom CNN to extract robust, multi-scale features based on color, size, and shape. To reduce redundancy and improve generalization, Binary Particle Swarm Optimization (BPSO) selects informative features prior to classification. We evaluate Support Vector Machines (SVM) and k-Nearest Neighbors (KNN); a cubic SVM attains 99.56% peak accuracy on the public Kaggle grape-leaf dataset. For context, we also benchmarked standard pretrained CNNs on the same data, observing validation accuracies of 34.04% (VGG16), 34.04% (VGG19), 97.95% (Xception), 94.91% (Darknet), and 98.44% (ResNet-50); B2-GraftingNet matches or exceeds these while remaining lighter and faster to train and deploy. To enhance transparency and actionability, we pair Grad-CAM, LIME, and occlusion-sensitivity visualizations with a local gpt-oss:20b assistant (served via Ollama) that converts evidence into plain, grower-focused guidance and supports interactive chat validated by horticulturists. Results are further checked against expert-annotated ground-truth labels, confirming high accuracy and computational efficiency. Overall, B2-GraftingNet offers a reliable, interpretable, and scalable solution for early grape-leaf disease detection. The complete setup (code, model, web platform, configuration, and assets) is available on Zenodo: https://doi.org/10.5281/zenodo.17353656.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像から検出する深層学習手法を開発し、複数モデル・専門家アノテーションと比較検証しているため、植物フェノタイピング手法が中心である。
abstractWe propose B2-GraftingNet, a deep learning framework for automated detection of grape leaf diseases.
Reproduction assets foundThe paper's Data Availability Statement points to a public Zenodo deposit containing the grape leaf images used for disease classification, which is a paper-specific, publicly actionable asset. The underlying Kaggle source dataset is also cited, but the Zenodo record is the authors' own public deposit matching an exactDataset · publicICCK Journal of Image Analysis and Processing
reproducible runs, API examples for mobile image https://zenodo.org/records/18401218.
uploads and programmatic retrieval of classifications
and explainability overlays, as well as additional Funding
figures and code listings that mirror the production
This work was supported without any funding.
repository.
Conflicts of Interest
4 Conclusion
Syed Adil Hussain Shah is affiliated with the
In this study, we introdOpen asset ↗Zenodo · 18401218pdf-layout-page:16 lines:1-68Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 6 Sept 2026
Grape cluster compactness is a key trait that influences fruit quality, yield, and disease susceptibility. Understanding the genetic basis of this trait is essential for optimizing vineyard management and improving grapevine cultivars. In this study, we performed quantitative trait locus (QTL) mapping to identify genomic regions associated with cluster architecture and yield components in a bi-parental population derived from Vitis vinifera cv. Riesling × Cabernet Sauvignon. A total of 138 full-sibling progeny were evaluated over two growing seasons at Oakville, Napa Valley, California. Traditional yield-related traits were measured, including cluster number, total cluster weight, and average cluster weight. Additionally, an image-based phenotyping pipeline leveraging the foundation model Segment Anything Model (SAM) was employed to segment individual berries, measure their size and shape, and compute cluster compactness with minimal manual intervention. Trait correlations revealed that compact clusters tended to have a higher berry count but smaller berry size, highlighting the role of compactness in modulating cluster structure. Heritability estimates varied across traits, with berry dimensions and compactness displaying moderate to high heritability, indicating strong genetic control. Two parental linkage maps were constructed using a pseudo-test cross strategy. QTL mapping identified multiple loci associated with cluster architecture and yield components, with several stable QTLs detected across both years, with marker effects ranging from 7.6% to 22.1%. Notably, a QTL for cluster compactness was found in both seasons on chromosome 1 in Cabernet Sauvignon. Other stable QTLs were associated with berry size (chromosomes 6 and 17) and berry count (chromosome 5 in Cabernet Sauvignon and chromosome 7 in Riesling). Additional QTLs were detected in a single year, reflecting the influence of environmental variation. Our findings provide valuable insights into the application of foundation models requiring no prior training and minimal intervention for high-quality segmentation and enhance our understanding of the genetic architecture of cluster compactness and yield traits. The genomic regions identified in this study offer promising targets for breeding programs aimed at improving grape quality and disease resistance.
Why it matches plant phenotyping methodsSAMを用いた画像解析パイプラインで個々の果粒を分割し、サイズ・形状と房のコンパクトネスを算出する方法が、研究の主要な技術的要素として明示されている。
abstractAdditionally, an image-based phenotyping pipeline leveraging the foundation model Segment Anything Model (SAM) was employed to segment individual berries, measure their size and shape, and compute cluster compactness with minimal manual intervention.
Accurate prediction of photosynthetic parameters is pivotal for precision viticulture, as it enables non-invasive monitoring of plant physiological status and informed management decisions. In this study, spectral reflectance data were used to predict key photosynthetic parameters such as assimilation rate (A), effective photosystem II (PSII) quantum yield (ΦPSII), and electron transport rate (ETR), as well as stem and leaf water potential (Ψstem and Ψleaf), in Vitis vinifera (cv. Müller-Thurgau) grown in an experimental vineyard in Lower Franconia (Germany). Measurements were obtained on 25 July, 7 August, and 12 August 2024 using a LI-COR LI-6800 system and a PSR+ hyperspectral spectroradiometer. Various machine learning models (SVR, Lasso, ElasticNet, Ridge, PLSR, a simple ANN, and Random Forest) were evaluated, both as standalone predictors and as base learners in a stacking ensemble regressor with a Random Forest meta-learner. First derivative reflectance (FDR) preprocessing enhanced predictive performance, particularly for ΦPSII and ETR, with the ensemble approach achieving R2 values up to 0.92 for ΦPSII and 0.85 for A at 1 nm resolution. At coarser spectral resolutions, predictive accuracy declined, though FDR preprocessing provided some mitigation of the performance loss. Diurnal patterns revealed that morning to mid-morning measurements, particularly between 9:00 and 11:00, captured peak photosynthetic activity, making them optimal for assessing vine vigor, while midday water potential declines indicated favorable timing for irrigation scheduling. These findings demonstrate the potential of integrating hyperspectral data with ensemble machine learning and FDR preprocessing for accurate, scalable, and high-throughput monitoring of grapevine physiology, supporting real-time vineyard management and the use of cost-effective sensors under diverse environmental conditions.
Why it matches plant phenotyping methodsハイパースペクトル測定と機械学習によるブドウの光合成・水ポテンシャル推定が研究の中心であり、複数モデルの性能評価と前処理比較も実施しているため。
abstractspectral reflectance data were used to predict key photosynthetic parameters such as assimilation rate (A), effective photosystem II (PSII) quantum yield (ΦPSII), and electron transport rate (ETR), as well as stem and leaf water potential (Ψstem and Ψleaf)
The accurate estimation of grapevine biophysical parameters is important for decision support in precision viticulture. This study addresses the use of unmanned aerial vehicle (UAV) multispectral data and machine learning (ML) techniques to estimate leaf area index (LAI), pruning wood biomass, and yield, across mixed-variety vineyards in the Douro Region of Portugal. Data were collected at three phenological stages, from veraison to maturation and two modeling approaches were tested: one using only spectral features, and another combining spectral and geometric features derived from photogrammetric elevation data. Multiple linear regression (MLR) and five ML algorithms were applied, with feature selection performed using both forward and backward selection procedures. Logarithmic transformations were used to mitigate data skewness. Overall, ML algorithms provided better predictive performance than MLR, particularly when geometric features were included. At harvest-ready, Random Forest achieved the highest accuracy for LAI (R2 = 0.83) and yield (R2 = 0.75), while MLR produced the most accurate estimates for pruning wood biomass (R2 = 0.83). Among geometric variables, canopy area was the most informative. For spectral data, the Modified Soil-Adjusted Vegetation Index (MSAVI) and the Soil-Adjusted Vegetation Index (SAVI) were the most relevant. The models performed well across grapevine varieties, indicating that UAV-based monitoring can serve as a practical, non-invasive, and scalable approach for vineyard management in heterogeneous vineyards.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像と機械学習により、ブドウのLAI、バイオマス、収量を推定する手法を開発・比較しており、表現型取得と推定が研究の中心である。
abstractThis study addresses the use of unmanned aerial vehicle (UAV) multispectral data and machine learning (ML) techniques to estimate leaf area index (LAI), pruning wood biomass, and yield
Accurate estimation of canopy geometric and structural characteristics, such as leaf area (LA), is essential for improving resource efficiency in fruit tree crop management. LA is a key biophysical parameter, influencing physiological processes like carbon fixation, evapotranspiration, and light interception, as well as fruit quality and yield. However, its measurement is complex due to the substantial number of leaves and the three-dimensional nature of tree canopies.An alternative approach, the Projected Tree Row Surface (PTRS), has shown a strong correlation with LA and has been recognized by the scientific community. Despite its robustness, the original PTRS method requires time-consuming manual data collection, which limits its practical application in the field.This study introduces a novel automated methodology for calculating the PTRS, validated using high-resolution ground-truth data providing LA values at 0.1-m intervals along the tree rows. When evaluated on almond, pear, and apple trees as well as vineyards, the method achieved remarkably high correlations between PTRS and LA, with coefficients up to r = 0.97 and r = 0.99 at optimal resolutions (0.1 –0.2 m PTRS per 1 m row section). These results demonstrate that the approach delivers consistent and reliable measurements of LA under diverse field conditions, enabling real-time, high-resolution assessment of tree-row canopies.The automated PTRSₙ approach enables fast and efficient LA estimation and can be adapted to any point cloud dataset. It supports flexible resolution to balance accuracy and processing time and can be applied to full rows, individual trees, or canopy segments. This methodology represents a step forward in automating LA assessment and supports the development of real-time applications in precision agriculture.
Why it matches plant phenotyping methods果樹・ブドウ樹冠の葉面積を推定するLiDARベースの自動PTRS手法を開発し、実測LAで検証しており、植物形質取得法が研究の中心である。
abstractThis study introduces a novel automated methodology for calculating the PTRS, validated using high-resolution ground-truth data providing LA values at 0.1-m intervals along the tree rows.
Sustainable vineyard management requires precise and efficient application of plant protection products to minimise environmental impact while ensuring plant health. This study presents a variable-rate spraying system that integrates an RGB-D camera with a GPU-equipped edge computing platform to enable accurate, real-time adjustment of spray flow rates in vineyards. A tensor-based representation of RGB-D data is employed to accelerate the entire processing pipeline. Based on this structure, a fast approximate meshing method is applied to rapidly generate 3D meshes from point clouds. To incorporate semantic information from RGB images, an instance segmentation model is used to detect grapevine canopies and trellis posts. The resulting canopy masks are used to isolate the canopy meshes, while the trellis posts serve as reference planes for canopy volume estimation via mesh projection. Based on the computed volume, pulse-width modulation signals are generated to dynamically control spray flow rates. Field experiments were conducted to evaluate the system’s effectiveness and real-time performance. The results demonstrated that the estimated canopy volume is a reliable indicator for regulating application rates. Compared to uniform-rate spraying, the proposed system reduced plant protection product consumption by 57.4% while ensuring adequate droplet coverage. Additionally, the system demonstrated satisfactory real-time performance even on entry-level hardware. Overall, the proposed variable-rate spraying system offers an accurate, real-time, and cost-effective solution for precision viticulture, highlighting its potential for commercial deployment in sustainable vineyard management.
Why it matches plant phenotyping methodsRGB-D画像からブドウ樹冠を分離し、3Dメッシュ投影で樹冠体積という植物形質を推定する技術が中心であり、リアルタイム性能と散布制御への有効性も評価している。
abstractan instance segmentation model is used to detect grapevine canopies and trellis posts
BACKGROUND: Early detection of grapevine downy mildew (GDM) using unmanned aerial vehicle (UAV) imagery remains highly challenging due to limited onboard computing power, natural light overexposure, limited imaging resolution that obscures subtle early lesions, pronounced symptom differences between leaf surfaces, and severe canopy occlusion. MODELS: To address these challenges, a lightweight GDM detection network (GDM-Net) for UAV-based applications is proposed. A CLAHE–Unsharp Masking–Gamma Correction (CUG) image enhancement framework is proposed to achieve synergistic multi-dimensional feature enhancement by expanding the regional dynamic range, emphasizing high-frequency lesion details, and redistributing image brightness. This effectively alleviates the issue of weak and difficult-to-recognize early lesion features. To enhance feature extraction with high efficiency and accuracy, the YOLOv11n backbone integrates an Adaptive Instance-aware Feature Interaction (AIFI) mechanism inspired by RT-DETR, achieving sparse and efficient feature interaction with minimal computational cost. Furthermore, an Adaptive Dual-Path Fusion Attention (ADFA) module is designed in the neck architecture to handle dense canopy occlusion. It performs multiscale convolutional feature extraction, cross-attention interaction, and adaptive dynamic fusion between feature pathways, thereby improving the discriminative capacity under occlusion and clustering conditions. RESULTS: Experimental results demonstrate that the CUG image enhancement framework significantly improves early lesion visibility and feature separability under complex illumination conditions. GDM-Net achieves accuracy and lightweight ability, demonstrating superior onboard deployment feasibility compared with other mainstream models. It achieves a 5.6% improvement in mAP@50 and a 6.5% improvement in mAP@50:95, while requiring only 7.5 GFLOPs of computational cost, addressing a key bottleneck in UAV-based GDM detection. CONCLUSIONS: GDM-Net provides an efficient and practical solution for UAV-based crop disease monitoring by integrating image information enhancement and improved lightweight detection. Its balance of computational efficiency and detection precision highlights its potential for broader applications in precision agriculture and intelligent crop health management systems.
Why it matches plant phenotyping methodsUAV画像からブドウ葉の病斑・べと病状態を検出する画像強調および軽量深層学習手法を開発し、精度と計算効率を評価しており、植物病害表現型の取得手法が中心である。
abstracta lightweight GDM detection network (GDM-Net) for UAV-based applications is proposed
Analysis of leaf hemispherical radiative properties for retrieval of its biochemical and mineral nutrients could lead to a powerful monitoring approach for precise farm management. This study explores the potential of leaf spectral modeling techniques for estimation of key biochemical and nutritional traits in grapevine leaves. Hyperspectral data spanning the 400-2500 nm range were collected from around 1000 leaf grapevine leaf samples across three growing seasons. Certain traits, including leaf structural parameter (Nstruct), anthocyanins, carotenoids, and chlorophyll, were imputed using the PROSPECT-PRO radiative transfer model in the inverse mode to enrich the dataset. An imputation model was developed to address missing labels for part of the dataset, employing a Convolutional Neural Network (CNN) with 23 principal components derived from the spectral data as inputs. This model enabled the completion of the dataset by predicting missing trait values, providing a comprehensive foundation for subsequent modeling efforts. For the primary trait prediction models, the spectral data were then reduced from 2101 bands to 204 bands through band merging based on pairwise correlations. Two predictive modeling approaches were evaluated: a single-trait model, where each trait is predicted independently, and a multi-trait model, where all traits are predicted simultaneously. Both models employed a hybrid of CNN and Long Short-Term Memory (LSTM) networks designed to capture spatial and sequential patterns in spectral data. The single-trait model utilized CNN-LSTM architecture with a single output node, requiring independent training for each trait. In contrast, the multi-trait model employed the same architecture but featured 16 output nodes, enabling the simultaneous prediction of all traits. A weighting strategy was implemented to balance the influence of fully measured and imputed samples during training, ensuring reliable predictions. The multi-trait model demonstrated superior predictive performance across most traits, achieving a higher coefficient of determination (R 2 ) and RPD (Residual Predictive Deviation), and lower normalized root mean squared error (NRMSE) values than the single-trait models. Some traits, such as nitrogen, phosphorus, Nstruct, and manganese benefited significantly in the multi-trait model with R 2 values of 0.42, 0.81, 0.90, and 0.62, respectively, compared to 0.26, 0.64, 0.25, and 0.30 in single-trait models. The results highlight the advantages of multi-trait modeling in leveraging shared spectral information and inter-trait dependencies, offering an efficient and accurate approach to predicting grapevine traits.
Why it matches plant phenotyping methodsブドウ葉のハイパースペクトルデータから生化学・栄養形質を推定するスペクトル計測およびCNN-LSTM解析手法が研究の中心であり、単一形質モデルと多形質モデルの性能比較・検証も行っている。
abstractThis study explores the potential of leaf spectral modeling techniques for estimation of key biochemical and nutritional traits in grapevine leaves.
Early identification of grapevine diseases is critical for reducing yield losses and ensuring sustainable viticulture. CNNs trained on benchmark datasets such as PlantVillage often achieve near-perfect accuracy, yet this performance fails to translate to real-world field conditions where lighting, backgrounds, and lesion appearance vary widely. To address challenges of data scarcity and imbalance, this study introduces VitiForge, a novel procedural synthetic imagery pipeline for generating realistic synthetic grape leaf textures representing healthy, Black Rot, Esca, and Leaf Blight conditions. VitiForge is systematically evaluated against GAN-based augmentation through a data ablation study on PlantVillage and FieldVitis, a curated field dataset, using MobileNetV2, InceptionV3, and ResNet50V2 classifiers. Results show that VitiForge significantly improves performance in low-data regimes, enabling model training even without real samples, whereas GAN augmentation proves more effective once sufficient real data is available. On field imagery, VitiForge often matched or surpassed GAN-based methods, particularly when paired with MobileNetV2. These findings highlight the complementary roles of procedural and GAN-based synthetic data: VitiForge offers flexibility and scalability under cross-domain and data-scarce conditions, while GANs enhance realism and variability when ample data exists. Together, they support the development of robust and generalizable models for automated grape disease detection in precision agriculture.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像から識別するための合成画像生成パイプラインを開発し、GAN augmentationと比較評価しているため、植物病害状態のフェノタイピング手法が中心である。
abstractthis study introduces VitiForge, a novel procedural synthetic imagery pipeline for generating realistic synthetic grape leaf textures representing healthy, Black Rot, Esca, and Leaf Blight conditions.
Reproduction assets foundThe paper introduces FieldVitis, a curated field grapevine leaf image dataset assembled from public sources, explicitly deposited on Zenodo with a DOI matching an allowed URL. No explicit public availability of the VitiForge pipeline code or trained models is stated in the supplied blocks.Dataset · publicThe introduction of FieldVitis, a curated dataset of grapevine leaves collected from multiple public sources to reflect the real-world variability of vineyard imagery, providing a valuable benchmark for evaluating model generalization under realistic field conditions. It is available in Zenodo at https://doi.org/10.5281/zenodo.17307846 .Open asset ↗Zenodo · 10.5281/zenodo.17307846lines:310-320Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Abstract Efficient irrigation management is fundamental to sustainable crop production, particularly under increasing temperatures and limited water availability. In vineyards, water stress significantly influences grapevine development and productivity. Controlled water stress is intentionally applied in deficit-irrigated systems to regulate yield and enhance fruit quality. Therefore, vineyards must be routinely monitored to prevent excessive stress that could cause detrimental effects. In this study, we developed a machine-learning framework based on the eXtreme Gradient Boosting (XGB) machine-learning model to estimate grapevine leaf water potential (Y leaf ) using meteorological data and high-resolution imagery from small unmanned aerial systems (sUAS) over commercial vineyards of different varieties and in different climatic zones in California. The framework incorporates key meteorological and image-derived features, including maximum air temperature in the 24 hours prior to the flight, air temperature at the time of flight, the difference between these two temperatures, as well as canopy temperature derived from sUAS thermal imagery. These features were included to indirectly capture plant-water-weather interaction during the 24-hour period preceding data collection, enhancing the model’s practical applicability. The XGB model demonstrated robust performance, achieving an RMSE of 0.16 MPa, a bias of -0.06 MPa, and a correlation coefficient of 0.83 while minimizing computational cost. Model generalizability was further validated in an independent vineyard, demonstrating its potential for commercial application in precision irrigation and vineyard water management. Our research highlights the potential for broader applicability, particularly in addressing flash drought and promoting adaptive water resource management.
Why it matches plant phenotyping methodssUAS熱画像と気象データからブドウ葉の水ポテンシャルを推定する機械学習手法を開発し、独立圃場で妥当性を検証しており、植物表現型取得が中心である。
abstractwe developed a machine-learning framework based on the eXtreme Gradient Boosting (XGB) machine-learning model to estimate grapevine leaf water potential (Y leaf ) using meteorological data and high-resolution imagery from small unmanned aerial systems (sUAS)
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 6 Sept 2026
Abstract In agroecosystems, the variable expression of crop functional traits is expected to play a role in key processes, including plant nutrient cycling and water acquisition, that confer ecosystem resistance and/ or resilience to environmental change. The ability to estimate crop trait data is therefore critical to predict crop responses to environmental change, enabling more informed diagnosis of crop performance and on-farm management strategies. Yet, many traditional methods for quantifying plant traits are time-consuming and resource-intensive, limiting sample sizes and study durations. In response, high-throughput phenotyping— specifically reflectance spectroscopy— has emerged as a key element of plant trait research, capable of estimating plant traits more rapidly. However, little is known about whether or not reflectance spectroscopy can detect within-species variation in resource acquisition and plant-water traits. Using wine grapes ( V. vinifera subsp. vinifera ) as a focal crop, this study aimed to assess the ability of reflectance spectroscopy and the subsequent partial least squares regression modelling approach to quantify intraspecific variation in 12 functional traits across 12 different cultivars. Results showed significant differences in traits, especially in the photosynthetic and hydraulic traits, among closely related cultivars, falling along a resource-conservative to resource-acquisitive axis of variation. We also found that reflectance differentiated this fine-scale trait variation, specifically in leaf chemical and morphological traits, contributing to higher accuracy, and indicating that this HTP approach is viable for detailed trait estimation in diverse agroecosystems.
Why it matches plant phenotyping methods反射分光とPLS回帰によるブドウの複数機能形質推定を主目的とし、ハイスループット表現型解析手法の性能・実用性を評価しているため。
abstractthis study aimed to assess the ability of reflectance spectroscopy and the subsequent partial least squares regression modelling approach to quantify intraspecific variation in 12 functional traits across 12 different cultivars.
• A smartphone-based 3D imaging approach was developed for canopy and berry cluster volume estimation in vineyards. • Machine learning-based segmentation using Gradient Boosting for canopy and YOLO11 with Structure from Motion (SfM) for clusters enabled accurate feature extraction. • Point cloud processing was utilized for accurate surface reconstruction and volume estimation. • The method provides an affordable and accessible alternative to traditional high-cost sensors for precision viticulture applications. Accurate estimation of vine canopy and berry cluster volumes is essential for precision viticulture, as it supports better vineyard management, yield prediction, and resource allocation. Traditional methods, such as manual measurements or expensive sensor-based systems, are often inaccessible to small and mid-scale growers. This study explores the use of smartphone-based 3D imaging and advanced machine learning techniques as an affordable and accessible alternative for estimating wine grape canopy and berry cluster volumes. In this study, point cloud data was collected using an iPhone 14 Pro Max to capture the spatial structure of grape canopies and berry clusters. Two separate datasets were used to evaluate canopy and cluster volumes independently, ensuring comprehensive analysis and validation. Canopy volume estimation involved segmentation using the Gradient Boosting Classifier, followed by computation of 3D point volumes, achieving an RMSE of 0.23 m³ and 98% classification accuracy for canopy point clouds. Berry clusters were segmented using YOLO11, and 3D point clouds were reconstructed using Structure from Motion (SfM) to create watertight meshes. Cluster volumes validated by water‑displacement ground truth yielded an RMSE of 14.68 cm. These findings demonstrate the potential of smartphone-based solutions to support precision viticulture through accurate estimation of vine canopies and berry clusters, which is expected to enhance vineyard productivity and berry quality.
Why it matches plant phenotyping methodsスマートフォン3D画像、機械学習セグメンテーション、SfM、点群処理を用いてブドウ樹冠・果房体積を推定し、実測値で検証する手法開発が中心である。
abstractA smartphone-based 3D imaging approach was developed for canopy and berry cluster volume estimation in vineyards.
To enhance crop yield, detecting leaf diseases has become a crucial research focus. Deep learning and computer vision excel in digital image processing. Various techniques grounded in deep learning have been utilized for detecting plant leaf diseases; however, achieving high accuracy remains a challenge. Basic convolutional neural networks (CNNs) in deep learning struggle with issues such as the abnormal orientation of images, rotation, and others, resulting in subpar performance. CNNs also need extensive data covering a wide range of variations to deliver strong performance. CapsNet is an innovative deep-learning architecture designed to address the limitations of CNNs. It performs well without needing a vast amount of data in various variations. CapsNets have their limitations, such as the encoder network considering every element in the image and the crowding issue. Due to this, they perform well on simple image recognition tasks but struggle with more complex images. To address these challenges, we introduced a new CapsNet model known as CCFM-CapsNet. This model incorporates CLAHE to reduce image noise and CDH to extract crucial features. Also, max-pooling and dropout layers are incorporated in the original CapsNet model for identifying and classifying diseases in apples, bananas, grapes, corn, mangoes, pepper, potatoes, rice, tomato and also for classifying fashion-MNIST and CIFAR-10 datasets. The proposed CCFM-CapsNet demonstrates significantly high validation accuracies, achieving 99.53%, 95.24%, 99.75%, 97.40%, 99.13%, 100%, 99.77%, 100%, 98.54%, 93.48%, and 82.34% with corresponding parameters in millions(M) 4.68M, 4.68M, 4.68M, 4.68M, 4.79M, 4.63M, 4.66M, 4.68M, 4.84M, 2.39M, and 4.84M for the datasets aforementioned respectively, outperforming the traditional CapsNet and other advanced CapsNet models. Consequently, the CCFM-CapsNet model can be utilized effectively as a smart tool for identifying plant diseases and also in achieving Sustainable Development Goal 2 (Zero Hunger), which aims to end global hunger by the year 2030.
Why it matches plant phenotyping methods植物葉画像から病害状態を推定する深層学習モデルを提案・評価しており、画像取得・分類手法が研究の中心であるため。
abstractTo address these challenges, we introduced a new CapsNet model known as CCFM-CapsNet.
Accelerating grapevine breeding for disease resistance and climate adaptation remains constrained by long generation cycles. We benchmarked genomic (SNP), phenomic (NIRS), and metabolomic (untargeted LC-MS) prediction for 24 agronomic traits in a biparental population phenotyped over three years. Seven statistical frameworks and four tissue x timepoint combinations (wood; vineyard leaves at budbreak and flowering; greenhouse leaves at flowering) were evaluated, together with feature-wise BLUPs across samples. Cross-year and cross-population analyses with two additional populations assessed temporal robustness and transferability. Genomic prediction was most accurate (up to r = 0.83), metabolomic prediction was intermediate (up to r = 0.59), and phenomic prediction was lowest (up to r = 0.39) despite its lower acquisition cost. Metabolite features were more heritable than NIR wavelengths, for which most unexplained variation remained residual under the fitted model. Multi-omics integration produced limited overall gains. These results support genomic selection as the primary approach, with metabolomic or phenomic screening considered only for traits and sampling designs that show reproducible predictive signal.
Why it matches plant phenotyping methodsブドウ育種集団の複数形質について、NIRSによるフェノミック測定を含む予測モデルを比較・検証し、交差年・集団で頑健性と転移性も評価しているため、形質推定法の技術的ベンチマークが中心である。
abstractWe benchmarked genomic (SNP), phenomic (NIRS), and metabolomic (untargeted LC-MS) prediction for 24 agronomic traits in a biparental population phenotyped over three years.
Sugar content is a crucial indicator of grape ripeness and grading, and developing non-contact and non-destructive sugar content detection devices is essential for grape-picking robots and sorting platforms. Spectroscopy, which can detect the chemical composition of grapes, has become a key technology for developing non-destructive testing devices. In this paper, we collected 2,880 randomly labeled multispectral images of Sunshine Rose grapes with a Changguang Yuchen MS600 PRO multispectral camera and measured the sugar content (in Brix values) of the labeled grapes with a handheld refractometer, using data exclusively from this grape variety. To address noise and misalignment issues in the multispectral images, we proposed preprocessing methods including Gaussian denoising and ECC (Enhanced Correlation Coefficient) algorithm registration. Based on a ResNet-50 residual network, we constructed a grape sugar content prediction regression model Improved-Res with SE (Squeeze-and-Excitation) attention modules, DSC (Depthwise Separable Convolutions), and Inception modules. The model's performance was evaluated by MSE (Mean Squared Error), MAE (Mean Absolute Error), and R 2 (R-Square) metrics. We compared the performance of four feature extraction methods combined with four traditional machine learning models, as well as seven deep learning models. The results showed that among traditional machine learning methods, the combination of color histogram feature extraction and the XGBoost regression achieved the best performance, with MSE, MAE, and R 2 of 1.35, 0.90 Brix, and 0.78, respectively. Among deep learning methods, the ResNet-50 model demonstrated the best performance, with MSE, MAE, and R 2 of 0.95, 0.96 Brix, and 0.84, respectively. Effective improvements of SE attention module, depthwise separable convolutions, and Inception module in the ResNet-50 model was confirmed through ablation experiments: the proposed Improved-Res model achieved MSE, MAE, and R 2 of 0.49, 0.55 Brix, and 0.92, respectively, which significantly outperformed traditional machine learning methods and classical deep learning models.
Why it matches plant phenotyping methodsブドウ果実の糖度という植物形質をマルチスペクトル画像から非破壊推定する前処理・深層学習モデルを開発し、複数手法との比較とアブレーション検証を行っており、フェノタイピング手法が中心である。
abstractdeveloping non-contact and non-destructive sugar content detection devices is essential for grape-picking robots and sorting platforms.
Grading grapevine downy mildew severity is essential for the precise application of pesticides. Since typical symptoms appear on the abaxial (underside) surface of grape leaves, and lesion area proportion determines severity, it is necessary to analyze lesion characteristics and develop adaxial-to-abaxial lesion inversion methods to build lightweight yet accurate grading models. This study proposes a comprehensive disease grading framework for grape downy mildew. First, a convolutional neural network (CNN)-based classification model is developed with specialized modules and coordinate attention to enhance feature extraction and semantic richness for improved lesion identification. Second, a novel K-Means++-CNN-Vote Consolidation lesion extraction method is introduced. In this framework, K-Means++ segments leaf sub-images, CNNs classify lesion types, and a voting mechanism consolidates results-addressing challenges posed by irregular lesion shapes and blurred boundaries. Finally, an abaxial lesion inversion framework is established by constructing a morphological feature mapping between the adaxial and abaxial surfaces, utilizing mapping functions and lesion generation techniques to infer the abaxial lesion distribution from the adaxial images. Experimental results showed disease grading accuracies of 82.16% (combined adaxial and abaxial), 79.74% (adaxial only), and 84.59% (abaxial only), with a model size of 5.08 MB. Lesion segmentation accuracies reached 89.29% (adaxial and abaxial), 76.92% (adaxial), and 64.47% (abaxial), while the adaxial-to-abaxial lesion inversion achieved an 80% similarity. This study provides methodological support for the online grading of grapevine downy mildew and offers a scientific basis for precise disease control.
Why it matches plant phenotyping methodsブドウ葉の病斑面積・分布という植物病害状態を画像から抽出・推定する手法を開発し、病害重症度の評価精度も検証しており、フェノタイピング手法が中心である。
abstracta novel K-Means++-CNN-Vote Consolidation lesion extraction method is introduced
The identification of archaeological fruits and seeds is crucial for understanding the relationships between humans and plants within the cultural and biological history of both wild and cultivated species. We compared the relative performance of a deep learning approach, namely convolutional neural networks (CNN), and outline analyses via geometric morphometrics using elliptical Fourier transforms (EFT) at identifying pairs of plant taxa. We used their seeds and fruit stones that are the most abundant organs in archaeobotanical assemblages, and whose morphological identification, chiefly between wild and domesticated types, allows to document their domestication and biogeographical history. We used existing modern datasets of four plant taxa (barley, olive, date palm and grapevine) corresponding to photographs of two orthogonal views of their seeds that were analysed separately to offer a larger spectrum of shape diversity. Sample sizes ranged from 473 to 1,769 seeds per class, which constitute a relatively small dataset for training CNNs models yet typical within archaeobotanical research. On these eight datasets, we compared the performance of CNN and EFT coupled with linear discriminant analyses. Our objectives were twofold: i) to test whether CNN can beat geometric morphometrics in taxonomic identification and if so, ii) to test which minimal sample size is required. We ran simulations on the full datasets and also on subsets, starting from 50 images in each binary class. For the CNN network, we deliberately used a candid approach relying on pre-parameterised VGG19 network. For EFT, we used a state-of-the art morphometrical pipeline. The main difference rests in the data used by each model: our CNN used bare photographs where EFT used outline coordinates. This "pre-distilled" geometrical description of seed outlines is often the most time-consuming part of morphometric studies. Results show that our CNN beats EFT in most cases, even for very small datasets. We finally discuss the potential of CNNs for archaeobotany, and how bioarchaeological studies could embrace both approaches, used in a complementary way, to better assess and understand the past history of species.
Why it matches plant phenotyping methods種子・果実石の画像形態を対象に、CNNと幾何学的形態計測を比較し、分類性能と必要サンプル数を検証する方法中心の研究である。植物器官の形状という観測可能な形質の抽出・識別を扱う。
abstractWe compared the relative performance of a deep learning approach, namely convolutional neural networks (CNN), and outline analyses via geometric morphometrics using elliptical Fourier transforms (EFT) at identifying pairs of plant taxa.
Climate change and evolving land management practices are reshaping soil–plant interactions critical for sustainable viticulture. These interactions are driven by soil texture, hydrogeochemical gradients, and climatic conditions, influencing grapevine traits like nutrient and water content. Integrating innovative methods, this study explores the relationship between soil variability and grapevine characteristics in the Médoc wine region, France. The research combines hyperspectral imaging, electromagnetic induction (EMI), and electrical resistivity tomography (ERT) with traditional soil and leaf sampling. Hyperspectral data, using visible-near infrared (VNIR) wavelengths, reliably estimated leaf traits such as nitrogen and water content, yielding strong predictive relationships (R2 up to 0.8). These findings suggest VNIR-based indices are cost-effective for monitoring grapevine physiology. Geophysical data revealed significant soil textural gradients, delineating sand, transitional (loam, sandy loam), and clay textural soil classes. Apparent electrical conductivity (ECa) and inverted electrical conductivity (EC) correlated with soil texture and grapevine traits, particularly at depths around 50 cm, aligning with primary root zones. However, interannual variability in correlations emphasised the influence of weather conditions and phenological stages, highlighting the need to align data acquisition with vine growth phases. The integration of hyperspectral imaging and geophysical methods provides a novel framework for linking soil and plant parameters. This interdisciplinary approach enhances the spatial resolution and scalability of vineyard monitoring, offering actionable insights for precision viticulture. Future work should expand datasets and refine predictive models to improve the understanding of soil–plant dynamics under changing environmental conditions. These findings underscore the potential of combining hyperspectral and geophysical data to develop climate-resilient vineyard management strategies, advancing precision agriculture, and sustainable viticulture practices.
Why it matches plant phenotyping methodsハイパースペクトル画像によりブドウ葉の窒素・水分などの植物形質を推定し、予測性能を評価している。土壌調査も含むが、植物形質の取得・推定手法が主要な技術的貢献である。
abstractHyperspectral data, using visible-near infrared (VNIR) wavelengths, reliably estimated leaf traits such as nitrogen and water content, yielding strong predictive relationships (R2 up to 0.8).
The PlantCity dataset addresses significant agricultural yield losses in Pakistan from plant diseases. It provides 10,667 high-resolution images of leaves from 12 key crops: apple, apricot, bean, cherry, maize, fig, grape, loquat, pear, tomato, walnut, and persimmon. The images are organized into 52 classes (41 diseased and 11 healthy) and augmented to a total of 52,273 images. Data was collected in real-field conditions in Charsadda (34.15°N, 71.74°E, typical temperature 40-44 °C) and Chitral (35.85°N, 71.79°E, typical temperature 25-30 °C) from April to July 2023-2024. The dataset enables the development of deep learning models for automated disease classification and captures a range of environmental factors, including high temperatures that can exacerbate disease symptoms. It utilizes smartphone-based computer vision to facilitate early disease identification, thereby supporting precision farming and sustainable agriculture in Pakistan.
Why it matches plant phenotyping methods植物葉の病害状態を画像から分類するデータセットが研究の中心であり、植物病害フェノタイピング用の画像データセットとして適格です。
abstractThe PlantCity dataset addresses significant agricultural yield losses in Pakistan from plant diseases.
Reproduction assets foundThe paper is a Data in Brief article describing the PlantCity plant leaf image dataset (10,667 original images, 52 classes, 12 crops, collected in Pakistan). The dataset itself is the paper's core phenotyping asset and is publicly deposited on Mendeley Data with a direct URL provided in the article.Dataset · publicon of diseases, pests, or environmental stress in plant leaves.
Data source location
Charsadda (Village Sarki) chosen for tomato and Chitral (Village Danin) for the other 11 crops, Khyber Pakhtunkhwa, Pakistan
Data accessibility
Repository name: Mendeley Data
Data identification number: 10.17632/w8kh2xkspx.2
Direct URL to data: https://data.mendeley.com/datasets/w8kh2xkspx/1
Related research article
None
1
Value of the Data
•
The PlantCity dataset is comprehensive, consisting of 10,667 high-resolution images across 52 classes (41 diseased, 11 healthy) from 12 crop species, collected from Charsadda (tomato disease symptoms) and Danin Chitral (selected for its agro-climatic suitability for fruOpen asset ↗Mendeley Data · 10.17632/w8kh2xkspx.2lines:1-48Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
Boron is an essential micronutrient for grapevine growth, yet excessive levels can impair photosynthesis, reduce yields, and diminish fruit quality. This study evaluated the potential of leaf spectroscopy combined with machine learning to identify boron-tolerant rootstocks rapidly and cost-effectively. We screened both commercial grapevine rootstocks and wild Vitis germplasm under boron treatments ranging from 0.5 to 8 ppm, measuring leaf boron accumulation, stomatal conductance, photosystem II efficiency, and leaf reflectance. The results revealed substantial genotypic variation in boron exclusion, with some genotypes maintaining low leaf boron concentration despite high substrate concentrations. Classification models (partial least squares discriminant analysis and random forest classification) outperformed regression models (partial least squares regression and random forest regression) in distinguishing boron-excluding genotypes, achieving 68 % to 79 % accuracy within just eight days after stress initiation. Reflectance-based vegetation indices such as the Normalized Difference Vegetation Index, Photochemical Reflectance Index, Structure Insensitive Pigment Index, and Chlorophyll Index indicated that boron stress reduces chlorophyll levels and may induce carotenoid accumulation, suggesting a photosynthetic tolerance mechanism. Although quantitative prediction of leaf boron concentration proved more challenging, simulations showed that even modest prediction accuracies (~60 %) can substantially boost genetic gains if larger populations are screened and selection intensities are increased. These findings underscore the value of leaf spectroscopy for high-throughput phenotyping, allowing breeders to rapidly identify and advance boron-tolerant rootstocks.
Why it matches plant phenotyping methods葉分光と機械学習を用いた耐性根株の迅速な表現型推定・選抜が中心であり、反射スペクトルからホウ素耐性や関連生理形質を高スループットに評価する方法を実質的に適用・検証している。
abstractThis study evaluated the potential of leaf spectroscopy combined with machine learning to identify boron-tolerant rootstocks rapidly and cost-effectively.
Accurate assessment of a crop’s water requirement is essential for optimising irrigation scheduling and increasing the sustainability of water use. The crop coefficient (Kc) is a dimensionless factor that converts reference evapotranspiration (ET0) into actual crop evapotranspiration (ETc) and is widely used for irrigation scheduling. The Kc reflects canopy cover, phenology, and crop type/variety, but is difficult to measure directly in heterogeneous perennial systems, such as vineyards. Remote sensing (RS) products, especially open-source satellite imagery, offer a cost-effective solution at moderate spatial and temporal scales, although their application in vineyards has been relatively limited due to the large pixel size (~100 m2) relative to vine canopy size (~2 m2). This study aimed to improve grapevine Kc predictions using vegetation indices derived from harmonised Sentinel-2 imagery in combination with spectral unmixing, with ground data obtained from canopy light interception measurements in three winegrape cultivars (Shiraz, Cabernet Sauvignon, and Chardonnay) in the Barossa and Eden Valleys, South Australia. A linear spectral mixture analysis approach was taken, which required estimation of vine canopy cover through beta regression models to improve the accuracy of vegetation indices that were used to build the Kc prediction models. Unmixing improved the prediction of seasonal Kc values in Shiraz (R2 of 0.625, RMSE = 0.078, MAE = 0.063), Cabernet Sauvignon (R2 = 0.686, RMSE = 0.072, MAE = 0.055) and Chardonnay (R2 = 0.814, RMSE = 0.075, MAE = 0.059) compared to unmixed pixels. Furthermore, unmixing improved predictions during the early and late canopy growth stages when pixel variability was greater. Our findings demonstrate that integrating open-source satellite data with machine learning models and spectral unmixing can accurately reproduce the temporal dynamics of Kc values in vineyards. This approach was also shown to be transferable across cultivars and regions, providing a practical tool for crop monitoring and irrigation management in support of sustainable viticulture.
Why it matches plant phenotyping methodsSentinel-2のスペクトルアンミキシングと回帰モデルにより、ブドウ樹の樹冠被覆・季節的Kcを推定し、品種間・地域間で精度検証している。水管理応用を含むが、植物状態の取得・推定手法が中心的である。
abstractThis study aimed to improve grapevine Kc predictions using vegetation indices derived from harmonised Sentinel-2 imagery in combination with spectral unmixing
Grapevine Leafroll Disease (GLD) poses a significant economic burden on the wine industry in major wine-producing regions. Conventional methods of phenotyping GLD are inefficient and delay vineyard management decisions. The emergence of affordable Unmanned Aerial Vehicles (UAV) provides unprecedented opportunities for GLD high-throughput phenotyping. However, detecting GLD-infected grapevines at the canopy level using UAV images is still a challenge due to the subtle differences between GLD canopy features and background. In this paper, we propose a GLD Detector (GLDD) for mapping GLD epidemics from UAV images. A new attention mechanism module, namely, Channel Attention with Transformers (CAT) is proposed to alleviate the difficulty of extracting high-resolution features from the elongated canopy. We redesigned YOLOv7-tiny for GLDD and conducted a series of ablation experiments to evaluate its performance. Experimental results show that GLDD outperforms YOLOv7-tiny by 3.1% and YOLOv6-tiny by 6.5%, reaching an accuracy of 88.2%. In comparison to several one-stage object detectors such as YOLOv5, YOLOX, PP-YOLOE, and YOLO-FaceV2, GLDD obtains the best detection results. Additionally, compared to convolutional-based detectors such as Faster-RCNN and transformer-based detector SwinT, GLDD performs better in speed and accuracy. Furthermore, GLD-infected grapevine distribution is also mapped by using GLDD detection results at the field scale.
Why it matches plant phenotyping methodsUAV画像からブドウ樹の葉巻病感染状態を推定する検出手法を開発・比較評価しており、植物病害状態のフェノタイピングが中心です。
abstractConventional methods of phenotyping GLD are inefficient and delay vineyard management decisions.
Grapevine winter pruning is a labor-intensive and repetitive process that significantly influences grape yield and quality at harvest and produced wine. Due to its complexity and repetitive nature, the task demands skilled labor that needs to be trained, as in many other agricultural sectors. This paper encompasses an approach that targets using a robotic system to perform autonomous grapevine winter pruning using a vision system and artificial intelligence. In our previous work, we presented a 2D neural network that segmented images of grapevines into 5 different classes of plant organs during their dormant season. In this paper, we expand into the third dimension, introducing point clouds into our algorithm. The 3D approach creates instance-segmented point clouds using depth images and segmentation masks obtained with our 2D neural network. After the 3D reconstruction, the system extracts thickness measurement and uses agronomic knowledge to place pruning points for balanced pruning. The study not only delineates the integration of 2D and 3D methods but also scrutinizes their efficacy in pruning point identification. The real-world performance of the created system was evaluated and statistically analyzed on data collected during field trials in the winter pruning season 2022/2023, where the system was used in a potted vineyard to prune a set of test vines, where the positive success rate is 54.2%. Moreover, as one of the main contributions, the paper underscores a unique facet of adaptability, presenting a customizable framework that empowers end-users to fine-tune parameters according to the expected balanced pruning. This adaptability extends to variables such as the number of nodes to retain on pruned spurs and the preferred cane thickness, encapsulating the versatility of the 3D approach.
Why it matches plant phenotyping methods2D画像分割と3D点群再構成を統合し、ブドウ樹器官の厚さを抽出して剪定点を生成・評価する手法が研究の中心であり、植物形質の取得と技術性能検証を含む。
abstractThe 3D approach creates instance-segmented point clouds using depth images and segmentation masks obtained with our 2D neural network.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · OpenAlex · checked 15 Sept 2026
High-throughput phenotyping technologies increase the efficiency of breeding programs, but with larger datasets, errors can accumulate. Plant breeders often conduct quantitative trait locus (QTL) mapping, where large sample size and accurate quantitative response estimates are important for detecting small-effect QTLs. This study examined how phenotype error, inconsistency, and replication changed QTL magnitude and location. Three real sets of phenotype data were used from microscopy robot analysis of grapevine powdery mildew ( Erysiphe necator) severity, which previously resulted in discovery of large ( R 2 = 85%), intermediate ( R 2 = 45%), and small ( R 2 = 9%) effect QTLs. Custom R scripts were written to induce several realistic sources of error, inconsistency, and varied replication. The results were remarkably robust to these changes. Swapping or shifting 2% of samples or changing disease severity by 50% on one replicate had negligible impact on QTLs. Unreplicated simulations produced the largest logarithm of the odds score range (5.55 to 8.27) and mean logarithm of the odds score deviation (−1.72 to −3.22; Cohen's D = 1.48 to 2.12). The large-effect-size QTL ( REN12) was always detected. The intermediate-effect-size QTL ( REN13) was detected except when three of the eight replicates were analyzed individually. Even for the small-effect-size locus ( NYVPLG9), error scenarios rarely (2 of 9,000 cases) eliminated significant QTL detection, versus no replication (9 of 10). Thus, the benefits of data volume associated with high-throughput phenotyping technologies outweigh the cost of the increased errors tested here. Instead, the focus should be on examining how each experimental replicate contributes to the results of the QTL mapping analysis.
Why it matches plant phenotyping methods高スループット画像表現型測定で生じる誤差・反復数・不整合がQTL解析結果に与える影響を系統的に評価しており、表現型取得データの技術的妥当性検証が中心である。
abstractThis study examined how phenotype error, inconsistency, and replication changed QTL magnitude and location.
GrapevineField / plotRGB / grayscaleObject detectionGrowth / development / phenology
Bud break is a critical phenological stage in muscadine grapevines, marking the start of the growing season and the increasing need for irrigation management. Real-time bud detection enables irrigation to match muscadine grape phenology, conserving water and enhancing performance. This study presents BudCAM, a low-cost, solar-powered, edge-computing camera system based on Raspberry Pi 5 and integrated with LoRa radio board, developed for real-time bud detection. Nine BudCAMs were deployed at Florida A&M University Center for Viticulture and Samll Fruit Research from mid February to mid March, 2024, monitoring three wine cultivars (A-27, noble, and Floriana) with three replicates each. Muscadine grape canopy images were captured every 20 minutes between 7:00 to 19:00, generating 2656 high-resolution (4656×3456 pixels) bud break images as database for bud detection algorithm development. The dataset was divided into 70% training, 15% validation, and 15% test. YOLOv11 models were trained using two primary strategies: a direct single-stage detector on tiled raw images and a refined two-stage pipeline that first identifies the grapevine cordon. Extensive evaluation of multiple model configurations identified top performers for both the single-stage (mAP@0.5=86.0%) and two-stage (mAP@0.5=85.0%) approaches. Further analysis revealed that preserving image scale via tiling was superior to alternative inference strategies like resizing or slicing. Field evaluations during the 2025 growing season confirmed the system’s effectiveness, with the two-stage model showing greater robustness to environmental noise like lens fog. A time-series filter smooths the raw daily counts to reveal a clear phenological trend for visualization. In its final deployment, the autonomous BudCAM system captures an image, runs inference on-device, and transmits the bud count in under three minutes, demonstrating a complete, field-ready solution for precision vineyard management.
Why it matches plant phenotyping methodsブドウの芽数・芽吹きという植物の表現型を、エッジカメラ、画像データセット、検出アルゴリズム、時系列処理で取得・推定するシステムを開発・評価しており、方法が研究の中心である。
abstractThis study presents BudCAM, a low-cost, solar-powered, edge-computing camera system based on Raspberry Pi 5 and integrated with LoRa radio board, developed for real-time bud detection.
Accurate and early detection of plant leaf diseases is crucial for ensuring crop health and improving agricultural productivity. This work proposes a hybrid deep learning model that combines ResNet18, Inception blocks, and fully connected Capsule layers to classify leaf images of apple, grape, and corn plants into healthy or diseased categories. ResNet18 is used as the backbone for deep feature extraction, while Inception modules enhance the network’s ability to capture multi-scale patterns. Capsule layers are employed at the final stage to retain spatial relationships and pose information, improving the model's ability to recognize complex disease features. The model is trained and evaluated using images from the PlantVillage dataset, with separate configurations for each crop. The proposed model achieved validation accuracies of 99.84% for apple, 100% for grape, and 97.27% for corn. Performance is further assessed using precision, recall, and F1-score, and compared against a baseline ResNet18 model. The results demonstrate that the proposed architecture significantly improves classification accuracy and feature understanding, making it a strong candidate for real-world agricultural disease monitoring systems.
Why it matches plant phenotyping methods植物葉画像から健全・病害状態を推定する画像ベースの深層学習手法を開発・評価しており、病害表現型の抽出が研究の中心である。
abstractThis work proposes a hybrid deep learning model that combines ResNet18, Inception blocks, and fully connected Capsule layers to classify leaf images of apple, grape, and corn plants into healthy or diseased categories.
• RGB-D camera and Jetson platform enable precise, adaptive vineyard spraying. • Canopy volume estimation reduces plant protection product use by 57.4%. • Real-time system adjusts spray rates for efficient, sustainable vineyard management. • Instance segmentation and 3D meshing provide accurate canopy and trellis detection. • Jetson’s parallel computing accelerates processing for fast, reliable results. Sustainable vineyard management requires precise and efficient application of plant protection products to minimise environmental impact while ensuring plant health. This study presents a variable-rate spraying system that integrates an RGB-D camera with a GPU-equipped edge computing platform to enable accurate, real-time adjustment of spray flow rates in vineyards. A tensor-based representation of RGB-D data is employed to accelerate the entire processing pipeline. Based on this structure, a fast approximate meshing method is applied to rapidly generate 3D meshes from point clouds. To incorporate semantic information from RGB images, an instance segmentation model is used to detect grapevine canopies and trellis posts. The resulting canopy masks are used to isolate the canopy meshes, while the trellis posts serve as reference planes for canopy volume estimation via mesh projection. Based on the computed volume, pulse-width modulation signals are generated to dynamically control spray flow rates. Field experiments were conducted to evaluate the system’s effectiveness and real-time performance. The results demonstrated that the estimated canopy volume is a reliable indicator for regulating application rates. Compared to uniform-rate spraying, the proposed system reduced plant protection product consumption by 57.4% while ensuring adequate droplet coverage. Additionally, the system demonstrated satisfactory real-time performance even on entry-level hardware. Overall, the proposed variable-rate spraying system offers an accurate, real-time, and cost-effective solution for precision viticulture, highlighting its potential for commercial deployment in sustainable vineyard management.
Why it matches plant phenotyping methodsRGB-D画像、インスタンスセグメンテーション、3Dメッシュによりブドウ樹冠を抽出し、樹冠体積という植物形質を推定する手法とリアルタイム基盤が研究の中心であるため。
abstractThis study presents a variable-rate spraying system that integrates an RGB-D camera with a GPU-equipped edge computing platform to enable accurate, real-time adjustment of spray flow rates in vineyards.
Quantifying crop responses to increasing temperatures is critical for predicting the productivity and sustainability of agricultural systems under environmental change. Physiological trait data associated with maximum Rubisco carboxylation ( V cmax ) and maximum electron transport ( J max ) rates are especially important predictors of crop response to elevated temperatures. However, when generating V cmax and J max data, steady-state methods of gas exchange measurements are time-consuming; thus, non-steady-state methods have been developed to obtain these measurements faster, prospectively allowing for trait data collection of considerably more varieties of crops. Globally important and geographically widespread vineyards are of particular interest due to the high economic value and the susceptibility of these managed systems to climate warming, especially in Canada, where the annual rate of warming far exceeds global averages. In this study, we examined the efficacy of the high-throughput, non-steady-state dynamic assimilation technique (DAT) for obtaining V cmax and J max data from wine grapes. Specifically, we measured V cmax and J max (alongside leaf nitrogen [N] concentrations and leaf mass per unit area [LMA]) across seven of the world's most common wine grape ( Vitis vinifera L.) varieties, namely, Cabernet franc, Cabernet sauvignon, Merlot, Pinot noir, Riesling, Sauvignon blanc, and Viognier. Our results show that V cmax and J max estimates derived from the DAT were strongly correlated to those obtained through the steady-state method ( r 2 = 0.748 and 0.908, respectively), and J max did not differ significantly between the two methods. Additionally, leaf N explained 43%-46% and 56%-58% of the variation in V cmax and J max , respectively, across both methods. Our results suggest that the DAT represents a viable tool for rapidly estimating intraspecific variation in important physiological traits and allows for increased replication and the inclusion of additional varieties when evaluating the responses of wine grape and other crops to climate warming.
Why it matches plant phenotyping methodsワインブドウの生理形質を高速取得する動的同化技術（DAT）を定常法と比較検証しており、植物表現型の測定法が中心的である。
abstractwe examined the efficacy of the high-throughput, non-steady-state dynamic assimilation technique (DAT) for obtaining V cmax and J max data from wine grapes.
Reproduction assets foundThe paper's physiological trait data (Vcmax, Jmax, leaf N, LMA for seven wine grape varieties) are openly deposited in the University of Toronto Borealis Dataverse, per the Data Availability Statement. No author analysis code or trained models are reported.Dataset · publicThe data that support the findings of this study are openly available in the Borealis Repository—University of Toronto Dataverse at https://doi.org/10.5683/SP3/URPVFF .Open asset ↗Borealis Repository—University of Toronto Dataverse · 10.5683/SP3/URPVFFlines:277-347Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Today, one of the most important first steps in attaining sustainable agriculture and guaranteeing food security is the detection of plant diseases. Quantitative analysis of plant physiology is now feasible thanks to developments in computer vision and imaging technologies. On the other hand, manual diagnosis requires a lot of work and in-depth plant pathology knowledge. Numerous innovative methods for identifying and classifying plant diseases have been widely used. In this study, we propose a novel hybrid classification method that combines (q,τ)-Nabla calculus quantum deformation-based features with deep learning feature representations to classify diseases in grapevine leaves. The methodology of this study relies on:•Nabla calculus quantum deformation features are utilized to extract robust handcrafted features that capture local texture and structural variations associated with disease symptoms.•Deep features are extracted using a pre-trained convolutional neural network, which captures high-level semantic information from leaf images.The concatenated feature vectors are then fed into a machine learning classifier for final prediction. Test results on a dataset of grapevine leaf disease show that the proposed method outperforms individual approaches, in accuracy. The proposed method helps minimize financial losses and support effective plant disease management, thereby improving crop yield and contributing to food security.
Why it matches plant phenotyping methodsブドウ葉画像から病徴に基づく病害状態を抽出・分類する新規画像解析手法を開発しており、植物フェノタイピング手法が中心である。
abstractwe propose a novel hybrid classification method that combines (q,τ)-Nabla calculus quantum deformation-based features with deep learning feature representations to classify diseases in grapevine leaves.
Downy mildew is one of the most destructive diseases in wine-growing regions, severely reducing yield and fruit quality. Traditional detection methods rely on expert scouting, which is labour-intensive, subjective and often imprecise. In this context, Artificial intelligence (AI) offers a promising alternative, enabling the use of conventional RGB images for field level disease detection. Therefore, this study proposes a deep learning-based approach to identify leaves with downy mildew symptoms using canopy-level RGB imagery. A comprehensive dataset was generated from fourteen commercial blocks with varying cultivars and disease intensities in northern Spain. RGB images of the canopy were collected under different lighting conditions throughout the day to increase the dataset variability. The YOLOv4 (You Only Look Once) algorithm was trained using this heterogeneous dataset to enhance robustness of the model. The model achieved a mAP of 67 %, an F1-score of 0.69 and an IoU of 62 % in the testing process applied on full canopy images. The number of infected leaves predicted by the model closely matched expert annotation reaching a determination coefficient R² of 0.93. Detection performance remained consistent across different infection levels, suggesting the model’s adaptability to a wide range of conditions. Furthermore, the model was able to accurately localise symptomatic leaves within the full canopy. The results of this study demonstrate that RGB images of the whole canopies can be effectively used to detect downy mildew symptoms, offering a practical approach for in-field disease monitoring. The proposed methodology facilitates the development of automated, on-the-go disease detection systems using mobile platforms such as agricultural robots or vehicles, thereby enabling real-time crop health assessment.
Why it matches plant phenotyping methodsブドウ樹冠のRGB画像からうどんこ病症状葉を検出・計数する深層学習手法を開発し、専門家アノテーションと性能検証を行っており、植物病害状態の取得が中心である。
abstractthis study proposes a deep learning-based approach to identify leaves with downy mildew symptoms using canopy-level RGB imagery
Predicting grapevine phenological stages (GPHS) is critical for precisely managing vineyard operations, including plant disease treatments, pruning, and harvest. Solutions commonly used to address viticulture challenges rely on image processing techniques, which have achieved significant results. However, they require the installation of dedicated hardware in the vineyard, making it invasive and difficult to maintain. Moreover, accurate prediction is influenced by the interplay of climatic factors, especially temperature, and the impact of global warming, which are difficult to model using images. Another problem frequently found in GPHS prediction is the persistent issue of missing values in viticultural datasets, particularly in phenological stages. This paper proposes a semi-supervised approach that begins with a small set of labeled phenological stage examples and automatically generates new annotations for large volumes of unlabeled climatic data. This approach aims to address key challenges in phenological analysis. This novel climatic data-based approach offers advantages over common image processing methods, as it is non-intrusive, cost-effective, and adaptable for vineyards of various sizes and technological levels. To ensure the robustness of the proposed Pseudo-labelling strategy, we integrated it into eight machine-learning algorithms. We evaluated its performance across seven diverse datasets, each exhibiting varying percentages of missing values. Performance metrics, including the coefficient of determination (R²) and root-mean-square error (RMSE), are employed to assess the effectiveness of the models. The study demonstrates that integrating the proposed Pseudo-labeling strategy with supervised learning approaches significantly improves predictive accuracy. Moreover, the study shows that the proposed methodology can also be integrated with explainable artificial intelligence techniques to determine the importance of the input features. In particular, the investigation highlights that growing degree days are crucial for improved GPHS prediction.
Why it matches plant phenotyping methodsブドウの生育ステージという植物状態を気候データから予測する半教師あり手法を開発し、複数アルゴリズムとデータセットで性能評価しており、フェノタイピング手法が中心である。
abstractThis paper proposes a semi-supervised approach that begins with a small set of labeled phenological stage examples and automatically generates new annotations for large volumes of unlabeled climatic data.
This study proposes a novel system for estimating the 3D structure of grapevines as part of a robotic pruning system. The system aims to accurately and efficiently estimate grapevine structures. Utilizing two RGB-D cameras based on Time-of-Flight (ToF) technology, depth images were captured from a wide field of view. This paper employs a minimum spanning tree (MST) to estimate the grapevine skeleton using a cost function that considers node distance, gravitropism, and connection smoothness. Notably, we developed a new component classification method that accurately classifies structural parts—cordons, shoots, and buds—using only skeletal information. The results demonstrated that the system could effectively distinguish between different parts of the grapevine using just the 3D skeletal structure. The system was evaluated on 10 grapevines in real-world vineyard environments. The proposed method achieved high alignment accuracy with manually constructed ground-truth skeletons, even under occlusion. For bud estimation, statistical analysis based on field data facilitated effective estimation. The proposed method outperformed existing approaches in processing speed, with an average processing time of 619 ms per grapevine. These results indicate the potential for real-time application in robotic pruning, enabling efficient structure estimation with high accuracy. Future work will focus on integration with other subsystems integrated within a robotic pruning system, which is expected to produce synergistic effects and enhance overall system performance.
Why it matches plant phenotyping methodsRGB-D画像からブドウ樹の3D構造、器官分類、芽数を推定する手法を開発・評価しており、ロボット剪定への応用を超えて植物形態フェノタイピングが中心である。
abstractThis study proposes a novel system for estimating the 3D structure of grapevines as part of a robotic pruning system.
Phenotyping is pivotal in biological and agronomical research, enabling the characterization of phenotypic traits in living organisms. Recent advancements have led to the development of innovative platforms that enhance the precision of phenotyping, integrating genetic and ecophysiological analyses for a comprehensive understanding of plant growth under controlled conditions. These technologies are instrumental in studying plant responses to environmental stresses, such as drought, which disrupts water balance in plants. This study focuses on the adaptability of grafted grapevines (Vitis vinifera L.) to drought stress, emphasizing the rootstock influence on scion performance. The experimental trial was performed at 'PhenoPlant,' a cutting-edge phenotyping platform at the University of Torino, DISAFA. PhenoPlant is a non-invasive, high-throughput tool that employs advanced technologies, including a PlantEye sensor for 3D vision and multispectral imaging, measurement of the potted-plant evapotranspiration by gravimetric technique, water potential assessment and Infra-Red Gas Analysis for leaf-to-atmosphere gas exchange detection. Grapevine responses to drought stress across eleven scion/rootstock combinations, featuring clones of Nebbiolo and Pinot Noir grafted onto rootstocks with varying drought tolerance were assessed. A 13-day drought-recovery experiment on grafted 1-year old plants, three months after in-pot-transplanting revealed significant differences in drought responses among rootstock/scion combinations. Drought-tolerant rootstocks (e.g., 1103 P, 110 R, 140Ru, M2) maintained stable spectrometric indices (e.g.: GLI, Green Leaf Index) mirroring morpho-physiological ones (e.g., Leaf Surface Angle - SA, Stomatal Conduction - gs, Stem Water Potential and Evapotranspiration), unlike their less tolerant counterparts (e.g., Kober 5BB, SO4, 420 A, Gravesac). In particular, after 10 days of water removal, a reduced variation in some traits was observed in tolerant combinations (SA: 39-44°; GLI ≈ 0.33-0.35; gs: 34.5-45.4 mmol H₂O·m⁻²·s⁻¹), while decreasing markedly in sensitive ones (SA: 27-35°; GLI: 0.28-0.32; gs: 8.6-10.8 mmol H₂O·m⁻²·s⁻¹), underscoring the rootstock's crucial role in drought response, independently from scion cultivar. These findings are vital for a fast and early assessment of multiple rootstock/scion combinations to optimize grapevine management and breeding programs for enhanced performance under water-limited conditions. Intrinsic limitations of the measurement system and aspects to be considered to export results from the platform to the vineyard are presented and discussed.
Why it matches plant phenotyping methods3D・マルチスペクトルセンサーを用いる高スループット表現型解析プラットフォームを中心に、干ばつ応答の早期評価と測定系の限界を検討しているため。
titleCan high-throughput 3D and multispectral phenotyping detect early grapevine responses to water stress events?
Grape is one of the most widely consumed fruit worldwide.There are several diseases present in grape leaf that reduces its yield regularly.Its early and accurate detection is crucial for effective yield outcomes.Classification techniques provided high potential for assisting farmers.This paper presents a four-class classification approach for grape leaf image classification using medium gaussian Support Vector Machine (SVM).SVMs are one of the powerful supervised learning models for classification task.In the proposed approach, features are extracted from grape leaf images from Niphad dataset.The images provide information for distinguishing data in four categories.The texture features of images like Difference theoretic texture features (DTTF), First Order Statistics (FOS), Fractal Texture (FT), Grey Level Difference Statistics (GLDS), Statistical Feature Matrix (SFM), Local Binary Patterns (LBP), Segmentation Based Fractal Texture Analysis (SFTA) and Tamura features are used to form a feature set for classification.In the proposed methodology evaluation is done using accuracy and sensitivity.The results are observed to outperform the state of the art methods.
Why it matches plant phenotyping methodsブドウ葉画像から病害状態を分類する画像ベースの表現型推定手法が研究の中心であり、特徴抽出、SVM分類、精度・感度評価を実施している。
abstractThis paper presents a four-class classification approach for grape leaf image classification using medium gaussian Support Vector Machine (SVM).
The present dataset is a collection of multispectral images designed for development of detection algorithms for grapevine diseases like Flavescence dorée (FD) and Esca (ED). Although FD severely threatens viticulture, there are few public datasets and none with multispectral data collected in the field. The collected images have been taken from a frontal perspective of vineyard plants that highlights details of leaves and trunks facilitating detailed disease analysis. The data were collected using a Micasense RedEdge-P multispectral camera, capturing six spectral bands across 172 image captures of three different grapevine varieties used in Lambrusco wines: Ancellotta, Marani, and Salamino. The dataset includes raw and processed images, calibration images for the multispectral camera, annotations detailing plant health conditions, and Python-based usage examples for researchers. Potential applications include the development of machine learning algorithms for automated disease detection, image alignment techniques, and background removal methods. The dataset is a valuable resource for advancing remote and proximal sensing in precision agriculture.
Why it matches plant phenotyping methodsブドウ病害という植物状態を対象に、マルチスペクトル画像・注釈・校正データを含む再利用可能なデータセットを構築しており、表現型取得基盤が研究の中心です。
titleA dataset for vineyard disease detection via multispectral imaging
Low-cost multispectral sensors have recently been commercialized, paving the way for decision support in a wide variety of agricultural applications (fertilization, grass cover management, etc.). Such sensors have seldom been tested for agricultural applications taking into account practical constraints (external environment, expected accuracy, etc.). This study proposes to investigate the measurement characteristics of the “AS7265x” (by AMS) which, with 18 spectral bands ranging from blue to near infrared light and a low cost, presents a real potential for a wide range of applications in agriculture, and especially in viticulture. This work proposes a holistic approach including tests of the “AS7265x” in controlled conditions to validate the measurement characteristics in terms of accuracy, repeatability and reproducibility as well as experiments carried out in the field to validate the potential of the sensor in assessing agronomic information in outdoor conditions. Three different “AS7265x” sensors were tested in controlled conditions to assess the accuracy and the repeatability of the 18 different spectral bands as well as the reproducibility of the measurements from one sensor to another. To assess the sensor’s suitability for field applications, two experiments were conducted using different vegetation indices (VIs). The first experiment involved proximal sensing to estimate weed coverage on the soil surface, categorized into five classes ranging from class 1 (0–20%) to class 5 (80–100%), and to evaluate plant vigor using the NDVI. The second experiment utilized contact mode with active light to estimate the nitrogen status of grapevine leaves using four adapted VIs: BGI, SIPI, NI_Tian, and NI_Wang. These two applications were chosen to explore the potential of the large diversity of wavebands available on the sensor as well as different acquisition modes (proximal detection, contact detection). For both experiments, results were compared to specific sensors intended as reference for the considered measurements. Regarding measurement characteristics in controlled conditions, results show that although accurate, the sensors present some bias specific to each spectral band and each sensor. These drawbacks require each sensor to be specifically calibrated before use which may limit their dissemination in agriculture. Once calibrated, results of proximal NDVI measurements performed with the sensor are consistent when compared to a Greenseeker (R²=0.87 for NDVI < 0.75). The sensor also allows to discriminate significantly (p < 0.05) two levels of vine vigour when plants were grown in different conditions. Regarding the measurement of nitrogen status, the sensor shows a good correlation (R² = 0.78) between NI_Wang index and the NBI reference value observed with the Dualex sensor. This work highlights the potential of “AS7265x” sensor to access objective agronomic information. It also highlights its limitations when it comes to acquire reference data. In this case, the quality of the sensor requires a calibration procedure specific to each sensor for each waveband.
Why it matches plant phenotyping methods低コストマルチスペクトルセンサーの精度・反復性・再現性を検証し、NDVIによるブドウ樹勢および葉の窒素状態の推定に適用している。センサー校正と植物形質取得が研究の中心である。
abstractThis study proposes to investigate the measurement characteristics of the “AS7265x” (by AMS)
Abstract Background and goals Measuring evapotranspiration (ETc) in vineyards is important to optimize vineyard irrigation and water management practices. Previous work demonstrated a strong correlation between the amount of shaded area under the vine at high noon and crop coefficient. This parameter can be measured with a photovoltaic sensor (Paso Panel) or by hand using grid paper. We aimed to develop a low-cost and easy-to-use smartphone-based alternative to measure shaded area under a vine. Methods and key findings Videos of the ground under a row of vines were recorded with a smartphone camera on a sunny day in the presence of resident vegetation which consisted of grasses and weeds. A novel computer vision-based algorithm using a segmentation machine learning model and structure-from-motion was developed to estimate the amount of shaded area present. Measurements were collected using a Paso Panel at the same time for comparison. Other Paso Panel measurements were collected to measure the relationship between electrical current and shaded area. Linear regression of this CV-based method to Paso Panel readings yields R2 = 0.68. Conclusions and significance A new model for relating Paso Panel current readings to shaded area was derived empirically. Adoption of the CV-based crop coefficient estimation method could improve spatial resolution of ETc estimates, potentially aiding adoption of variable rate irrigation.
Why it matches plant phenotyping methodsブドウ樹下の遮光面積をスマートフォン映像とコンピュータビジョンで推定し、作物係数・蒸発散量推定に利用する手法を開発・比較検証しており、植物状態の取得手法が中心である。
abstractWe aimed to develop a low-cost and easy-to-use smartphone-based alternative to measure shaded area under a vine.
Grapes are widely cultivated fruit crops, essential for fresh consumption, winemaking and dried product production. However, their yield and quality are significantly impacted by various fungal diseases. This paper provides a large dataset of 2,726 high-quality grape leaf disease images collected from grapes farm of Nashik, India in two years of span 2023 to 2025. The dataset is precisely annotated under the guidance and observation of agriculture domain expert and organized in a well-defined folder structure. The dataset captures the two major categories healthy leaves and unhealthy leaves, during cultivation period. A primary directory containing two main classes Heathy Leaf Images and Unhealthy Leaf images. Further unhealthy class is divided into three subfolders for disease class, namely Downy Mildew, Powdery Mildew and Bacterial Leaf Spot. These are the major fungal disease observed on grape crop causes substantially crop losses and ultimately impact on the yield production. Timely identification of these diseases can significantly reduce the risk of crop loss and help to improve quality of fruit with maximum yield production. This High-quality annotated image dataset can help to design standard advanced AI models for automated disease detection, classification, and prediction. The dataset was validated through a transfer learning approach using the ResNet-18 algorithm and demonstrated the remarkable classification accuracy of 96 %. These results validate the dataset’s quality and its suitability for deep learning-based grape disease detection. Overall, this open-access resource provides a valuable foundation for computer vision, machine learning, and agricultural technology researchers aims to enhance disease management practices in grape production. thus, this is an effective source of data for future studies and real-world applications in sustainable grape production.
Why it matches plant phenotyping methodsブドウ葉の健康状態と病害を画像として収集・注釈したデータセットが研究の中心であり、植物病害状態を直接評価する再利用可能な資源として構築・検証されている。
abstractThis paper provides a large dataset of 2,726 high-quality grape leaf disease images collected from grapes farm of Nashik, India in two years of span 2023 to 2025.
The increasing demand for agricultural products and rising production costs have intensified labour shortages in the agricultural sector. Manual harvesting remains essential for products with specific designations, such as wine grapes, where automated solutions cannot match human operators’ dexterity, speed, and care. Minimizing transportation time is also crucial for preserving produce quality and optimizing efficiency. This study aims to optimize harvesting efficiency and vineyard management through the design and implementation of a mobile robotic platform. The platform combines operator dexterity with robotic assistance, continuously tracking operators as they deposit harvested grapes into a harvesting box carried by a robot while gathering data for yield map development. Adaptable to various manual fruit-picking processes, the platform can be integrated into a collaborative harvesting assistance fleet. Field experiments conducted at the Bodegas Terras Gauda (UTM coordinates: 41.95, −8.80, O Rosal, Pontevedra, Spain) vineyard, indicated that operators using robotic assistance reduced their average harvesting time per box by 6 min, increased their total harvested yield by 72.50 kg after two hours (up to 50% more), and reduced manual labour costs by 22.50%. A yield map was developed with high-accuracy GNSS data and an industrial scale mounted on the robot. The map geolocates the weights collected with a maximum variability error of 0.11 kg and successfully expresses grapevine density variability within the same vineyard row. The system preserves produce quality during transportation and significantly eliminates physical strain among operators. These results demonstrate the potential of the robotic platform to improve the efficiency of manual harvesting while maintaining high-quality outcomes.
Why it matches plant phenotyping methodsロボット搭載スケールとGNSSにより収穫重量を地理参照し、ブドウの収量・密度変動を地図化する計測プラットフォームが研究の中心であるため。
abstractA yield map was developed with high-accuracy GNSS data and an industrial scale mounted on the robot.
The timely detection of viral pathogens in vineyards is a critical aspect of management. Diagnostic methods can be labor-intensive and may require specialized training or facilities. The emergence of artificial intelligence (AI) has the potential to provide innovative solutions for disease detection but requires a significant volume of high-quality data as input. With that purpose, we partnered with wine grape growers to collect a robust dataset of verified images. We used those images to train an AI model and develop a handheld application as a decision support tool for grapevine leafroll and red blotch diseases. The tool allows users to scan a grapevine canopy with a mobile device and view a confidence reading describing the likelihood that the imaged vine has visual symptoms consistent with leafroll, red blotch, or a healthy vine. The 86% accuracy under field conditions and generally positive user experience suggest there is potential for the trained use of AI as an investigative tool to quickly assess visual symptoms associated with these grapevine diseases.
Why it matches plant phenotyping methodsブドウの病徴を画像から推定するAIモデル、データセット、携帯アプリを開発・評価しており、植物病害状態の取得・判定手法が研究の中心である。
abstractWe used those images to train an AI model and develop a handheld application as a decision support tool for grapevine leafroll and red blotch diseases.
Accurate, quantitative phenotyping aids in the discovery of quantitative trait loci, particularly those with minor effects. Previously, we optimized replicated precision phenotyping of mapping families after inoculation of leaf discs with the grapevine powdery mildew pathogen ( Erysiphe necator ). Pathogen colonies were stained, and hyphal density was estimated using hyphal transects. This approach outperformed field evaluations and other controlled phenotyping methods but required one or two person-months of microscopy per experiment to evaluate resistance across 300 host genotypes. More recently, we combined advanced macrophotography, robotic sample positioning, and convolutional neural networks to produce a high-throughput phenotyping device, which was modified and commercialized as "Blackbird." Here, that device was tested for nondestructive image collection and computer vision quantification of foliar grapevine powdery mildew. Blackbird outpaced manual microscopy up to 60-fold and nondestructively generated time-series segregating phenotypes from 2 to 9 days postinoculation (dpi). Paired analysis of these phenotypes with RNase H2-amplicon sequencing haplotype markers targeting the Vitis core genome detected REN13 on chromosome 8. Genetic analysis of Blackbird convolutional neural network data explained a greater proportion of the phenotypic variance via hyphae at 4 dpi (24.5%) and conidia at 9 dpi (24.0%) than manual microscopy at 8 dpi (15.8%). As a moderate-effect resistance locus in the widely planted resistant variety 'Norton', which already produces commercial wine quality, REN13 could significantly delay epidemics and could be useful in grape breeding programs to increase the durability of stronger resistance loci (e.g., RUN1 , REN4 , or REN12 ) in resistance gene stacks while maintaining fruit quality.
Why it matches plant phenotyping methods画像取得、ロボット試料位置決め、CNNによる病徴定量を統合した装置を開発・改変し、手動顕微鏡法と比較検証しているため、植物病害表現型の取得・抽出法が中心である。
abstractwe combined advanced macrophotography, robotic sample positioning, and convolutional neural networks to produce a high-throughput phenotyping device
Early and accurate recognition of abiotic stress types is essential for accelerating the selection of stress-tolerant varieties and implementing effective management strategies. This study is motivated by the socio-economic relevance of vineyard and by the increasing need for stressor-specific fingerprint(s) to support the reliable identification of stress type (e.g., drought or salinity) within a high-throughput plant phenotyping domain. This paper presents a reanalysis of physiological and phenotyping data from drought and salt stress experiments in Vitis vinifera focussing the maximum photosynthetic efficiency ( F v / F m ) and leaf Dark Green color. The reanalysis suggests that salt-stressed vines might suffer additional (non-stomatal) limitations curbing net photosynthetic rate (Pn) severely as drought stress does at equivalent stomatal conductance ( g s ) levels. Through a Principal Component (PC) Analysis, physiological and colorimetric response variables were decomposed revealing that F v / F m and Dark Green dominates the non-stomatal PC (∼80%) clustering data between salt and drought experiments. Confusion matrices reveal that model based on F v / F m and Dark Green performed better (accuracy = 1, precision =1) than that based on Pn, g s , transpiration, and stem water potential. This study supports the potential use of F v / F m and Dark Green for early and non-destructive stress type identification.
Why it matches plant phenotyping methods既存の生理・色彩形質を用いてストレス種を識別する解析手法を再評価し、特徴量の比較と混同行列による性能検証を行っており、フェノタイピング手法の評価が中心である。
abstractThis paper presents a reanalysis of physiological and phenotyping data from drought and salt stress experiments
In times of highly effective and cost-efficient genotyping technologies routinely applied in plant research and breeding, the need for comparable high-throughput (HT) and high-resolution phenotyping tools has increased substantially. As a perennial plant, grapevines have very specific requirements for HT phenotyping. Depending on the trait, it can be applied in laboratories or greenhouses, but it is also very important under field conditions to rate the full phenotypic variability of traits like yield or plant vigour throughout the season. For more than a decade, researchers have strived to improve grapevine phenotyping by sensors and automation to dissolve the phenotyping bottleneck. The core goal of the present review is the illustration of promising and reliable opportunities for HT phenotyping in grapevine research and breeding. Therefore, different imaging sensor technologies and their data analysis, including artificial intelligence (AI), will be discussed, focusing on traits that are important for breeding new grapevine varieties. However, the expected outcome of any HT phenotyping approach is similar: transfer of a low-throughput method into an approach that acquires objective, precise, and reliable data for plant evaluation with high spatial and temporal resolution. Furthermore, the collection of large phenotypic data sets and their linkage with environmental or genomic data will provide new or extended insights into the response of grapevines to biotic and abiotic stresses and will significantly support the evaluation of traits, identification of new QTLs, or implementation of breeding strategies like genomic prediction. These advancements offer an improvement of precision and scalability within seedling selection and can additionally contribute to increased sustainability in viticulture.
Why it matches plant phenotyping methodsブドウ育種における高スループット表現型解析技術、画像センサー、データ解析、AIを中心に扱う方法論レビューであり、植物表現型解析手法が中核である。
abstractThe core goal of the present review is the illustration of promising and reliable opportunities for HT phenotyping in grapevine research and breeding.
Accurate canopy characterisation is crucial for the targeted application of plant protection products following the variable rate application (VRA) concept. In this study, two different canopy measurement systems were compared: ultrasonic (US) sensors and UAV-based photogrammetry. A specific device was developed to host a series of US sensors that could conduct a fully automatic canopy characterisation of two vine rows in a single pass. The results of canopy characterisation (canopy width, canopy height, leaf wall area, and tree row volume) were compared with those obtained after complete data processing of the images obtained using a multispectral camera embedded on a UAV. Results indicated that no significant differences have been obtained in the definition of main canopy parameters. Field tests indicated that US sensors offered stable canopy height readings but exhibited variability in width measurements due to factors like ground conditions and sensor placement. Compared to with UAV photogrammetry, US sensors provided comparable results for canopy height and width at a lower cost and with less precision. Therefore, the choice between US sensors and UAVs should consider the resolution requirements, cost, and field conditions. Field data were collected from two commercial vineyards in the Penedès region close to Barcelona (Spain). Before this, laboratory tests were performed using an artificial target to achieve an accurate evaluation of the US sensors. Overall, this study highlighted the potential of ground-based sensing systems for precise and repeatable canopy measurements, contributing to improved vineyard management practices and advanced technological integration for agricultural monitoring.
Why it matches plant phenotyping methodsブドウ樹冠の形態形質を取得する超音波センサーとUAV画像法を開発・比較検証しており、フェノタイピング手法が研究の中心である。
abstractA specific device was developed to host a series of US sensors that could conduct a fully automatic canopy characterisation of two vine rows in a single pass.
Grapevine winter pruning is a labor-intensive and repetitive process that significantly influences grape yield and quality at harvest and produced wine. Due to its complexity and repetitive nature, the task demands skilled labor that needs to be trained, as in many other agricultural sectors. This paper encompasses an approach that targets using a robotic system to perform autonomous grapevine winter pruning using a vision system and artificial intelligence. In our previous work, we presented a 2D neural network that segmented images of grapevines into 5 different classes of plant organs during their dormant season. In this paper, we expand into the third dimension, introducing point clouds into our algorithm. The 3D approach creates instance-segmented point clouds using depth images and segmentation masks obtained with our 2D neural network. After the 3D reconstruction, the system extracts thickness measurement and uses agronomic knowledge to place pruning points for balanced pruning. The study not only delineates the integration of 2D and 3D methods but also scrutinizes their efficacy in pruning point identification. The real-world performance of the created system was evaluated and statistically analyzed on data collected during field trials in the winter pruning season 2022/2023, where the system was used in a potted vineyard to prune a set of test vines, where the positive success rate is 54.2%. Moreover, as one of the main contributions, the paper underscores a unique facet of adaptability, presenting a customizable framework that empowers end-users to fine-tune parameters according to the expected balanced pruning. This adaptability extends to variables such as the number of nodes to retain on pruned spurs and the preferred cane thickness, encapsulating the versatility of the 3D approach.
Why it matches plant phenotyping methods2D画像分割と3D点群再構成を統合し、ブドウ樹器官の厚さを測定して剪定点を生成・評価する手法が研究の中心であるため、植物表現型計測手法として収載する。
abstractThe 3D approach creates instance-segmented point clouds using depth images and segmentation masks obtained with our 2D neural network.
Accurate, quantitative phenotyping aids in the discovery of quantitative trait loci, particularly those with minor effects. Previously, we optimized replicated precision phenotyping of mapping families after inoculation of leaf discs with the grapevine powdery mildew pathogen (Erysiphe necator). Pathogen colonies were stained, and hyphal density was estimated using hyphal transects. This approach outperformed field evaluations and other controlled phenotyping methods but required one or two person-months of microscopy per experiment to evaluate resistance across 300 host genotypes. More recently, we combined advanced macrophotography, robotic sample positioning, and convolutional neural networks to produce a high-throughput phenotyping device, which was modified and commercialized as "Blackbird." Here, that device was tested for nondestructive image collection and computer vision quantification of foliar grapevine powdery mildew. Blackbird outpaced manual microscopy up to 60-fold and nondestructively generated time-series segregating phenotypes from 2 to 9 days postinoculation (dpi). Paired analysis of these phenotypes with RNase H2-amplicon sequencing haplotype markers targeting the Vitis core genome detected REN13 on chromosome 8. Genetic analysis of Blackbird convolutional neural network data explained a greater proportion of the phenotypic variance via hyphae at 4 dpi (24.5%) and conidia at 9 dpi (24.0%) than manual microscopy at 8 dpi (15.8%). As a moderate-effect resistance locus in the widely planted resistant variety 'Norton', which already produces commercial wine quality, REN13 could significantly delay epidemics and could be useful in grape breeding programs to increase the durability of stronger resistance loci (e.g., RUN1, REN4, or REN12) in resistance gene stacks while maintaining fruit quality.
Why it matches plant phenotyping methods高スループット画像取得とコンピュータビジョンによるブドウうどんこ病表現型の定量化が研究の中心で、手動顕微鏡との性能比較・検証も行っている。
abstractwe combined advanced macrophotography, robotic sample positioning, and convolutional neural networks to produce a high-throughput phenotyping device
A decade after the discovery of grapevine red blotch virus (GRBV), there is ample evidence of its detrimental impacts on grapevine physiology, grape composition, and wine production. To mitigate the spread of GRBV in vineyards, roguing is recommended as a disease management response. The imperative to identify and remove diseased vines justifies the development of autonomous scouting. In this study, nearly 700 ground-based hyperspectral images, encompassing both symptomatic and asymptomatic vine canopies, were collected in a Cabernet Franc vineyard during two growing seasons, capturing pre- and post-veraison vine development stages. Spanning 230 bands from visible (VIS) to near-infrared (NIR) domains (510 to 900 nm with 1.7 nm width), canopy spectral signals were isolated from the background through semantic segmentation using U-Net. Simultaneously, the GRBV status of each vine was established in the laboratory through polymerase chain reaction. These two intertwined datasets were used for training various machine learning algorithms and their ensembles. In addition, strategies to reduce dataset size through spectral binning and testing three different feature selection methods (Recursive Feature Elimination, Univariate Feature Selection, and taking into consideration autocorrelation) were explored. Our findings revealed that hyperspectral imagery identified GRBV-infected vines with an accuracy of 75.7 % around harvest, coinciding with the peak of disease symptom expression, utilizing only 19 bands with a 16 nm bin width. Prior to veraison when most vines are asymptomatic, an accuracy of 74.2 % was achieved, employing 5 bands with a 16 nm bin width. This study substantiates the utility of hyperspectral images in the identification of GRBV-infected vines, offering a robust foundation for the development of a streamlined sensing system that holds great promise for the grape and wine industry in effectively scouting vineyards for GRBV.
Why it matches plant phenotyping methodsブドウ樹の感染状態をハイパースペクトル画像と機械学習で推定するセンシング・解析手法を開発・評価しており、植物病害状態の取得が研究の中心です。
Autonomous mobile robotic solutions are increasingly being explored in precision agriculture to aid human workers in labour-intensive or repetitive tasks. Moreover, the emergence of foundation models in vision-based AI domain presents an opportunity to perform automated interpretation of in-field collected data. This study presents a cost-effective mobile robotic research platform designed for autonomous vineyard inspection: it integrates mission planning, real-world navigation and a post-processing pipeline of multimodal data. The system, based on the Leo rover, is equipped with LiDAR, RGB cameras and GNSS-visual-inertial positioning, ensuring reliable operation in GNSS-degraded vineyard environments. We propose a novel methodology for automating several stages of the workflow using various open and in-situ collected data. The robotic platform and processing pipeline were validated through simulation and field experiments, demonstrating its capability for autonomous navigation, 3D reconstruction, AI-based fruit detection and an initial plant health assessment through Large Multimodal Models (LMM). Results show that while 3D mapping provides highresolution spatial data, AI-driven object detection and vision models require further domain adaptation for reaching reliable and trustable operation. The study highlights the feasibility of cost-effective mobile robotic solutions in vineyard monitoring and the potential of integrating AI to enhance agricultural automation.
Why it matches plant phenotyping methods自律型ロボットとマルチモーダル処理パイプラインを開発・検証し、3D再構成、果実検出、植物健全性評価という植物状態の取得を中核的に扱っているため。
abstractThis study presents a cost-effective mobile robotic research platform designed for autonomous vineyard inspection
In addition to purely qualitative trait descriptions, biometric studies have become established within ampelography for the identification and characterisation of grapevine varieties. This publication examines the suitability of a software-assisted foliometric method, based on the self-similarity of vine leaves and employing 24 vein branching nodes, for grapevine variety identification. Various classification methods were used, including PCA, discriminant analysis, and two different neural networks. The findings show that in Vitis vinifera, leaf venation alone provides enough distinguishing features to enable an almost perfect classification of the varieties analysed here—Grüner Veltliner and its parent varieties, Traminer and St. Georgener-Rebe (Mater Veltlinis). However, due to significant individual variability, single leaves are not sufficient. By generating composite leaf data through averaging a small number of leaves, accurate classification becomes feasible. The choice of features also allows for a reasonably precise description of differences in leaf venation, which were analysed in terms of their variability across different years of data collection, within a growing season, and along the primary shoot (from proximal to distal), with particular attention to varietal differences. Leaves of the shoot middle proved more suitable for identification, while the time of year when sampling occurs is of less importance, although differences between spring and summer leaves were observed. Furthermore, the method was employed to provide a comparative description of the leaf venation patterns of the three grapevine varieties. For most traits, Grüner Veltliner exhibited intermediate characteristics between its parent varieties.
Why it matches plant phenotyping methodsソフトウェア支援の葉脈形態計測法を開発・適用し、葉脈形質によるブドウ品種識別と変動性を評価しているため、植物フェノタイピング手法が中心である。
abstractThis publication examines the suitability of a software-assisted foliometric method, based on the self-similarity of vine leaves and employing 24 vein branching nodes, for grapevine variety identification.
Morphological image analysis has emerged as a powerful tool for assessing physical bunch characteristics in viticulture, particularly for estimating grape bunch weight, a key factor affecting vineyard yield and wine quality. Traditional manual sampling methods are labour-intensive, destructive, and prone to significant errors due to vineyard variability and environmental stresses such as water deficit. To address these challenges, this study investigates the potential of two-dimensional (2D) image analysis for non-destructive grape bunch weight estimation across varying levels of water stress. Images of 359 bunches from Cabernet-Sauvignon vines grown under different irrigation regimes, were analysed to extract 13 morphological features. A stepwise multiple regression model was developed to predict bunch weight based on key image-derived features, demonstrating strong explanatory power (adjusted R2 of the prediction = 0.824). The results indicate that features such as area, perimeter, and circularity are strong predictors of bunch weight. While the model demonstrated high accuracy overall, some deviations were observed in large weight categories indicating opportunities for further refinement. These findings demonstrate that image-based phenotyping can reliably estimate bunch weight across a range of water availability scenarios, supporting more precise and efficient vineyard management practices. Future research should focus on enhancing model robustness by integrating additional morphological descriptors and evaluating broader cultivar variability under field conditions.
Why it matches plant phenotyping methods画像からブドウ房の形態特徴を抽出し、房重を推定する手法が研究の中心であるため、植物フェノタイピング手法として採用。
abstractthis study investigates the potential of two-dimensional (2D) image analysis for non-destructive grape bunch weight estimation
Humans need food to sustain their lives. Therefore, agriculture is one of the most important issues in nations. Agriculture also plays a major role in the economic development of countries by increasing economic income. Early diagnosis of plant diseases is crucial for agricultural productivity and continuity. Early disease detection directly impacts the quality and quantity of crops. For this reason, many studies have been carried out on plant leaf disease classification. In this study, a simple and effective leaf disease classification method was developed. Disease classification was performed using seven state-of-the-art pretrained convolutional neural network architectures: VGG16, ResNet50, SqueezeNet, Xception, ShuffleNet, DenseNet121 and MobileNetV2. A simplified SqueezeNet model, GAPNet, was subsequently proposed for grape, apple and potato leaf disease classification. GAPNet was designed to be a lightweight and fast model with 337.872 parameters. To address the data imbalance between classes, oversampling was carried out using the synthetic minority oversampling technique. The proposed model achieves accuracy rates of 99.72%, 99.53%, and 99.83% for grape, apple and potato leaf disease classification, respectively. A success rate of 99.64% was achieved in multiplant leaf disease classification when the grape, apple and potato datasets were combined. Compared with the state-of-the-art methods, the lightweight GAPNet model produces promising results for various plant species.
Why it matches plant phenotyping methods植物葉の病害状態を画像から分類する軽量CNN手法を開発・評価しており、病害表現型の抽出が研究の中心です。
abstractIn this study, a simple and effective leaf disease classification method was developed.
Reproduction assets foundAuthors publicly release GAPNet implementation code via GitHub and Zenodo, and the paper's leaf image datasets (PlantVillage, New Plant Disease, Plant Pathology 2020) are publicly available at listed URLs.Code · publiccle, and approved the final draft.
Asuman Günay Yılmaz conceived and designed the experiments, analyzed the data, authored or reviewed drafts of the article, and approved the final draft.
Data Availability
The following information was supplied regarding data availability:
The data and code are available at GitHub and Zenodo:
- https://github.com/ozgenurr/GAPNet.git .Open asset ↗ozgenurr/GAPNetlines:727-762Dataset · public- Ozge Ozaras. (2025). ozgenurr/GAPNet: GAPNET (GAPNET). Zenodo. https://doi.org/10.5281/zenodo.15163686 .
The datasets are publicly available at:
- Plant Village Dataset: https://github.com/spMohanty/PlantVillage-Dataset/tree/master/raw/color .
- New Plant disease Dataset: https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset/data .
- Plant Pathology: https://www.kaggle.com/competitions/plant-pathology-2020-fgvc7/data .
References
Babalola, Kpai & Toygar (2023) Babalola FO, Kpai NI, Toygar Ö. Deep learning-based classification of apple leaf diseases uOpen asset ↗PlantVillage-Datasetlines:763-789Dataset · public- Ozge Ozaras. (2025). ozgenurr/GAPNet: GAPNET (GAPNET). Zenodo. https://doi.org/10.5281/zenodo.15163686 .
The datasets are publicly available at:
- Plant Village Dataset: https://github.com/spMohanty/PlantVillage-Dataset/tree/master/raw/color .
- New Plant disease Dataset: https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset/data .
- Plant Pathology: https://www.kaggle.com/competitions/plant-pathology-2020-fgvc7/data .
References
Babalola, Kpai & Toygar (2023) Babalola FO, Kpai NI, Toygar Ö. Deep learning-based classification of apple leaf diseases using AlexNet. Computer Science, IDAP-2023: International Artificial Intelligence and Data Processing SympOpen asset ↗lines:763-789Dataset · publicGAPNET (GAPNET). Zenodo. https://doi.org/10.5281/zenodo.15163686 .
The datasets are publicly available at:
- Plant Village Dataset: https://github.com/spMohanty/PlantVillage-Dataset/tree/master/raw/color .
- New Plant disease Dataset: https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset/data .
- Plant Pathology: https://www.kaggle.com/competitions/plant-pathology-2020-fgvc7/data .
References
Babalola, Kpai & Toygar (2023) Babalola FO, Kpai NI, Toygar Ö. Deep learning-based classification of apple leaf diseases using AlexNet. Computer Science, IDAP-2023: International Artificial Intelligence and Data Processing Symposium (IDAP-2023) 2023:67–74. doi: 10.53070/bbd.1349566.
BanjaOpen asset ↗lines:763-789Plant phenotyping relevance match · UnverifiedEurope PMC · checked 14 Sept 2026
The vine plant holds significant importance beyond grape farming due to its diverse products. Various grape-derived products, such as wine and molasses, highlight the vine plant's role as a valuable agricultural resource. Additionally, traditional cuisines around the world widely utilize grape leaves, contributing to their substantial economic value. However, diseases affecting grape leaves not only harm the plant and its yield but also render the leaves unsuitable for culinary use, leading to considerable economic losses for producers. Detecting diseases on grape leaves is a challenging and time-consuming task when performed manually. Thus, developing a deep learning-based model to automate the classification of grape leaf diseases is of critical importance. This study aims to classify the most common grape leaf diseases grape-scab (grape leaf blister mite) and downy mildew (grapevine downy mildew) alongside healthy leaves using deep learning techniques. Initially, we conducted a basic classification using pre-trained deep learning models. Subsequently, the Convolutional Block Attention Module (CBAM) and Squeeze-and-Excitation Networks (SE) were integrated into the most successful pre-trained classification model to enhance classification performance. As a result, the classification accuracy improved from 92.73% to 96.36%.
Why it matches plant phenotyping methodsブドウ葉の病害状態を画像から分類する深層学習手法が研究の中心であり、CBAM・SEの統合による性能向上も評価しているため、植物フェノタイピング手法として適格です。
abstractdeveloping a deep learning-based model to automate the classification of grape leaf diseases is of critical importance
The agro-industrial sector is experiencing a new wave of innovation driven by goods-inspecting devices designed to optimise operations, improve product quality, and reduce yield losses. Grape withering is widely used to concentrate berry juice for raisin and sweet wine production. While this process alters wine characteristics, it also introduces costs and risks, as pathogen infections can compromise the quality of the final product. This study presents a reliable three-dimensional analysis for assessing grape colour and bunch morphology to evaluate infection risk and drying performance. Twenty Vitis vinifera bunches were dried under environmental conditions. Colour analysis focused on the distribution of colours in healthy versus rotten bunches. Three-dimensional digital replicas were generated with two methods: i) photogrammetry, and ii) a recently developed artificial intelligence model. The point clouds and meshes output from the two approaches were compared, and morphometric traits were directly measured, including volume, surface area, and both horizontal and vertical sections for each bunch. Key geometrical descriptors of the bunch's horizontal sections and individual berries were found to be relevant for classifying the risk of bunch rot. Additionally, morphometric traits related to bunch compactness were linked to drying speed. A linear model incorporating three-dimensional descriptors was developed to estimate weight loss during withering, achieving an R² value of 0.98 and a relative error of 0.07. The artificial intelligence-based technique produced lower-quality models for grape reconstruction, but the selected morphometric traits remained effective.
Why it matches plant phenotyping methodsブドウ房の3D形態・色を画像から取得し、形態形質による腐敗リスクと乾燥性能の評価手法を開発・比較しているため、フェノタイピング手法が中心です。
abstractThis study presents a reliable three-dimensional analysis for assessing grape colour and bunch morphology to evaluate infection risk and drying performance.
Grapes are widely cultivated fruit crops, essential for fresh consumption, winemaking and dried product production. However, their yield and quality are significantly impacted by various fungal diseases. This paper provides a large dataset of 2,726 high-quality grape leaf disease images collected from grapes farm of Nashik, India in two years of span 2023 to 2025. The dataset is precisely annotated under the guidance and observation of agriculture domain expert and organized in a well-defined folder structure. The dataset captures the two major categories healthy leaves and unhealthy leaves, during cultivation period. A primary directory containing two main classes Heathy Leaf Images and Unhealthy Leaf images. Further unhealthy class is divided into three subfolders for disease class, namely Downy Mildew, Powdery Mildew and Bacterial Leaf Spot. These are the major fungal disease observed on grape crop causes substantially crop losses and ultimately impact on the yield production. Timely identification of these diseases can significantly reduce the risk of crop loss and help to improve quality of fruit with maximum yield production. This High-quality annotated image dataset can help to design standard advanced AI models for automated disease detection, classification, and prediction. The dataset was validated through a transfer learning approach using the ResNet-18 algorithm and demonstrated the remarkable classification accuracy of 96 % . These results validate the dataset's quality and its suitability for deep learning-based grape disease detection. Overall, this open-access resource provides a valuable foundation for computer vision, machine learning, and agricultural technology researchers aims to enhance disease management practices in grape production. thus, this is an effective source of data for future studies and real-world applications in sustainable grape production.
Why it matches plant phenotyping methodsブドウ葉の病徴・健全状態を画像で取得した注釈付きデータセットを提供し、分類モデルで検証しているため、植物病害表現型のデータセット開発・検証が中心です。
abstractThis paper provides a large dataset of 2,726 high-quality grape leaf disease images collected from grapes farm of Nashik, India in two years of span 2023 to 2025.
Reproduction assets foundThe paper is a data descriptor for the Niphad Grape Leaf Disease Dataset (NGLD), 2,726 annotated grape leaf images, publicly deposited on Mendeley Data with direct URL and DOI. No author analysis code is shared.Dataset · publicges were labelled sequentially for clear association within the dataset.
Data source location
Niphad Grapes farms, located at District Nashik 422209, MH-India
Longitude and Latitude: 20.0771° N, 74.1094° E
Data accessibility
Repository Name: Niphad Grape Leaf Disease Dataset (NGLD)
DOI: 10.17632/8nnd2ypcv3.5
Direct URL to Data: https://data.mendeley.com/datasets/8nnd2ypcv3/5
1.
Value of the Data
•
Comprehensive Dataset : The Dataset is comprehensive and consists of 2726 high-quality images, in four subfolder such as Downy Mildew, Powdery Mildew, Bacterial Leaf Spot and Healthy Grapes Leaf. Unlike existing public datasets that primarily focus on diseases such as Esca, Black Rot, and Leaf BligOpen asset ↗10.17632/8nnd2ypcv3.5lines:1-43Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
The present dataset is a collection of multispectral images designed for development of detection algorithms for grapevine diseases like Flavescence dorée (FD) and Esca (ED). Although FD severely threatens viticulture, there are few public datasets and none with multispectral data collected in the field. The collected images have been taken from a frontal perspective of vineyard plants that highlights details of leaves and trunks facilitating detailed disease analysis. The data were collected using a Micasense RedEdge-P multispectral camera, capturing six spectral bands across 172 image captures of three different grapevine varieties used in Lambrusco wines: Ancellotta, Marani, and Salamino. The dataset includes raw and processed images, calibration images for the multispectral camera, annotations detailing plant health conditions, and Python-based usage examples for researchers. Potential applications include the development of machine learning algorithms for automated disease detection, image alignment techniques, and background removal methods. The dataset is a valuable resource for advancing remote and proximal sensing in precision agriculture.
Why it matches plant phenotyping methodsブドウ樹の病害状態を対象とするマルチスペクトル画像データセットで、画像・校正・アノテーション・利用例を含む再利用可能な資源として構築されており、植物フェノタイピング手法の基盤が中心です。
titleA dataset for vineyard disease detection via multispectral imaging.
Reproduction assets foundThe paper is a data descriptor for a multispectral vineyard disease detection dataset deposited by the authors on Zenodo, including raw/processed images, annotations, and Python usage examples. The two Micasense GitHub repositories are generic vendor libraries, not paper-specific assets.Dataset · publicgio Emilia, Emilia-Romagna, Italy). It is managed by the RIMLab laboratory at the University of Parma, Parco Area delle Scienze 181/A, 43100 Parma, Italy.
Data accessibility
Repository name: A Dataset for Vineyard Disease Detection via Multispectral Imaging
Data identification number: 10.5281/zenodo.14936376
Direct URL to data: https://zenodo.org/records/14936376
Related research article
1
Value of the Data
•
The dataset features grapevine images which is a high-value plant used for wine production. Italy and other European nations are among the world's largest wine exporters. For this reason, diseases such as Flavescence Dorée (FD) and Esca, that cause severe damage to both the plant aOpen asset ↗Zenodo · 10.5281/zenodo.14936376lines:1-49Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Background Imaging sensors (e.g., multispectral cameras) mounted on unmanned aerial systems (UAS) have emerged as a powerful tool for deriving insights about agricultural fields, from plant morphology phenotyping to plant disease monitoring. Advances in computer vision-based image analysis have enabled researchers to rapidly and accurately isolate crop spectra in UAS images. Specialty crops often employ unique production styles, such as trellising or inter-cropping. This presents a barrier to using existing image processing methodologies developed for broad-acre, row cropped systems (i.e. corn, wheat, soybean). Here, we present MAUI, a customizable image processing workflow built for specialty crops. Using a pathology research vineyard and hemp breeding trial as test cases, MAUI streamlines the generation of multispectral orthomosaic time-series, the segmentation of crops at the unit of research interest, and the extraction of crop spectra for downstream analysis. Results We successfully used MAUI to collect and analyze UAS data at two field sites over two growing seasons. Of the five canopy segmentation methods we tested, a supervised deep convolutional neural network (DeepLabv3) and a vision foundation model (SAM) produced the most accurate crop masks for the vineyard and hemp images, with mean intersection over union (mIoU) values of 0.85 and 0.95, respectively. Segmentation accuracy decreased when we applied each method to the other dataset, highlighting the importance of modular, flexible segmentation workflows for UAS imaging analysis in specialty crops. Conclusion We present a modular framework to efficiently extract spectral data for specialty crops from UAS imagery. We highlight two kinds of segmentation applied to trellised and row cropping systems to demonstrate the modularity and versatility of the proposed methodology. MAUI improved spectral discrimination between individual plants and treatment groups for hemp and grapevine, respectively. With the containerized deployment package and open-source codebase, MAUI can be widely adopted by specialty crop researchers to facilitate the integration of UAS imagery analysis into routine research.
Why it matches plant phenotyping methods専門作物向けUAS画像から植物キャノピーを分割し、スペクトル形質を抽出するモジュール型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。
abstractHere, we present MAUI, a customizable image processing workflow built for specialty crops.
Grapes, highly nutritious and flavorful fruits, require adequate chlorophyll to ensure normal growth and development. Consequently, the rapid, accurate, and efficient detection of chlorophyll content is essential. This study develops a data-driven integrated framework that combines hyperspectral imaging (HSI) and convolutional neural networks (CNNs) to predict the chlorophyll content in grape leaves, employing hyperspectral images and chlorophyll a + b content data. Initially, the VGG16-U-Net model was employed to segment the hyperspectral images of grape leaves for leaf area extraction. Subsequently, the study discussed 15 different spectral preprocessing methods, selecting fast Fourier transform (FFT) as the optimal approach. Twelve one-dimensional CNN models were subsequently developed. Experimental results revealed that the VGG16-U-Net-FFT-CNN1-1 framework developed in this study exhibited outstanding performance, achieving an R2 of 0.925 and an RMSE of 2.172, surpassing those of traditional regression models. The t-test and F-test results further confirm the statistical robustness of the VGG16-U-Net-FFT-CNN1-1 framework. This provides a basis for estimating chlorophyll content in grape leaves using HSI technology.
Why it matches plant phenotyping methodsブドウ葉のクロロフィル含量という植物形質を、ハイパースペクトル画像、画像分割、スペクトル前処理、CNNで推定する手法を開発・評価しており、フェノタイピング手法が中心である。
abstractThis study develops a data-driven integrated framework that combines hyperspectral imaging (HSI) and convolutional neural networks (CNNs) to predict the chlorophyll content in grape leaves
Plant diseases significantly affect agricultural productivity by reducing both the quality and quantity of crops. The necessity for automated image‑based solutions stems from the labor‑intensive and subjectively error‑prone nature of traditional inspection methods performed by farmers or agricultural specialists. To maintain sustainable agriculture and prevent the spread of infections, the detection of plant leaf diseases should be performed early and accurately. Early identification of infections can also significantly reduce yield losses and minimize the excessive use of pesticides. Since leaf diseases frequently manifest as uneven texture patterns, spots, or distortions on the leaf surface, local texture capturing mechanisms have proven to be remarkably effective among many computational approaches. This study proposes a novel Deep Convolutional Neural Network (DCNN) to extract high‑level hidden feature representations from leaf images. To enhance performance, the deep features are combined with traditional handcrafted texture features known as the Uniform Local Binary Pattern (uLBP). The proposed model was trained and tested using three well‑known publicly available datasets: Apple Leaf, Tomato Leaf, and Grape Leaf. The model achieved test accuracies of 96%, 91%, and 96% on these datasets, respectively. The experimental results demonstrate that the proposed approach is an effective and practical method for early diagnosis of plant diseases. This system has potential for real‑world application by farmers and agricultural experts to support disease management and contribute to the development of more resilient crops and a sustainable agricultural industry.
Why it matches plant phenotyping methods植物葉画像から病害状態を推定する画像解析手法を提案・評価しており、病害表現型の取得・分類が研究の中心であるため。
abstractThis study proposes a novel Deep Convolutional Neural Network (DCNN) to extract high‑level hidden feature representations from leaf images.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 14 Sept 2026
Abstract Investigating the genetic architecture of important agronomic traits in grapevine, like berry quality and resilience to abiotic stress, has been hampered by bottlenecks in genotyping and phenotyping. To address these limitations, this study aimed to develop innovative tools to unravel the complex polygenic genomic architecture of these traits. Specifically, a high‐density 200K single nucleotide polymorphism array is developed and validated its effectiveness by genotyping 471 accessions from three F 1 breeding populations. A high‐throughput grape phenotyping tool is developed to accurately capture berry color, shape, and size. By integrating data from the two platforms, associated loci are identified over three growing seasons. Association mapping and haplotype analysis identified novel loci and candidate genes for berry shape ( bHLH017 ), soluble sugars ( ACT ), and organic acids ( ALMT1 and FUSC2 ), as well as vine cold tolerance ( NAC08 ), and fine‐mapped the flower sex determination locus. Furthermore, the functional role of NAC08 is validated, demonstrating that it activates the expression of a raffinose synthase gene, thereby increasing raffinose levels and conferring cold tolerance. Together, these augmented tools, the integrated data, and novel loci establish a better foundation for trait aggregation that will enhance breeding efficiency and boost the development of high‐quality grape varieties.
Why it matches plant phenotyping methodsブドウ果実の色・形・サイズを高精度に取得するハイスループット表現型解析ツールの開発が研究の中心であり、遺伝子型データとの統合応用も行っている。
abstractA high‐throughput grape phenotyping tool is developed to accurately capture berry color, shape, and size.
Abstract Crop pests significantly reduce crop yield and threaten global food security. Conventional pest control relies heavily on insecticides, leading to pesticide resistance and ecological concerns. However, crops and their wild relatives exhibit varied levels of pest resistance, suggesting the potential for breeding pest-resistant varieties. This study integrates deep learning (DL)/machine learning (ML) algorithms, plant phenomics, quantitative genetics, and transcriptomics to conduct genomic selection (GS) of pest resistance in grapevine. Building deep convolutional neural networks (DCNNs), we accurately assess pest damage on grape leaves, achieving 95.3% classification accuracy (VGG16) and a 0.94 correlation in regression analysis (DCNN-PDS). The pest damage was phenotyped as binary and continuous traits, and genome resequencing data from 231 grapevine accessions were combined in a Genome-Wide Association Studies, which maps 69 quantitative trait locus (QTLs) and 139 candidate genes involved in pest resistance pathways, including jasmonic acid, salicylic acid, and ethylene. Combining this with transcriptome data, we pinpoint specific pest-resistant genes such as ACA12 and CRK3, which are crucial in herbivore responses. ML-based GS demonstrates a high accuracy (95.7%) and a strong correlation (0.90) in predicting pest resistance as binary and continuous traits in grapevine, respectively. In general, our study highlights the power of DL/ML in plant phenomics and GS, facilitating genomic breeding of pest-resistant grapevine.
Why it matches plant phenotyping methodsブドウ葉の害虫被害を深層学習で画像評価し、分類・回帰性能を検証した植物フェノタイピング手法が中心である。
abstractBuilding deep convolutional neural networks (DCNNs), we accurately assess pest damage on grape leaves, achieving 95.3% classification accuracy (VGG16) and a 0.94 correlation in regression analysis (DCNN-PDS).
Reproduction assets foundThe authors publicly host the grape leaf image dataset, YOLO model weights, and all analysis scripts in a GitHub repository explicitly cited for data and code availability.Code · publicCode availability
All scripts performed in this study are available on Github: https://github.com/zhouyflab/Pest-Resistance .Open asset ↗zhouyflab/Pest-Resistancelines:183-224Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Abstract Background Horticultural crops propagated vegetatively are at risk of infection by wood-colonizing and vascular pathogens from infected cuttings. Diseases caused by such pathogens are difficult to diagnose. Because their chronic infections cannot be cured, disease diagnosis in the nursery could be an efficient approach to prevent spread to perennial plantings. Early detection of grapevine trunk diseases is confounded by a delay of up to a year before symptoms appear. This incubation period exceeds the 6 to 8 months grapevines are grown in the nursery; visual inspection for leaf symptoms is thus not a means of trunk-disease diagnosis. We evaluated hyperspectral imagery as a non-destructive means. Anatomical, physiological, and transcriptomic host responses occur within weeks of infection, and may be associated with changes in hyperspectral reflectance of asymptomatic leaves. If so, hyperspectral imagery might have promise in trunk-disease diagnosis in the nursery. For 14 weeks, we compared hyperspectral reflectance (410 to 1,000 nm) of asymptomatic leaves on potted plants, the woody stems of which were inoculated with fungi that cause trunk diseases Botryosphaeria dieback ( Neofusicoccum parvum ) and Esca ( Phaeomoniella chlamydospora and Tropicoporus texanus ), to those of non-inoculated controls. Results Destructive sampling of woody stems, at weeks 2, 8, and 14, confirmed the largest internal wood lesions in N . parvum -inoculated plants. Normalized difference spectral indices (NDSIs) of wavelengths in the visible (VIS) spectrum (e.g., 670 nm) and at the ‘red edge’ (700 – 730 nm) distinguished controls from inoculated plants, at weeks 8 and 9. By week 14, two pairs of treatments ( N . parvum and P. chlamydospora versus control and T. texanus ) were distinguished, based on separate Principal Component Analyses (PCAs) of the VIS and near-infrared (NIR) spectra, on the strength of associated NDSIs, and on overlap of their spectral curves. Partial least-squares discriminant analyses (PLS-DAs), under a 2-class model, identified VIS and NIR wavelengths that distinguished leaves of control plants versus each inoculation treatment, albeit with discriminant accuracies of 55 to 79%. Conclusions Further research is needed to substantiate the prospects of hyperspectral imaging as an early detection tool of grapevine trunk diseases with potential application at a commercial scale, under nursery conditions.
Why it matches plant phenotyping methodsブドウの無症状葉を対象に、ハイパースペクトル画像から感染状態を非破壊的に推定する手法を評価しており、植物病害表現型の取得・識別が中心である。
abstractWe evaluated hyperspectral imagery as a non-destructive means.
Grapevine phenotyping, that is the process of determining the physical properties (e.g., size, shape, and number) of grape bunches, provides valuable information for growth and health monitoring, yield estimation and efficient crop management in precision viticulture. Currently, grape bunch counting and sizing is done manually, which is labor intensive and often impractical for large-scale field applications. This paper describes a novel framework to automatically detect, count and estimate the volume/weight of grape bunches using RGB and depth data acquired in the field by a farmer robot. The proposed pipeline starts with the semantic segmentation of RGB images based on a pre-trained MANet architecture with EfficientnetB3 backbone to separate fruit from non-fruit regions. The segmented fruit mask is then projected onto the co-registered depth image to recover a depth mask, allowing for three-dimensional (3D) data association. After a pre-processing step to correct anomalies, such as corrupted and missing values, and to remove outliers, a depth gradient-based clustering algorithm is applied that detects individual grape bunch clusters. This enables the separation of adjacent and partially overlapping bunches. In addition, a method to reconstruct the whole 3D shape of a bunch is introduced, so as to provide an estimate of volume and weight. Experiments performed in a commercial vineyard in Italy are presented showing that, despite the low quality and high variability of the input images, the proposed approach is able to count grape bunch clusters with an average error of about 12% with respect to visual ground-truth and an average error less than 30% with respect to manual weight measurements. It is also shown that the processing framework can be applied to geo-referenced image sequences acquired by the farmer robot while traversing vineyard rows, thus providing an automated pipeline for the generation of high-resolution yield maps for precision viticulture applications.
Why it matches plant phenotyping methodsブドウ房の検出・計数・3D形状復元により体積・重量を推定する画像・深度ベースの表現型取得手法を開発し、精度検証も行っているため、方法が中心的である。
abstractThis paper describes a novel framework to automatically detect, count and estimate the volume/weight of grape bunches using RGB and depth data acquired in the field by a farmer robot.
Accurate detection and timely management of grapevine diseases, such as downy and powdery mildew, are essential for ensuring vineyard health and maximizing yield and quality. This study presents a novel approach using a ResNet50 model enhanced with batch normalization for precise identification and classification of grapevine leaf and fruit diseases. The dataset consists of 1,226 images categorized into five classes: Downy mildew diseased fruits (70 training, 18 testing), Downy mildew diseased leaves (534 training, 134 testing), Healthy leaves (183 training, 46 testing), Powdery mildew diseased fruits (93 training, 23 testing), and Powdery mildew diseased leaves (100 training, 25 testing). The model achieved an impressive accuracy of 95% in distinguishing between healthy and diseased grapevine leaves during rigorous evaluation and validation phases. Evaluation metrics—precision (94%), recall (96%), and F₁-score (95%)—highlight the model’s effectiveness compared to conventional methods. This research demonstrates the feasibility and superiority of deep learning in vineyard disease management, emphasizing its potential to revolutionize viticulture practices through automated, real-time disease detection and monitoring. These findings contribute to advancing agricultural sustainability and productivity through innovative technology applications in plant pathology.
Why it matches plant phenotyping methodsブドウ葉・果実の病害状態を画像から分類する深層学習手法を開発・評価しており、植物病害表現型の取得が研究の中心である。
abstractThis study presents a novel approach using a ResNet50 model enhanced with batch normalization for precise identification and classification of grapevine leaf and fruit diseases.
Bunch compactness (BC) is a complex, multi-trait characteristic that has been studied mostly in the context of wine grapes, with table grapes being scarcely considered. As these groups have marked phenotypic and genetic differences, including BC, the study of this trait is reported here using a genetically diverse collection of 116 Vitis vinifera L. cultivars and lines enriched for table grapes over two seasons. For this, 3D scanning-based morphological data were combined with ground measurements of 14 BC-related traits, observing high correlations among both approaches (R 2 > 0.90-0.97). The multivariate analysis suggests that the attributes 'berries per bunch', 'berry weight and width', and 'bunch weight and length' could be considered as the main descriptors for BC, optimizing evaluation times. Then, GWASs based on a set of 70,335 SNPs revealed that GBS analysis in this same population enabled the detection of several SNPs associated with different sub-traits, with a locus for 'berries per bunch' in chromosome (chr) 18 being the most prominent. Enrichment analysis of significant and frequent SNPs found simultaneously in several traits and seasons revealed the over-representation of discrete functions such as alpha-linolenic acid metabolism and glycan degradation. In summary, the utility of 3D automated phenotyping was validated for table grape backgrounds, and new SNPs and candidate genes associated with the BC trait were detected. The latter could eventually become a selection tool for grapevine breeding programs.
Why it matches plant phenotyping methodsブドウ房のコンパクトネスを対象に、3Dスキャンによる自動形態計測を地上測定と比較・検証し、育種利用可能な表現型評価法として実証しているため、方法が中心的です。
abstract3D scanning-based morphological data were combined with ground measurements of 14 BC-related traits, observing high correlations among both approaches (R 2 > 0.90-0.97).
Societal Impact Statement Characterizing variability in crop traits is key for understanding agroecosystem responses to environmental change. However, trait data are often time‐consuming to collect and therefore still limit our understanding and predictions of agriculture responses to environmental change. We tested the ability of reflectance spectroscopy—a high‐throughput technique—to rapidly amass trait data for multiple wine grape cultivars. Reflectance spectroscopy predicts important wine grape leaf traits including photosynthesis and biochemistry with a good degree accuracy, but in a fraction of the time compared to traditional techniques. Reflectance spectroscopy can therefore rapidly characterize wine grape phenotypes and, in doing so, inform predictions of how vines, clones and cultivars will respond to environmental change. Summary Reflectance spectroscopy has emerged as a powerful tool for non‐destructive and high‐throughput phenotyping in plants. While the ability of reflectance spectroscopy to predict traits across diverse plant species and ecosystems has received considerable attention, whether or not this technique is able to quantify within species trait variation—especially physiological traits—has been less extensively explored. Quantifying intraspecific variation in traits through reflectance spectroscopy is especially appealing in agroecology, where it may present an approach for better understanding crop performance, fitness and trait‐based responses to environmental conditions. We tested if reflectance spectroscopy coupled with partial least square regression (PLSR) predicts photosynthetic carbon assimilation ( A 420 ), RuBisCO carboxylation ( V cmax ) and electron transport ( J max ) rates, as well as leaf mass per area (LMA) and leaf nitrogen (N) concentrations, across six wine grape ( Vitis vinifera ) cultivars (Cabernet Franc, Cabernet Sauvignon, Merlot, Pinot noir, Viognier, Sauvignon blanc). PLSR models showed good capability in predicting intraspecific trait variation in wine grapes, explaining up to 55%, 58%, 62% and 62% of the variation in observed J max , V cmax , leaf N and LMA values, respectively. However, predictions of A 420 were less strong, with reflectance spectra explaining only up to 29% of the variation in this trait. Our results indicate that trait variation within species and crops is less well‐predicted by reflectance spectroscopy, than trait variation that exists among species. However, our results indicate that reflectance spectroscopy still presents a viable technique for quantifying trait variation in wine grapes specifically, and agroecosystems more broadly.
Why it matches plant phenotyping methods反射分光法とPLSRを用いてブドウ葉の光合成・生理・化学形質を非破壊推定し、予測性能を評価しており、植物表現型取得法が研究の中心である。
abstractReflectance spectroscopy has emerged as a powerful tool for non‐destructive and high‐throughput phenotyping in plants.
Aiming at the problems of low recognition rate of small target spots in grape leaf images and low detection accuracy due to low resolution of input images. In this paper, an improved recognition network based on YOLO v8 is constructed. In the constructed network, Spatial Pyramid Dilated Convolution (SPD-Conv) is used to replace each stepwise convolution layer and each pooling layer to better capture the detailed features of small targets. Meanwhile, the Efficient Multi-Scale Attention (EMA) Module is incorporated into the Neck part of YOLO v8 to make full use of the feature information of each detection layer and improve the accuracy of feature representation.The Plant Village dataset and the orchard image set are used to test the network performance of the improved model. The experimental test results show that the improved YOLO v8 has 92.64% precision, 93.28% recall and 96.17% AP. The size of model was a mere 7.1M. Compared to YOLO v8, the improvements are 2.38%, 1.91%, and 1.13%, respectively. Compared with the mainstream networks YOLO v4, YOLO v5, YOLO v6, and YOLO v7, precision is improved by 4.74%, 3.38%, 4.15%, and 4.69%, respectively. Therefore, the improved network proposed in this paper can improve the detection accuracy of small target objects and also identify the black rot disease of grape leaves more accurately.
Why it matches plant phenotyping methodsブドウ葉の黒腐病斑を画像から検出する改良YOLOv8を開発し、複数データセットで性能評価しており、植物病害状態の画像ベース表現型取得が中心である。
abstractan improved recognition network based on YOLO v8 is constructed
The agro-industrial sector is experiencing a new wave of innovation driven by goods-inspecting devices designed to optimise operations, improve product quality, and reduce yield losses. Grape withering is widely used to concentrate berry juice for raisin and sweet wine production. While this process alters wine characteristics, it also introduces costs and risks, as pathogen infections can compromise the quality of the final product. This study presents a reliable three-dimensional analysis for assessing grape colour and bunch morphology to evaluate infection risk and drying performance. Twenty Vitis vinifera bunches were dried under environmental conditions. Colour analysis focused on the distribution of colours in healthy versus rotten bunches. Three-dimensional digital replicas were generated with two methods: i) photogrammetry, and ii) a recently developed artificial intelligence model. The point clouds and meshes output from the two approaches were compared, and morphometric traits were directly measured, including volume, surface area, and both horizontal and vertical sections for each bunch. Key geometrical descriptors of the bunch's horizontal sections and individual berries were found to be relevant for classifying the risk of bunch rot. Additionally, morphometric traits related to bunch compactness were linked to drying speed. A linear model incorporating three-dimensional descriptors was developed to estimate weight loss during withering, achieving an R 2 value of 0.98 and a relative error of 0.07. The artificial intelligence-based technique produced lower-quality models for grape reconstruction, but the selected morphometric traits remained effective. • Grapevine bunches morphology evaluation through three-dimensional reconstruction. • Morphometric traits classified bunch dehydration speed and rot infection risk. • Colour analysis described withering progress and rotting spreading. • A fast artificial intelligence-based 3D acquisition technique was tested.
Why it matches plant phenotyping methodsブドウ房の3D画像再構成と色解析を用いて形態形質・感染リスク・乾燥性能を定量化し、 photogrammetryとAI手法を比較評価しているため、植物フェノタイピング手法が中心である。
abstractThis study presents a reliable three-dimensional analysis for assessing grape colour and bunch morphology to evaluate infection risk and drying performance.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 6 Sept 2026
Abstract Grape cluster compactness is a key trait that influence fruit quality, yield, and disease susceptibility. Understanding the genetic basis of this trait is essential for optimizing vineyard management and improving grapevine cultivars. In this study, we performed quantitative trait locus (QTL) mapping to identify genomic regions associated with cluster architecture and yield components in a bi-parental population derived from Vitis vinifera cv. Riesling × Cabernet Sauvignon. A total of 138 full-sibling progeny were evaluated over two growing seasons at Oakville, Napa Valley, California. Traditional yield-related traits were measured, including cluster number, total cluster weight, and average cluster weight. Additionally, an image-based phenotyping pipeline leveraging the foundation model Segment Anything Model (SAM) was employed to segment individual berries, measure their size and shape, and compute cluster compactness with minimal manual intervention. Trait correlations revealed that compact clusters tended to have a higher berry count but smaller berry size, highlighting the role of compactness in modulating cluster structure. Heritability estimates varied across traits, with berry dimensions and compactness displaying moderate to high heritability, indicating strong genetic control. Two parental linkage maps were constructed using a pseudo-test cross strategy. QTL mapping identified multiple loci associated with cluster architecture and yield components, with several stable QTLs detected across both years. Notably, a QTL for cluster compactness was found in both seasons on chromosome 1 in Cabernet Sauvignon. Other stable QTLs were associated with berry size (chromosomes 6 and 17) and berry count (chromosome 5 in Cabernet Sauvignon and chromosome 7 in Riesling). Additional QTLs were detected in a single year, reflecting the influence of environmental variation. Our findings provide valuable insights into the application of foundation models requiring no prior training and minimal intervention for high-quality segmentation and enhance our understanding of the genetic architecture of cluster compactness and yield traits. The genomic regions identified in this study offer promising targets for breeding programs aimed at improving grape quality and disease resistance.
Why it matches plant phenotyping methodsSAMを用いた画像ベースの果粒セグメンテーションと形状・サイズ・房のコンパクトネス推定が、遺伝解析のための主要な表現型取得手法として明示されているため。
abstractAdditionally, an image-based phenotyping pipeline leveraging the foundation model Segment Anything Model (SAM) was employed to segment individual berries, measure their size and shape, and compute cluster compactness with minimal manual intervention.
Smart farming technologies empower farmers to achieve sustainability by enabling data-driven decision-making. For instance, smart farming can contribute to farm sustainability through optimized pest control. By utilizing pest risk prediction models, farmers can conserve resources and minimize environmental impact by avoiding unnecessary treatments. However, effective crop pest control relies on timely treatments at specific phenological stages. Therefore, the ability to accurately predict phenological development becomes a crucial factor in increasing farm sustainability. This paper describes the design, development and evaluation of a set of Machine Learning models that predict the phenology of grapevines. The models were trained on multisourced data that combine 9 different datasets with different temporal and spatial resolutions. The authors evaluated and compared different machine learning algorithms to predict 9 different phenological stages of grapevines. The models that performed best also included data derived from Sentinel-2 images, which suggests that multispectral satellite images could be used to monitor and predict woody plant phenology. A key contribution of our proposal is the combination of multiple data sources and a fine-grained prediction aimed at distinguishing among 9 phenological states.
Why it matches plant phenotyping methodsブドウの9段階の生育フェノロジーを予測する機械学習手法を設計・開発・評価しており、植物状態の取得・推定が研究の中心であるため含める。
abstractThis paper describes the design, development and evaluation of a set of Machine Learning models that predict the phenology of grapevines.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 13 Sept 2026
Abstract Boron (B) is an essential micronutrient for grapevine growth, yet excessive levels can impair photosynthesis, reduce yields, and diminish fruit quality. In this study, we evaluated the potential of hyperspectral radiometry combined with machine learning to identify B-tolerant rootstocks rapidly and cost-effectively. We screened both commercial grapevine rootstocks and wild Vitis germplasm under B treatments ranging from 0.5 to 8 ppm, measuring leaf B accumulation, stomatal conductance, photosystem II efficiency, and leaf reflectance (R 380 –R 1100 nm). Our results revealed substantial genotypic variation in B exclusion, with some genotypes maintaining low leaf B content despite high external concentrations. Classification models (Partial Least Squares Discriminant Analysis and Random Forest classification) outperformed regression models (Partial Least Squares Regression and Random Forest regression) in distinguishing B-excluding genotypes, achieving moderate to high accuracy within just eight days after stress initiation. Vegetation indices such as Normalized Difference Vegetation Index (NDVI), Photochemical Reflectance Index (PRI), Structure Insensitive Pigment Index (SIPI), and Chlorophyll Index (CI) indicated that B stress reduces chlorophyll levels and may induce carotenoid accumulation, suggesting a photosynthetic tolerance mechanism. Although quantitative prediction of leaf B content proved more challenging, simulations showed that even modest prediction accuracies can substantially boost genetic gains if larger populations are screened, and selection intensities are increased. These findings underscore the value of hyperspectral radiometry for high-throughput phenotyping, allowing breeders to rapidly identify and advance B-tolerant rootstocks.
Why it matches plant phenotyping methodsブドウ台木のホウ素耐性を対象に、ハイパースペクトル計測と機械学習による表現型スクリーニング手法を開発・評価しており、表現型取得と分類が研究の中心である。
abstractwe evaluated the potential of hyperspectral radiometry combined with machine learning to identify B-tolerant rootstocks rapidly and cost-effectively.
Leaf chlorophyll content (LCC) is a key indicator for assessing the growth of grapes. Hyperspectral techniques have been applied to LCC research. However, quantitative prediction of grape LCC using this technique remains challenging due to baseline drift, spectral peak overlap, and ambiguity in the sensitive spectral range. To address these issues, two typical crop leaf hyperspectral data were collected to reveal the spectral response characteristics of grape LCC using standardization by variables (SNV) and multiple far scattering correction (MSC) preprocessing variations. The sensitive spectral range is determined by Pearson's algorithm, and sensitive features are further extracted within that range using Extreme Gradient Boosting (XGBoost), Recursive Feature Elimination (RFE), and Principal components analysis (PCA). Comparison of the prediction ability of Random Forest Regression (RFR) algorithm, Support Vector Machine Regression (SVR) model, and Genetic Algorithm-Based Neural Network (GA-BP) on grape LCC based on sensitive features. A SNV-RFE-GA-BP framework for predicting hyperspectral LCC in grapes is proposed, where [Formula: see text]=0.835 and NRMSE = 0.091. The analysis results show that SNV and MSC treatments improve the correlation between spectral reflectance and LCC, and different feature screening methods have a greater impact on the model prediction accuracy. It was shown that SNV-based processed hyperspectral data combined with GA-BP has great potential for efficient chlorophyll monitoring in grapevine. This method provides a new framework theory for constructing a hyperspectral analytical model of grapevine key growth indicators.
Why it matches plant phenotyping methodsブドウ葉のクロロフィル含量という植物形質を、ハイパースペクトル計測と前処理・特徴選択・機械学習で推定する手法を開発・比較しており、フェノタイピング手法が研究の中心である。
abstractTo address these issues, two typical crop leaf hyperspectral data were collected to reveal the spectral response characteristics of grape LCC using standardization by variables (SNV) and multiple far scattering correction (MSC) preprocessing variations.
Downy mildew is a critical disease in viticulture, typically identified through manual inspection of individual leaves in the field by experts. The combination of artificial intelligence techniques with mobile platforms can optimise non-invasive detection. This work focused on employing semantic segmentation deep neural networks to detect visual symptoms of downy mildew in high-resolution grapevine images under field conditions. Vineyard canopy images were collected from 14 plots using both manual and mobile platform methods. The study compared six architectures and six encoders using transfer learning, as well as two SegNet AdHoc architectures. To address imbalance problems, simple data augmentation, MixUp, oversampling, and undersampling techniques were employed. The results were adjusted through test-time augmentation. The study found that the U-Net architecture, using the MobileVit-S encoder and the Dice loss function, was particularly efficient. The U-Net architecture with light-weight encoders exhibited potential for real-time applications. The robustness of the model was improved by combining oversampling and undersampling with simple data augmentation during training. The classification of areas with and without disease symptoms achieved an accuracy of 86% and an f1-score of 82%. Additionally, the number of symptoms in grapevine canopy images was detected with an NRMSE of 12%. In conclusion, the proposed methodology shows promise for efficiently early assessing grapevine downy mildew under field conditions. This approach could be applied to other crop diseases and pests, taking advantage of the complexity of the dataset to strengthen the robustness of the model in real-world scenarios.
Why it matches plant phenotyping methodsブドウ葉・キャノピー画像から病徴領域と症状数を推定するセマンティックセグメンテーション手法が研究の中心であり、植物病害状態の表現型計測と技術評価に該当する。
abstractThis work focused on employing semantic segmentation deep neural networks to detect visual symptoms of downy mildew in high-resolution grapevine images under field conditions.
The research presented in this study offers a contribution to the field of viticulture by testing at lab scale an innovative approach for monitoring grape ripening using an autonomous proximal sensing technology. By leveraging an IoT spectral sensing system, termed i-Grape, the research aims to remotely monitor vineyards and provide real-time data on grape ripening status. This system, consisting of tailored optical, host, and controller modules, offers a novel solution for continuous monitoring throughout the crop season, overcoming limitations associated with traditional sampling methods. The study conducted comprehensive sampling in the viticulture area of the Douro Valley, collecting data from cv. Touriga Nacional and Touriga Franca. Both optical and wet-chemistry analyses were performed on the grape samples to develop predictive models for ripening parameters, including Total Soluble Solids (TSS), Potential Alcohol (PA), pH, Titratable Acidity (TA), Total Polyphenols (TP), and Extractable Anthocyanins (EA). Exploratory analysis of the optical data revealed insights into the behaviour of the spectral readouts over time, highlighting the evolution of grape ripening and the potential interference factors that need to be addressed for accurate modelling. Pre-processing techniques, including background subtraction and Log10 transformation, were employed to enhance the quality of the optical data and improve model performance. Overall, predictive PLS models with good performance were obtained for the estimation of the technological ripening parameters (RPD = 2.76 and R² = 0.86 for TSS; RPD = 2.58 and R² = 0.85 for PA; RPD = 3.65 and R² = 0.92 for TA; RPD = 2.27 and R² = 0.79 for pH), establishing a solid ground for the application of this sensing strategy in the field. For the phenolic parameters (TP and EA), the performance of the models is still insufficient (RPD = 1.28 and R² = 0.51 for TP; RPD = 1.55 and R² = 0.58 for EA). A comparison with existing literature highlighting the advancements achieved in terms of predictive performance and operational capabilities has been reported. The potential of the i-Grape system to revolutionize grape ripening monitoring by offering a cost-effective, non-destructive, and scalable solution for vineyard management has been demonstrated at lab scale. In conclusion, the research laid the groundwork for further advancements in optical sensing technology for viticulture, opening up avenues for future research in optimizing hardware design, data processing algorithms, and field implementation strategies to realize the full potential of IoT-based solutions in precision agriculture.
Why it matches plant phenotyping methodsブドウ果実の成熟形質を対象に、IoT分光センシングシステムと予測モデルを開発・評価しており、形質取得手法が研究の中心です。
abstracttesting at lab scale an innovative approach for monitoring grape ripening using an autonomous proximal sensing technology
In agricultural practices, plant phenotyping using object detection models is gaining attention, plant phenotyping is a technology that accurately measures the quality and condition of cultivated crops from images, contributing to the improvement of crop yield and quality, as well as reducing environmental impact. However, collecting the training data necessary to create generic and high-precision models is extremely challenging due difficulties associated with annotations and the diversity of domains. Such difficulties arise from the unique shapes and backgrounds of plants, as well as the significant changes in appearance due to environmental conditions and growth stages. Furthermore, it is difficult to transfer training data across different crops, and although machine learning models effective for specific environments, conditions, and crops have been developed, they cannot be widely applied in real-world conditions. Therefore, in this study, we propose a generative artificial intelligence data augmentation method (D4) and investigated its application towards a shoot detection task in a vineyard. D4 uses a pre-trained text-guided diffusion model based on a large number of original images culled from video data collected by unmanned ground vehicles or other means, and a small number of annotated datasets. The proposed method generates new annotated images with background information adapted to the target domain while retaining annotation information necessary for object detection. In addition, D4 overcomes the lack of training data in agriculture, including the difficulty of annotation and diversity of domains. We confirmed that this generative data augmentation method improved the mean average precision by up to 28.65% for the BBox detection task and the average precision by up to 13.73% for the keypoint detection task for vineyard shoot detection. D4 generative data augmentation is expected to simultaneously solve the cost and domain diversity issues of training data generation for agricultural applications and improve the generalization performance of detection models.
Why it matches plant phenotyping methodsブドウのシュート検出を対象に、テキスト誘導拡散モデルによるドメイン適応型データ拡張手法を開発・評価しており、画像から植物器官を抽出する方法が研究の中心である。
abstractTherefore, in this study, we propose a generative artificial intelligence data augmentation method (D4) and investigated its application towards a shoot detection task in a vineyard.
The rapid advancement of artificial intelligence has paved the way for innovative solutions in agriculture, particularly in crop disease detection. Diagnosing plant diseases often rely on manual inspection and expert knowledge that time-consuming and prone to errors. As agriculture faces increasing challenges from pests, diseases and climate change, there is a pressing require for efficient, automated systems to monitor crop health. In this manuscript, Development of an AI-Based Pyramid Convolutional Neural Network Model with ResNet and Coati Optimization for Multi-Crop, Multi-Disease Identification in Leaf Images is (PCNN-ResNet-COA) proposed. Initially the data is collected from Plant Disease Classification Merged Dataset. This dataset includes both healthy and diseased leaves for various crop types. The collected images are organized into categories based on the crop type such as crop, grape and soybean and disease condition healthy or various types of diseases. Then the categorized images are fed to Pyramid Convolutional Neural Network with Residual Network (PCNN-ResNet), for identifying and classifying the Leaf Images as Corn_healthy, Corn_northern_leaf_blight, Corn_gray_leaf_spot, Grape_healthy, Grape_leaf_blight, Grape_black_rot, Grape_black_measles, corn_common_rust, soybean_bacterial_blight, Soybean_downy_mildew, Soybean_mosaic_virus, Soybean_powdery_mildew, Soybean_healty, Soybean_rust, and Soybean_southern_blight. In general, PCNN-ResNet does not express any adaption of optimization methods for determining optimal parameters to assure precise detection and classification of Leaf Images. Coati Optimization Algorithm (COA) is proposed for improving the weight parameter of PCNN-ResNet classifier that accurately predicts crop yield. The proposed PCNN-ResNet-COA method is implemented and analyzed with help of performance metrics like accuracy, precision, F1-score, computational time is evaluated. The proposed ResNet-COA approach attains 18.97%, 24.57% and 32.68% higher accuracy and 19.84%, 24.93% and 31.62% lower computational time with existing method respectively.
Why it matches plant phenotyping methods葉画像から植物病害状態を自動分類する新規CNN手法を開発・評価しており、植物フェノタイピング手法が中心です。
titleDevelopment of an AI-Based Pyramid Convolutional Neural Network Model with ResNet and Coati Optimization for Multi-Crop, Multi-Disease Identification in Leaf Images
Reproduction assets foundThe paper's leaf-image disease classification experiments are built entirely on the public Kaggle 'Plant Disease Classification Merged Dataset' (88 classes, >76,000 images), which is explicitly cited with its Kaggle URL in the references. No author code, models, or checkpoints are reported as available.Dataset · publicThe input data are obtained from Plant Disease Classification Merged Dataset [17]. A huge number of images, at
least one healthy plant and one disease per plant, the most prevalent diseases, annotated images, laboratory and
field photographs, significant staple foods and the plant species with the highest worldwide production were the
self-imposed conditions for the dataset.Open asset ↗Plant Disease Classification Merged Datasetpdf-layout-page:4 lines:1-64Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Published8 Feb 20252025 International Conference on Emerging Systems and Intelligent Computing (ESIC)Cited by 0 · OpenAlex ↗
Plant infections destroy and impair the quality of crops, and the pesticides used to treat them pollute the soil, rendering it unfit for planting. Image processing and deep learning technologies may be used to identify disease spots on grape leaves. The detection's precision and efficiency, on the other hand, remain problems. In this paper, the performance of a few well-known CNN models implemented utilizing transfer learning, such as ResNet18, VGG16, and GoogleNet were compared with capsule network. The local entity characteristics are first found by the Capsule layers. For the purpose of collecting aggregate data like "disease type" and "disease stage," the capsule layer makes use of geographic information and the frequency of local-level characteristics. Colored pictures of strong and unhealthy leaves were taken from the public dataset and then used to train the models. The proposed work focused on the novelty concept of identifying the stages of diseases. The Capsule based classification technique can give competitive benefits with an effective classification of accuracy and stages of leaf disease related to other models.
Why it matches plant phenotyping methodsブドウ葉の画像から病害の種類と進行段階という植物の病態を推定する深層学習手法を開発・比較しており、表現型取得・抽出が中心です。
abstractImage processing and deep learning technologies may be used to identify disease spots on grape leaves.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 13 Sept 2026
ABSTRACT Water scarcity is a major threat to crop production and quality. Improving drought tolerance through variety selection requires a deeper understanding of plant ecophysiological responses, but large-scale phenotyping remains a bottleneck. This study assessed the potential of high-throughput tools (spectroscopy and poro-fluorometry) to predict leaf morphological and ecophysiological traits in a grapevine diversity panel grown in pots under well-watered outdoor conditions and under three contrasting soil water treatments in a greenhouse. We found a certain complementarity between measuring devices. Spectrometers could accurately predict leaf mass per area, water content, and water quantity (R² > 0.58), while the poro-fluorometer was efficient for predicting net CO₂ assimilation (R² > 0.72), regardless of the water treatment. The prediction of leaf mass per area using spectrometers appeared to be quite robust across both outdoor and greenhouse experiments, while the prediction of water use efficiency was dependent on the water treatment, with much better predictions under moderate (R² > 0.73) than severe water deficit. Calibrated models were then applied to the full diversity panel using only high-throughput measurements to estimate trait values and their broad-sense heritability. Leaf mass per area, also measured directly, showed similar heritability whether based on observed or predicted data. Heritability estimates for predicted traits reached up to 0.5. Overall, our findings support the use of spectroscopy and poro-fluorometry as reliable, non-destructive tools for high-throughput phenotyping, enabling genetic studies on drought-related traits in grapevine.
Why it matches plant phenotyping methods分光法とポロフルオロメトリーによる葉の形態・生理形質の高スループット推定を開発・評価し、予測精度と頑健性を検証しているため、植物フェノタイピング手法が中心である。
abstractThis study assessed the potential of high-throughput tools (spectroscopy and poro-fluorometry) to predict leaf morphological and ecophysiological traits
Viticulture benefits significantly from rapid grape bunch identification and counting, enhancing yield and quality. Recent technological and machine learning advancements, particularly in deep learning, have provided the tools necessary to create more efficient, automated processes that significantly reduce the time and effort required for these tasks. On one hand, drone, or Unmanned Aerial Vehicles (UAV) imagery combined with deep learning algorithms has revolutionised agriculture by automating plant health classification, disease identification, and fruit detection. However, these advancements often remain inaccessible to farmers due to their reliance on specialized hardware like ground robots or UAVs. On the other hand, most farmers have access to smartphones. This article proposes a novel approach combining UAVs and smartphone technologies. An AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training. By leveraging UAV-captured data for training, the proposed model not only accelerates the detection process but also enhances the accuracy and adaptability of grape bunch detection across different devices, surpassing the efficiency of traditional and purely UAV-based methods. To this end, using a dataset of UAV videos recorded during early growth stages in July (BBCH77-BBCH79), the X-Decoder segments vegetation in the front of the frames from their background and surroundings. X-Decoder is particularly advantageous because it can be seamlessly integrated into the AI pipeline without requiring changes to how data is captured, making it more versatile than other methods. Then, YOLO is trained using the videos and further applied to images taken by farmers with common smartphones (Xiaomi Poco X3 Pro and iPhone X). In addition, a web app was developed to connect the system with mobile technology easily. The proposed approach achieved a precision of 0.92 and recall of 0.735, with an F1 score of 0.82 and an Average Precision (AP) of 0.802 under different operation conditions, indicating high accuracy and reliability in detecting grape bunches. In addition, the AI-detected grape bunches were compared with the actual ground truth, achieving an R 2 value as high as 0.84, showing the robustness of the system. This study highlights the potential of using smartphone imaging and web applications together, making an effort to integrate these models into a real platform for farmers, offering a practical, affordable, accessible, and scalable solution. While smartphone-based image collection for model training is labour-intensive and costly, incorporating UAV data accelerates the process, facilitating the creation of models that generalise across diverse data sources and platforms. This blend of UAV efficiency and smartphone precision significantly cuts vineyard monitoring time and effort.
Why it matches plant phenotyping methodsスマートフォン画像とUAVデータを用いてブドウ房を検出・計数するAIパイプラインを開発・評価し、実測値との比較も行っているため、植物器官形質の取得手法が中心である。
abstractAn AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training.
Reproduction assets foundThe paper's Data Availability Statement points to a public, paper-specific dataset (EscaYard: geotagged smartphone vineyard images, phytosanitary status, UAV 3D point clouds and orthomosaics) published as a Data Brief with a DOI, which directly underpins the smartphone/UAV grape detection phenotyping analysis. No code,Dataset · publicData is available at https://doi.org/10.1016/j.dib.2024.110497 [ 55 ].Open asset ↗10.1016/j.dib.2024.110497lines:202-204Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Feb 2025Computers and Electronics in Agriculture.
The digital techniques, spreading across the agriculture sector, allow access to helpful information from fields, crops, and routine operations. Additionally, artificial intelligence is essential to automate and optimise laborious activities. The present study is focused on estimating the diameter of the grapevine (Vitis vinifera L. cv. Tempranillo) trunks through several artificial intelligence approaches. Then, benefits and constraints were investigated for the in-field application. Several RGB images of vines were acquired under different conditions from two fields planted with vines of different ages. The same high-resolution camera was used to take RGB images in two acquisition modes, manually and on-the-go. Additionally, different camera angles were compared to detect any distortion effect on the analysis. Finally, the impact of some common disturbance factors was compared. The trunk measurement through image analysis followed two phases. First, a YOLOv4 algorithm was set up to detect all the trunks. Then, two main approaches using deep learning were considered to estimate the diameter of each trunk. Four convolutional neural networks were trained to estimate the trunk’s diameter through a regression model. On the other hand, two semantic segmentation models were trained to localise the pixels of the trunk, and a tape in the trunk marked the diameter measurement. The uncertainty analysis detected the most deleterious disturbance factors. The actual and predicted diameter regression was compared among the approaches. Xception was proved as the most accurate architecture for the estimation through a regression model. The semantic segmentation model showed a higher R-squared value than the Xception, which resulted in 0.842 and 0.619, respectively. The semantic segmentation model’s normalised root mean square error was lower than the Xception-based regression one, 0.071 and 0.098, respectively. Moreover, the light condition, the vine age, and the acquisition mode showed relevant interference for the trunk diameter estimation. The trunk diameter is a key factor for monitoring the vines’ vigour and the reserve stocks. An automated trunk diameter estimation would be essential for mapping the vineyard variability at a large scale.
Why it matches plant phenotyping methodsブドウ樹幹径という植物形質を、画像取得・物体検出・深層学習回帰・セマンティックセグメンテーションで推定する手法を開発・比較し、誤差や外乱要因も評価しているため、方法が中心である。
abstractThe present study is focused on estimating the diameter of the grapevine (Vitis vinifera L. cv. Tempranillo) trunks through several artificial intelligence approaches.
Viticulture benefits significantly from rapid grape bunch identification and counting, enhancing yield and quality. Recent technological and machine learning advancements, particularly in deep learning, have provided the tools necessary to create more efficient, automated processes that significantly reduce the time and effort required for these tasks. On one hand, drone, or Unmanned Aerial Vehicles (UAV) imagery combined with deep learning algorithms has revolutionised agriculture by automating plant health classification, disease identification, and fruit detection. However, these advancements often remain inaccessible to farmers due to their reliance on specialized hardware like ground robots or UAVs. On the other hand, most farmers have access to smartphones. This article proposes a novel approach combining UAVs and smartphone technologies. An AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training. By leveraging UAV-captured data for training, the proposed model not only accelerates the detection process but also enhances the accuracy and adaptability of grape bunch detection across different devices, surpassing the efficiency of traditional and purely UAV-based methods. To this end, using a dataset of UAV videos recorded during early growth stages in July (BBCH77-BBCH79), the X-Decoder segments vegetation in the front of the frames from their background and surroundings. X-Decoder is particularly advantageous because it can be seamlessly integrated into the AI pipeline without requiring changes to how data is captured, making it more versatile than other methods. Then, YOLO is trained using the videos and further applied to images taken by farmers with common smartphones (Xiaomi Poco X3 Pro and iPhone X). In addition, a web app was developed to connect the system with mobile technology easily. The proposed approach achieved a precision of 0.92 and recall of 0.735, with an F1 score of 0.82 and an Average Precision (AP) of 0.802 under different operation conditions, indicating high accuracy and reliability in detecting grape bunches. In addition, the AI-detected grape bunches were compared with the actual ground truth, achieving an R² value as high as 0.84, showing the robustness of the system. This study highlights the potential of using smartphone imaging and web applications together, making an effort to integrate these models into a real platform for farmers, offering a practical, affordable, accessible, and scalable solution. While smartphone-based image collection for model training is labour-intensive and costly, incorporating UAV data accelerates the process, facilitating the creation of models that generalise across diverse data sources and platforms. This blend of UAV efficiency and smartphone precision significantly cuts vineyard monitoring time and effort.
Why it matches plant phenotyping methodsスマートフォン画像とUAVデータを用いてブドウ房を検出・計数するAIパイプラインを開発し、実測値との比較で性能検証しているため、植物形質取得法が中心である。
abstractAn AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training.
Quantifying crop responses to increasing temperatures is critical for predicting the productivity and sustainability of agricultural systems under environmental change. Physiological trait data associated with Rubisco carboxylation ( V cmax ) and electron transport ( J max ) rates are especially important predictors of crop response to elevated temperatures. However, when generating V cmax and J max data, steady-state methods of gas exchange measurements are time-consuming; thus, non-steady-state methods have been developed to obtain these measurements faster, prospectively allowing for trait data collection of considerably more varieties of crops. Globally important and geographically widespread vineyards are of particular interest due to the high economic value and the susceptibility of these managed systems to climate warming, especially in Canada, where the annual rate of warming far exceeds global averages. In this study, we examined the efficacy of the high-throughput, non-steady-state dynamic assimilation technique (DAT) for obtaining V cmax and J max data from wine grapes. Specifically, we measured V cmax and J max (alongside leaf nitrogen [N] concentrations and leaf mass per unit area [LMA]) across seven of the world’s most common wine grape ( Vitis vinifera L.) varieties, namely, Cabernet franc, Cabernet sauvignon, Merlot, Pinot noir, Riesling, Sauvignon blanc, and Viognier. Our results show that V cmax and J max estimates derived from the DAT were strongly correlated to those obtained through steady-state methods ( r 2 =0.748 and 0.908, respectively), and J max did not differ significantly between the two methods. Additionally, leaf N explained 43-46% and 56-58% of the variation in V cmax and J max , respectively, across both methods. Our results suggest the DAT represents a viable tool for rapidly estimating intraspecific variation in important physiological traits and allows for increased replication and the inclusion of additional varieties when evaluating the responses of wine grape and other crops to elevated temperatures.
Why it matches plant phenotyping methods作物の生理形質（V cmax、J max）を迅速に取得する動的同化法を開発・検証し、定常状態法との比較で妥当性を評価しているため、植物フェノタイピング手法が中心である。
abstractwe examined the efficacy of the high-throughput, non-steady-state dynamic assimilation technique (DAT) for obtaining V cmax and J max data from wine grapes.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 6 Sept 2026
In agriculture, promptly and accurately identifying leaf diseases is crucial for sustainable crop production. To address this requirement, this research introduces a hybrid deep learning model that combines the visual geometric group version 19 (VGG19) architecture features with the transformer encoder blocks. This fusion enables the accurate and précised real-time classification of leaf diseases affecting grape, bell pepper, and tomato plants. Incorporating transformer encoder blocks offers enhanced capability in capturing intricate spatial dependencies within leaf images, promising agricultural sustainability and food security. By providing farmers and farming stakeholders with a reliable tool for rapid disease detection, our model facilitates timely intervention and management practices, ultimately leading to improved crop yields and mitigated economic losses. Through extensive comparative analyses on various datasets and filed tests, the proposed depth wise separable convolutional-TransNet (DSC-TransNet) architecture has demonstrated higher performance in terms of accuracy (99.97%), precision (99.94%), recall (99.94), sensitivity (99.94%), F1-score (99.94%), AUC (0.98) for Grpae leaves across different datasets including bell pepper and tomato. Furthermore, including DSC layers enhances the computational efficiency of the model while maintaining expressive power, making it well-suited for real-time agricultural applications. The developed DSC-TransNet model is deployed in NVIDIA Jetson Nano single board computer. This research contributes to advancing the field of automated plant disease classification, addressing critical challenges in modern agriculture and promoting more efficient and sustainable farming practices.
Why it matches plant phenotyping methods葉画像から植物病害を分類する深層学習モデルを開発し、複数データセットで比較評価するとともにエッジデバイスへ実装しており、植物病害状態の取得・推定手法が研究の中心である。
abstractthe proposed depth wise separable convolutional-TransNet (DSC-TransNet) architecture has demonstrated higher performance in terms of accuracy (99.97%), precision (99.94%), recall (99.94), sensitivity (99.94%), F1-score (99.94%), AUC (0.98)
Assessing the health status of vegetation is of vital importance for all stakeholders. Multi-spectral and hyper-spectral imaging systems are tools for evaluating the health of vegetation in laboratory settings, and also hold the potential of assessing vegetation of large portions of land. However, the literature lacks benchmark datasets to test algorithms for predicting plant health status, with most researchers creating tailored datasets. This work presents a dataset composed of multi-spectral images, hyper-spectral reflectance values, and measurements of weight, chlorophyll, and nitrogen content of leaves at five different drying stages, from avocado, olive, and grape trees, which are common crops in the Valparaíso region of Chile. This dataset is a valuable asset for developing tools in the field of precision agriculture and assessing the general health status of vegetation.
Why it matches plant phenotyping methods植物のマルチスペクトル・ハイパースペクトル画像と葉の水分状態・化学形質を含む評価用データセットを構築しており、フェノタイピング手法開発のためのベンチマークが中心である。
abstractThis work presents a dataset composed of multi-spectral images, hyper-spectral reflectance values, and measurements of weight, chlorophyll, and nitrogen content of leaves at five different drying stages
Reproduction assets foundThe paper's multispectral images, hyperspectral reflectance, and trait measurements (weight, chlorophyll, nitrogen, fuel moisture) are publicly deposited on Figshare with an explicit DOI. The authors' sample Matlab code is included within that dataset. The MicaSense imageprocessing repository is a generic third-party工具Dataset · publicAll the data is available at this repository DOI: https://doi.org/10.6084/m9.figshare.26950660.v2.Open asset ↗figshare · 10.6084/m9.figshare.26950660.v2pdf-page:14 lines:1-35Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Climate change poses fundamental challenges to viticulture, such as more frequent droughts in Central Europe. This development requires precise, site-specific methods to determine plant water status. Especially in steep sloped vineyards, the spatial variability of drought stress can be high and depends on different factors such as slope, aspect and soil characteristics. Most established methods for determining plant water status are destructive, labor-intensive, or provide point-in-time measurements, or e.g. non-destructive modeling approaches need to be well referenced. UAV campaigns using thermal and multispectral imagery, as well as in-field sensor networks, provide non-destructive solutions with high spatio-temporal resolution. This study aims to combine both solutions to measure the high spatial and temporal variability of drought stress in a steep sloped vineyard. The goal is to develop a continuous, cross-scale, and resource-efficient method that can be used directly for irrigation scheduling or as a reference method for cross-scale modeling approaches at high spatial resolution. During the growing season of 2022, UAV campaigns were conducted every two weeks to generate thermal and multispectral imagery over a vineyard of 1 ha in Saxony, Germany. The vineyard was divided into five management zones (MZ), which differ in terms of slope, aspect, soil characteristics and grape varieties. A monitoring system has been established in each management zone to continuously collect data on local climate, as well as soil and plant water properties. Simultaneously with the UAV campaigns, the water status and physiological stage of the vines were determined as reference measurements. Therefore, predawn leaf water potential (Ψpd) was measured using a Scholander pressure chamber. Based on the processed aerial images and the in-situ sensor-based measurements the Crop Water Stress Index (CWSI) was computed and then validated by comparing it’s values to in-field reference measurements such as soil water status and Ψpd. Weather and plant physiological in-situ measurements were also integrated into a grapevine water balance model to derive quantitative information on plant and soil water status. In-situ measurements of plant and soil water potentials correlated well with the results of the modeling approach. This was a good representation of the spatial heterogeneity of the vineyard, especially the differences in plant water availability between MZs. CWSI values from the UAV campaigns will be compared with the in-situ measurements in terms of spatial variability, and also temporal variability to reproduce drought and other seasonal events. The combination of sensor data, simulation modeling and UAV-based thermal and multispectral imagery offers great potential to provide site-specific information with high spatio-temporal resolution about the plant water status. In particular, the inclusion of UAV campaigns can help to optimize the implemented sensor network and minimize the number of in-situ reference measurements. However, this cross-scale method also depends on a large number of influencing factors that need to be considered and discussed in depth in order to allow a valid assessment of drought stress dynamics and to set thresholds for irrigation or other management measures.
Why it matches plant phenotyping methodsUAV熱・マルチスペクトル画像、センサーネットワーク、水収支モデルを統合してブドウの水分状態を推定する手法を開発し、圃場基準測定で検証している。植物表現型の取得・推定が研究の中心である。
abstractThe goal is to develop a continuous, cross-scale, and resource-efficient method that can be used directly for irrigation scheduling or as a reference method for cross-scale modeling approaches at high spatial resolution.
Assessing vines' vigour is essential for vineyard management and automatization of viticulture machines, including shaking adjustments of berry harvesters during grape harvest or leaf pruning applications. To address these problems, based on a standardized growth class assessment, labeled ground truth data of precisely located grapevines were predicted with specifically selected Machine Learning (ML) classifiers (Random Forest Classifier (RFC), Support Vector Machines (SVM)), utilizing multispectral UAV (Unmanned Aerial Vehicle) sensor data. The input features for ML model training comprise spectral, structural, and texture feature types generated from multispectral orthomosaics (spectral features), Digital Terrain and Surface Models (DTM/DSM- structural features), and Gray-Level Co-occurrence Matrix (GLCM) calculations (texture features). The specific features were selected based on extensive literature research, including especially the fields of precision agri- and viticulture. To integrate only vine canopy-exclusive features into ML classifications, different feature types were extracted and spatially aggregated (zonal statistics), based on a combined pixel- and object-based image-segmentation-technique-created vine row mask around each single grapevine position. The extracted canopy features were progressively grouped into seven input feature groups for model training. Model overall performance metrics were optimized with grid search-based hyperparameter tuning and repeated-k-fold-cross-validation. Finally, ML-based growth class prediction results were extensively discussed and evaluated for overall (accuracy, f1-weighted) and growth class specific- classification metrics (accuracy, user- and producer accuracy).
Why it matches plant phenotyping methodsUAVマルチスペクトル画像、特徴抽出、画像セグメンテーション、機械学習、交差検証を統合し、ブドウの樹勢・樹冠体積という植物形質を推定する手法が研究の中心である。
abstractbased on a standardized growth class assessment, labeled ground truth data of precisely located grapevines were predicted with specifically selected Machine Learning (ML) classifiers
Seedlessness in table grapes is a desirable trait for consumers. Plant growth regulators (PGRs) have been extensively utilized to induce seedlessness. However, the efficacy of these PGRs is not uniformly successful. In addition, the seedlessness is difficult to detect by cutting and counting technique. The shortwave-near infrared spectroscopy (SW-NIRS), coupled with suitable chemometric analysis, is a non-destructive method for sorting and prediction of seedlessness grapes. The NIRS is higher efficiency than original technique in term of accuracy, measuring time and waste reduction.•The SW-NIR spectra of 240 grape berries were recorded. Each reflectance spectrum was acquired in the wavenumber of 3996-12,489 cm -1 . After that all grape berries were cut and count for seedlessness sorting.All spectral together with seedlessness sorting were be analysis by chemometrics.•The NIR spectral data were analyzed using principal component analysis (PCA). In addition, supervised self-organizing map (SSOM) and quadratic discriminant analysis (QDA) were applied to classify the seedlessness.•The PCA results represented a negative tendency to classify the seedlessness. Clear classification tendency can be obtained from SOMs. Good predictive results from SSOM were obtained, as it gave a percentage correctly classified of 97.14 and 94.64% for training and test sample sets, respectively.
Why it matches plant phenotyping methodsブドウ果実の種なし形質をSW-NIRSとケモメトリクスで非破壊推定・分類する手法が研究の中心であり、精度評価も行っている。
abstractThe shortwave-near infrared spectroscopy (SW-NIRS), coupled with suitable chemometric analysis, is a non-destructive method for sorting and prediction of seedlessness grapes.
The hairiness of the leaves is an essential morphological feature within the genus Vitis that can serve as a physical barrier. A high leaf hair density present on the abaxial surface of the grapevine leaves influences their wettability by repelling forces, thus preventing pathogen attack such as downy mildew and anthracnose. Moreover, leaf hairs as a favorable habitat may considerably affect the abundance of biological control agents. The unavailability of accurate and efficient objective tools for quantifying leaf hair density makes the study intricate and challenging. Therefore, a validated high-throughput phenotyping tool was developed and established in order to detect and quantify leaf hair using images of single grapevine leaf discs and convolution neural networks (CNN). We trained modified ResNet CNNs with a minimalistic number of images to efficiently classify the area covered by leaf hairs. This approach achieved an overall model prediction accuracy of 95.41%. As final validation, 10,120 input images from a segregating F1 biparental population were used to evaluate the algorithm performance. ResNet CNN-based phenotypic results compared to ground truth data received by two experts revealed a strong correlation with R values of 0.98 and 0.92 and root-mean-square error values of 8.20% and 14.18%, indicating that the model performance is consistent with expert evaluations and outperforms the traditional manual rating. Additional validation between expert vs. non-expert on six varieties showed that non-experts contributed to over- and underestimation of the trait, with an absolute error of 0% to 30% and -5% to -60%, respectively. Furthermore, a panel of 16 novice evaluators produced significant bias on set of varieties. Our results provide clear evidence of the need for an objective and accurate tool to quantify leaf hairiness.
Why it matches plant phenotyping methodsブドウ葉の毛密度という形態形質を画像とCNNで自動定量する高スループット手法を開発し、専門家評価および大規模集団で検証しており、表現型取得・抽出法が研究の中心である。
abstractTherefore, a validated high-throughput phenotyping tool was developed and established in order to detect and quantify leaf hair using images of single grapevine leaf discs and convolution neural networks (CNN).
Reproduction assets foundThe authors publicly released the ResNet CNN training code, the leaf disc image datasets, and the full leaf hair quantification pipeline in a GitHub repository, directly reproducing this paper's phenotyping analysis.Code · publicAll datasets and the code to train the CNNs are available in the GitHub repository.Open asset ↗lines:91-99Dataset · publicThe script of the ResNet CNN along with the images are available in the GitHub repository: https://github.com/1708nagarjun/ResNet-CNN-Leaf-hair.Open asset ↗1708nagarjun/ResNet-CNN-Leaf-hairlines:143-183Plant phenotyping relevance match · UnverifiedbioRxiv · checked 15 Sept 2026
Cold hardiness is a crucial physiological parameter that determines the survival of grapevines during the dormant season. Accurate modeling and large-scale prediction of grapevine cold hardiness are essential for assessing the potential geographic distribution of grapevine cultivation, quantifying the impact of climate change on grapevine habitats, and ensuring the sustainability of the grape and wine industries in cool climate regions worldwide. However, until now, no comprehensive database has been available. In this research, we combined advanced automated machine learning techniques with extensive historical and current weather data to create an integrative database for grapevine cold hardiness: VineColD (https://cornell-tree-fruit-physiology.shinyapps.io/VineColD/). We developed the NYUS.2.1 model, an automated machine learning-based system for predicting grapevine cold hardiness and in this study, applied it to global historical weather data from 17,985 curated weather stations spanning 30{degrees} to 55{degrees} in both hemispheres from 1960 to 2024, resulting in the development of an integrative grapevine cold hardiness database and monitoring system. VineColD integrates both a global historical dataset and a daily updated regional cold hardiness system, offering a comprehensive resource to study grape cold hardiness for 54 grapevine cultivars. The platform provides multiple download options, from single-station data to complete datasets, and the interactive multi-functional R Shiny application facilitates data analysis and visualization. VineColD delivers critical insights into the impact of climate change on grapevine cultivation and supports a range of analytical functions, making it a valuable tool for grape growers and researchers.
Why it matches plant phenotyping methodsブドウの耐寒性という植物生理形質を予測する機械学習モデルを開発し、全球データベースと監視プラットフォームとして提供しており、形質推定手法と再利用可能な基盤が研究の中心である。
abstractWe developed the NYUS.2.1 model, an automated machine learning-based system for predicting grapevine cold hardiness
Grapevines are subjected to many physiological and environmental stresses that influence their vegetative and reproductive growth. Water stress, cold damage, and pathogen attacks are highly relevant stresses in many grape-growing regions. Precision viticulture can be used to determine and manage the spatial variation in grapevine health within a single vineyard block. Newer technologies such as remotely piloted aircraft systems (RPASs) with remote sensing capabilities can enhance the application of precision viticulture. The use of remote sensing for vineyard variation detection has been extensively investigated; however, there is still a dearth of literature regarding its potential for detecting key stresses such as winter hardiness, water status, and virus infection. The main objective of this research is to examine the performance of modern remote sensing technologies to determine if their application can enhance vineyard management by providing evidence-based stress detection. To accomplish the objective, remotely sensed data such as the normalized difference vegetation index (NDVI) and thermal imaging from RPAS flights were measured from six commercial vineyards in Niagara, ON, along with the manual measurement of key viticultural data including vine water stress, cold stress, vine size, and virus titre. This study verified that the NDVI could be a useful metric to detect variation across vineyards for agriculturally important variables including vine size and soil moisture. The red-edge and near-infrared regions of the electromagnetic reflectance spectra could also have a potential application in detecting virus infection in vineyards.
Why it matches plant phenotyping methodsRPASのマルチスペクトル・熱画像を用いてブドウの水分ストレス、低温ストレス、サイズ、ウイルス感染を推定・検証することが研究の中心であり、植物形質・状態の取得手法を評価している。
abstractThe main objective of this research is to examine the performance of modern remote sensing technologies to determine if their application can enhance vineyard management by providing evidence-based stress detection.
In the agricultural sector, pesticides are used to prevent disease transmission and protect crop yields. However, due to the diverse range of diseases, the human observation can often lead to misidentification. It is essential for a timely and precise disease classification approach without human intervention. Classifying the plant leaf diseases with an automated system is the significant need in this scenario. In this work, a hybrid classification model for the categorisation of plant leaf diseases is presented. Preprocessing, segmentation, feature extraction and classification of leaf diseases are the four steps in this method. In this work, crops such as grapes and mango are considered. Primarily, preprocessing the input image by utilising Gaussian filtering methods, which enhances the quality of image. The filtered image is then put through a segmentation process using the MBIRCH framework. The segmented image is then used to extract a number of features, including GLCM, ILGBHS, colour, shape and deep features using the VGG16 and AlexNet networks. Following the procedure, the hybrid model—which combines Bi‐GRU and DCNN with TL—is applied to the acquired features, and the final classified result is determined by the enhanced fusion score method.
Why it matches plant phenotyping methods植物葉の病害状態を画像から抽出・分類する前処理、分割、特徴抽出、分類のワークフローが研究の中心であり、植物病害フェノタイピング手法に該当する。
abstractIn this work, a hybrid classification model for the categorisation of plant leaf diseases is presented.
Climate change significantly impacts viticulture by harming plant and fruit growth, resulting in lower quality and storage issues. Therefore, there is growing scientific interest in the carbon fluxes of vineyard management activities, including efforts to measure carbon capture and storage in annual biomass. Precision Viticulture and Machine Vision techniques can help assess variations in vines’ biomass, which relate to the vines’ carbon balance. The present study examined the feasibility of a light detection and ranging (LiDAR) for vines’ (Vitis vinifera L. cv. Riesling) annual biomass reconstruction and its role in the annual carbon cycle. The leaves dry weight showed a high correlation coefficient of R² = 0.87 with the LiDAR-based leaf area estimation. Thus, making the proposed sensing system reliable for biomass elemental carbon assessment. Nevertheless, no significant correlation was found for the monitoring of leaf area/fruit ratio. The proposed study showcases the potential and the limits of LiDAR-based vine biomass assessment.
Why it matches plant phenotyping methodsLiDARによるブドウ樹の葉面積・年間バイオマス再構成を評価し、乾物重との相関でセンシング手法の信頼性を検証しているため、植物形質取得法が中心である。
abstractThe present study examined the feasibility of a light detection and ranging (LiDAR) for vines’ (Vitis vinifera L. cv. Riesling) annual biomass reconstruction and its role in the annual carbon cycle.
In agricultural practices, plant phenotyping using object detection models is gaining attention, plant phenotyping is a technology that accurately measures the quality and condition of cultivated crops from images, contributing to the improvement of crop yield and quality, as well as reducing environmental impact. However, collecting the training data necessary to create generic and high-precision models is extremely challenging due difficulties associated with annotations and the diversity of domains. Such difficulties arise from the unique shapes and backgrounds of plants, as well as the significant changes in appearance due to environmental conditions and growth stages. Furthermore, it is difficult to transfer training data across different crops, and although machine learning models effective for specific environments, conditions, and crops have been developed, they cannot be widely applied in real-world conditions. Therefore, in this study, we propose a generative artificial intelligence data augmentation method (D4) and investigated its application towards a shoot detection task in a vineyard. D4 uses a pre-trained text-guided diffusion model based on a large number of original images culled from video data collected by unmanned ground vehicles or other means, and a small number of annotated datasets. The proposed method generates new annotated images with background information adapted to the target domain while retaining annotation information necessary for object detection. In addition, D4 overcomes the lack of training data in agriculture, including the difficulty of annotation and diversity of domains. We confirmed that this generative data augmentation method improved the mean average precision by up to 28.65% for the BBox detection task and the average precision by up to 13.73% for the keypoint detection task for vineyard shoot detection. D4 generative data augmentation is expected to simultaneously solve the cost and domain diversity issues of training data generation for agricultural applications and improve the generalization performance of detection models. • Proposed novel data augmentation method D4 using text-guided diffusion model. • Analyzed detection accuracy using D4 for BBox detection and keypoint detection. • D4 improved detection accuracy and demonstrated effectiveness of domain adaptation. • D4 uses automatic selection with DreamSim to maintain generated image quality. • D4 overcomes lack of training data in agriculture.
Why it matches plant phenotyping methodsブドウのシュートを画像から検出する植物フェノタイピング向けのデータ拡張・ドメイン適応手法を開発し、検出精度を検証しており、表現型取得手法が中心である。
abstractTherefore, in this study, we propose a generative artificial intelligence data augmentation method (D4) and investigated its application towards a shoot detection task in a vineyard.
In agriculture, plant disease detection at an early stage is significant. Plant leaves are an important factor in plant disease detection. Early disease prevention in-line harvest loss benefit is given to plant growers. The odd leaves can be visible after getting infected, and if the software can tell accurately the disease infestation result. Forecasting plant leaf disease works better for an early disease warning. The infected leaf needs to be permanently removed, or it will affect the plant. As leaves are a significant part and produce food from the sun, the infected leaves show different patterns. Many proposing such AI and ML models have forecasted the infected leaves of many plants. This document focuses on forecasting diseases in plant leaves mainly for four types of plants: tomato, potato, apple & grape. Diseases in plant leaves can cause significant damage to plants and can have a detrimental effect on both the quantity and quality of production whether we consider a lower scale kitchen gardening or a high scale agriculture farming. The research begins with an overview of the common plant leaf diseases found in each of these four plants, along with their symptoms and causes. It then discusses the importance of early detection and management of these diseases to ensure healthy growth. This helps a naïve gardener as well as an experienced farmer to prevent a plant or a whole field from getting infected or diseased. Plant disease detection approaches are crucial for prevention and management. Various techniques in plant disease detection using AI- based machine learning and deep learning methods are reviewed comprehensively. Keywords— Agriculture farming, Kitchen Garden, Plant leaf diseases, Forecasting diseases.
Why it matches plant phenotyping methods植物葉の病徴・病害をAI/機械学習で検出・予測する手法を包括的にレビューしており、植物の病害状態を観測・推定するフェノタイピング手法が中心です。
titleA Comprehensive Review of Research on Disease Prediction in Plant Leaves
Grapevines ( Vitis vinifera L.) are one of the most economically relevant crops worldwide, yet they are highly vulnerable to various diseases, causing substantial economic losses for winegrowers. This systematic review evaluates the application of remote sensing and proximal tools for vineyard disease detection, addressing current capabilities, gaps, and future directions in sensor-based field monitoring of grapevine diseases. The review covers 104 studies published between 2008 and October 2024, identified through searches in Scopus and Web of Science, conducted on 25 January 2024, and updated on 10 October 2024. The included studies focused exclusively on the sensor-based detection of grapevine diseases, while excluded studies were not related to grapevine diseases, did not use remote or proximal sensing, or were not conducted in field conditions. The most studied diseases include downy mildew, powdery mildew, Flavescence dorée , esca complex, rots, and viral diseases. The main sensors identified for disease detection are RGB, multispectral, hyperspectral sensors, and field spectroscopy. A trend identified in recent published research is the integration of artificial intelligence techniques, such as machine learning and deep learning, to improve disease detection accuracy. The results demonstrate progress in sensor-based disease monitoring, with most studies concentrating on specific diseases, sensor platforms, or methodological improvements. Future research should focus on standardizing methodologies, integrating multi-sensor data, and validating approaches across diverse vineyard contexts to improve commercial applicability and sustainability, addressing both economic and environmental challenges.
Why it matches plant phenotyping methodsブドウの病害という植物状態を対象に、リモートセンシング・近接センシングによる検出手法を体系的に評価するレビューであり、フェノタイピング手法が中心です。
abstractThis systematic review evaluates the application of remote sensing and proximal tools for vineyard disease detection
Purpose Diagnosing the crop diseases by farmers accurately with the naked eye can be challenging. Timely identification and treating these diseases is crucial to prevent complete destruction of the crops. To overcome these challenges, in this work a light-weight automatic crop disease detection system has been developed, which uses novel combination of residual network (ResNet)-based feature extractor and machine learning algorithm based classifier over a real-time crop dataset. Design/methodology/approach The proposed system is divided into four phases: image acquisition and preprocessing, data augmentation, feature extraction and classification. In the first phase, data have been collected using a drone in real time, and preprocessing has been performed to improve the images. In the second phase, four data augmentation techniques have been applied to increase the size of the real-time dataset. In the third phase, feature extraction has been done using two deep convolutional neural network (DCNN)-based models, individually, ResNet49 and ResNet41. In the last phase, four machine learning classifiers random forest (RF), support vector machine (SVM), logistic regression (LR) and eXtreme gradient boosting (XGBoost) have been employed, one by one. Findings These proposed systems have been trained and tested using our own real-time dataset that consists of healthy and unhealthy leaves for six crops such as corn, grapes, okara, mango, plum and lemon. The proposed combination of Resnet49-SVM and ResNet41-SVM has achieved accuracy of 99 and 97%, respectively, for the images that have been collected from the city of Kurukshetra, India. Originality/value The proposed system makes novel contribution by using a newly proposed real time dataset that has been collected with the help of a drone. The collected image data has been augmented using scaling, rotation, flipping and brightness techniques. The work uses a novel combination of machine learning methods based classification with ResNet49 and ResNet41 based feature extraction.
Why it matches plant phenotyping methodsドローン画像から植物の健康状態・病害を推定する画像解析システムの開発が研究の中心であり、特徴抽出、分類、データセット構築と性能評価を含むため。
abstracta light-weight automatic crop disease detection system has been developed, which uses novel combination of residual network (ResNet)-based feature extractor and machine learning algorithm based classifier over a real-time crop dataset.
GrapevineField / plotRootMorphology / geometry measurement2D/3D reconstructionGrowth / time-series analysisRoot system architectureWater status / transpiration
Understanding root system architecture (RSA) is essential for improving crop resilience to climate change, yet assessing root systems of woody perennials under field conditions remains a challenge. This study introduces a pipeline that combines field excavation, in situ 3-dimensional digitization, and transformation of RSA data into an interoperable format to analyze and model the growth and water uptake of grapevine rootstock genotypes. Eight root systems of each of 3 grapevine rootstock genotypes ("101-14", "SO4", and "Richter 110") were excavated and digitized 3 and 6 months after planting. We validated the precision of the digitization method, compared in situ and ex situ digitization, and assessed root loss during excavation. The digitized RSA data were converted to root system markup language (RSML) format and imported into the CPlantBox modeling framework, which we adapted to include a static initial root system and a probabilistic tropism function. We then parameterized it to simulate genotype-specific growth patterns of grapevine rootstocks and integrated root hydraulic properties to derive a standard uptake fraction (SUF) for each genotype. Results demonstrated that excavation and in situ digitization accurately reflected the spatial structure of root systems, despite some underestimation of fine root length. Our experiment revealed significant genotypic variations in RSA over time and provided new insights into genotype-specific water acquisition capabilities. Simulated RSA closely resembled the specific features of the field-grown and digitized root systems. This study provides a foundational methodology for future research aimed at utilizing RSA models to improve the sustainability and productivity of woody perennials under changing climatic conditions.
Why it matches plant phenotyping methods圃場での根系3次元デジタル化、精度検証、データ形式変換、モデル化を統合した根系表現型取得・解析パイプラインが研究の中心である。
abstractThis study introduces a pipeline that combines field excavation, in situ 3-dimensional digitization, and transformation of RSA data into an interoperable format to analyze and model the growth and water uptake of grapevine rootstock genotypes.
Abstract Assessing the health status of vegetation is of vital importance for all stakeholders. Multi-spectral and hyper-spectral imaging systems are tools for evaluating the health of crops across large areas, particularly when deployed on robotic platforms such as unmanned aerial vehicles (UAVs). However, the literature lacks benchmark datasets to test algorithms for predicting plant health status, with most researchers creating tailored datasets. This work presents a dataset composed of multi-spectral images, hyper-spectral reflectance values, and measurements of weight, chlorophyll, and nitrogen content of leaves at five different drying stages, from avocado, olive, and vineyard trees, which are common crops in the Valparaíso region of Chile. This dataset is a valuable asset for developing tools in the field of precision agriculture and assessing the general health status of vegetation.
Why it matches plant phenotyping methods植物の健康状態を推定するためのマルチスペクトル・ハイパースペクトル画像と葉の形質測定を組み合わせた評価用データセットが主題であり、植物フェノタイピング手法のベンチマーク資源に該当する。
abstractThis work presents a dataset composed of multi-spectral images, hyper-spectral reflectance values, and measurements of weight, chlorophyll, and nitrogen content of leaves at five different drying stages
Reproduction assets foundThe paper is a dataset descriptor; its complete plant-phenotyping measurements (multispectral leaf images, hyperspectral reflectance, chlorophyll, nitrogen, weight/FMC across five drying stages for avocado, olive, and vineyard) are publicly deposited on figshare under DOI 10.6084/M9.FIGSHARE.26950660, along with aMatlåDataset · publicAll the data is available at this repository DOI: 10.6084/M9.FIGSHARE.26950660Open asset ↗figshare · 10.6084/M9.FIGSHARE.26950660pdf-page:15 lines:1-59Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Dec 2024Computers and Electronics in Agriculture.
This work presents a framework based on convolutional neural networks (CNNs) to estimate root traits (length, diameter, and color) from minirhizotron (MR) imagery. The proposed framework uses a set of reusable sub-network modules to compose different networks for object (i.e., root) detection and attribute (i.e., trait) estimation for per-root and per-image root phenotyping tasks. It provides a solution without requiring root segmentation. The first step in per-root phenotyping involves detecting the roots in the image; the traits of each detected root are then estimated. Per-image root phenotyping estimates aggregated root trait values, including total root length (TRL), mean root diameter, and percentage of white root. Regression-based and objects' points-detection-based variations are demonstrated for both per-root and per-image root trait estimation. Five network architectures are presented, two of which were previously used for TRL estimation (and are now evaluated for estimating mean root diameter and white root percentage), and three of which are new. The proposed framework is demonstrated on an annotated grapevine root dataset comprising 531 images, made publicly available as part of this paper. All images were acquired in situ using an MR system and annotated with Rootfly software. Regression-based modules used for individual detected roots yielded errors of 8.8%, 15.5%, and 23.5% for color, length, and diameter, respectively. The points-detection-based modules resulted in errors of 9.1%, 14.9%, and 25.0% for the same parameters. The image-level estimates showed errors of 11.5%-16.5% for white root percentage, 13.7%-16.0% for TRL, and 17.6%-22.1% for mean root diameter. We demonstrate that aggregating per-root estimations of diameter and color obtained with the new suggested architectures improves the per-image estimations of these traits relative to the direct per-image estimation that does not include per-root estimations. To demonstrate further the practicality of the suggested framework in deriving the vertical distribution of various root traits, an additional dataset of 132 root images from two different grapevine graft combinations was annotated (and also made publicly available as part of this paper). In this dataset, the per-image root traits were estimated for different soil depths and visually compared with human annotation results.
Why it matches plant phenotyping methodsCNNによるミニリゾトロン画像からの根形質推定フレームワークを開発し、複数のネットワーク構成、誤差評価、公開データセットで検証しており、植物フェノタイピング手法が研究の中心である。
abstractThis work presents a framework based on convolutional neural networks (CNNs) to estimate root traits (length, diameter, and color) from minirhizotron (MR) imagery.
Optimization of water inputs is possible through precision irrigation based on prescription maps. The crop water stress index (CWSI) is an indicator of spatial and dynamic changes in plant water status that can serve irrigation management decision-making. The driving hypothesis was that in-season CWSI maps based on combined static and spatial-dynamic variables could be used to delineate irrigation MZs. A primary incentive was to minimize thermal-imaging campaigns and to complement CWSI maps between campaigns with cost-effective multi-spectral imaging campaigns producing normalized difference vegetative index (NDVI) maps. A spatial machine-learning model based on a random-forest (RF) algorithm combined with spatial statistical methods was developed to predict the spatial and temporal variability in CWSI of single vines in a vineyard. Model criteria and objectives included the reduction of sample data and input variables to a minimum without impacting prediction accuracy, consideration of only variables readily available to farmers, and accounting for spatial location and spatial processes. The model was developed and tested on data from a ‘Cabernet Sauvignon’ vineyard in Israel over two years. Prediction of CWSI was driven by terrain parameters, slope, aspect and topographical wetness index, soil apparent electrical conductivity (ECa), and NDVI. Spatial models based on RF were found to support CWSI prediction. Adding a geospatial component significantly improved model performance and accuracy, particularly when raw data was represented as z-scores or when z-scores were used as weights. NDVI, followed by ECa, aspect, or slope, was the most important variable predicting CWSI in the non-spatial models. The stronger the variable importance of NDVI, the better the model performed. The weaker the effect of NDVI in predicting CWSI, the stronger the effect of terrain and soil variables. In the spatial models, based on z-transformed values or on weighted values, the most important variable in predicting CWSI was either NDVI or location. The model, based on a limited and readily accessible number of variables, can serve as the basis for user-friendly decision support tools for precision irrigation. Additional research is needed to evaluate alternative prediction variables and to account for case studies in more geographical locations to address overfitting specific input data. Socio-economic and cost-benefit considerations should be integrated to examine whether precision irrigation management based on such models has the desired effects on water consumption and yield.
Why it matches plant phenotyping methods単一ブドウ樹の水分状態（CWSI）を推定する空間機械学習モデルを開発・検証しており、植物表現型の取得・推定手法が研究の中心である。
abstractA spatial machine-learning model based on a random-forest (RF) algorithm combined with spatial statistical methods was developed to predict the spatial and temporal variability in CWSI of single vines in a vineyard.
Infrared spectroscopy provides extensive spectral and chemical information for plant material. However, the raw spectral data of fresh grapevine organs are poorly understood. This study investigated the spectral properties of grapevine shoots, leaves, and berries collected throughout the growing season. Near infrared (NIR) spectral analysis was performed using a solid probe (NIR-SP) and a rotating integrating sphere (NIR-RS) on samples collected from multiple cultivars and vintages from commercial growing sites at various phenological stages. Clustering patterns of the spectral properties were investigated using principal component analysis (PCA), and unsupervised self-organising maps (SOM) as a novel approach. Separation based on grapevine organ type was seen using both PCA and unsupervised SOM. Thereafter, unsupervised SOM analysis showed organ-specific differences between phenological stages (berries and shoots) and lignification (shoots). Supervised SOM was performed for classification purposes as a novel application on viticultural data for the prediction of grapevine organ and accurate prediction of organ type at 90.3% was observed for the NIR-SP dataset collected from 2019 to 2021. The prediction of individual phenological stages proved more challenging, but when phenological stages were grouped together, a prediction of 85.6% for the NIR-RS grape berry dataset was found. The prediction of shoot lignification yielded accurate results of 74.4% and 89.9% for the NIR-SP 2019–2021 and NIR-RS 2020–2021 datasets, respectively. Furthermore, spectral variable selection with orthogonal partial least squares discriminant analysis (OPLS-DA) and S-plots improved the discrimination of shoots and leaves up to 14% for the NIR-RS 2020–2021 dataset. Lignification predictions also improved by 9.8% for the NIR-SP shoot 2019–2021 (92.3%) and 12.7% for the NIR-RS shoot 2020–2021 (95.9%) datasets. Clustering methods, especially unsupervised SOM, showed important groupings in the raw spectral data for fresh grapevine organs. Organ type, phenological stage, and lignification were highlighted for showing prominent separation over multiple vintages.
Why it matches plant phenotyping methodsNIR分光とSOM、PCA、OPLS-DAを用いて、ブドウの器官、フェノロジー段階、シュートの木化を推定・分類する手法を中心的に開発・評価している。
abstractNear infrared (NIR) spectral analysis was performed using a solid probe (NIR-SP) and a rotating integrating sphere (NIR-RS)
GrapevineField / plotStem / branchPhysiological trait estimationWater status / transpiration
Efficient water management is pivotal for viticulture sustainability. Decision support tools can advise on how to optimize irrigation or on the feasibility of growing grapes in rainfed conditions, but reliable algorithms for assessing vine water status are required. In this context, the aim of the current study was to upgrade a soil water balance model specific for vineyards by incorporating meteorological, soil and vine vigor in equations that transform the fraction of transpirable soil water into midday stem water potential (Ψₛₜₑₘ). The model's sensitivity to variations in the magnitude of input parameters was analyzed. Furthermore, the model was tested in a broad scope of Spanish vineyards with different grapevine cultivars (both red and white), rootstocks, plant age, soil and climatic conditions, and water regimes, totaling 129 scenarios. The model was only slightly sensitive to variations in the magnitude of most inputs, except for the fraction of transpirable water at which leaf stomatal conductance begin to decline. Moreover, the model satisfactorily reproduced the evolution of Ψₛₜₑₘ over the growing season, although it slightly overestimated the measured Ψₛₜₑₘ values, as the slopes of the fitted regression lines were lesser than 1 on most occasions, 76 out of 129. Nonetheless, the coefficients of determination for these relationships were greater than 0.9, except for 21 datasets. Mean errors averaged 0.024 ± 0.015 MPa, while root mean square errors averaged 0.27 ± 0.01 MPa. The index of agreement was greater than 0.75 in 51 datasets, with only three datasets showing an index of agreement lower than 0.5. Nevertheless, the deviations between observed and simulated Ψₛₜₑₘ values did not alter the classification of the water stress undergone by grapevines. This upgraded model could constitute the core of a decision support system for water management in vineyards, applicable to both rainfed and irrigated conditions.
Why it matches plant phenotyping methodsブドウの茎水ポテンシャルという植物生理状態を推定する土壌水分収支モデルを改良し、129条件で感度分析と技術検証を行っており、フェノタイピング手法が研究の中心である。
abstractthe aim of the current study was to upgrade a soil water balance model specific for vineyards by incorporating meteorological, soil and vine vigor in equations that transform the fraction of transpirable soil water into midday stem water potential (Ψₛₜₑₘ).
GrapevineLeafPhysiological trait estimationWater status / transpiration
Quantifying drought tolerance in crops is critical for agriculture management under environmental change, and drought response traits in grape vine have long been the focus of viticultural research. Turgor loss point (π tlp ) is gaining attention as an indicator of drought tolerance in plants, though estimating π tlp often requires the construction and analysis of pressure-volume (P-V) curves which are very time consuming. While P-V curves remain a valuable tool for assessing π tlp and related traits, there is considerable interest in developing high-throughput methods for rapidly estimating π tlp , especially in the context of crop screening. We tested the ability of a dewpoint hygrometer to quantify variation in π tlp across and within 12 clones of grape vine (Vitis vinifera subsp. vinifera) and one wild relative (Vitis riparia), and compared these results to those derived from P-V curves. At the leaf-level, methodology explained only 4-5% of the variation in π tlp while clone/species identity accounted for 39% of the variation, indicating that both methods are sensitive to detecting intraspecific π tlp variation in grape vine. Also at the leaf level, π tlp measured using a dewpoint hygrometer approximated π tlp values (r 2 = 0.254) and conserved π tlp rankings from P-V curves (Spearman's ρ = 0.459). While the leaf-level datasets differed statistically from one another (paired t-test p = 0.01), average difference in π tlp for a given pair of leaves was small (0.1 ± 0.2 MPa (s.d.)). At the species/clone level, estimates of π tlp measured by the two methods were also statistically correlated (r 2 = 0.304), did not deviate statistically from a 1:1 relationship, and conserved π tlp rankings across clones (Spearman's ρ = 0.692). The dewpoint hygrometer (taking ∼ 10-15 min on average per measurement) captures fine-scale intraspecific variation in π tlp , with results that approximate those from P-V curves (taking 2-3 h on average per measurement). The dewpoint hygrometer represents a viable method for rapidly estimating intraspecific variation in π tlp , and potentially greatly increasing replication when estimating this drought tolerance trait in grape vine and other crops.
Why it matches plant phenotyping methodsブドウの乾燥耐性形質である膨圧損失点を、露点 hygrometer で高スループットに推定する方法を開発・P-V曲線と比較検証しており、表現型取得法が研究の中心です。
titleA high-throughput approach for quantifying turgor loss point in grapevine.
• UAV-based localization and mapping methods have been benchmarked in vineyards. • Five evaluation metrics were developed for agricultural scenarios. • Lighting variation impacts point cloud resolution. • Deep learning enhances SLAM for efficient plant phenotyping. UAVs equipped with various sensors offer a promising approach for enhancing orchard management efficiency. Up-close sensing enables precise crop localization and mapping, providing valuable a priori information for informed decision-making. Current research on localization and mapping methods can be broadly classified into SfM, traditional feature-based SLAM, and deep learning-integrated SLAM. While previous studies have evaluated these methods on public datasets, real-world agricultural environments, particularly vineyards, present unique challenges due to their complexity, dynamism, and unstructured nature. To bridge this gap, we conducted a comprehensive study in vineyards, collecting data under diverse conditions (flight modes, illumination conditions, and shooting angles) using a UAV equipped with high-resolution camera. To assess the performance of different methods, we proposed five evaluation metrics: efficiency, point cloud completeness, localization accuracy, parameter sensitivity, and plant-level spatial accuracy. We compared two SLAM approaches against SfM as a benchmark. Our findings reveal that deep learning-based SLAM outperforms SfM and feature-based SLAM in terms of position accuracy and point cloud resolution. Deep learning-based SLAM reduced average position error by 87% and increased point cloud resolution by 571%. However, feature-based SLAM demonstrated superior efficiency, making it a more suitable choice for real-time applications. These results offer valuable insights for selecting appropriate methods, considering illumination conditions, and optimizing parameters to balance accuracy and computational efficiency in orchard management activities.
Why it matches plant phenotyping methodsブドウ園でのUAV画像によるSfM・SLAM手法を比較検証し、植物レベルの空間精度や点群完全性などを評価しており、植物フェノタイピングの取得・解析基盤が中心である。
abstractDeep learning enhances SLAM for efficient plant phenotyping.
Traditional methods for assessing vine water status, such as the Scholander pressure chamber, are time-consuming, punctual and labour-intensive. The development of alternative methods which are accurate, reliable and can provide real-time information on vine water status is a necessity for farmers all over the world. This study proposes the use of plant electrophysiology as a novel approach for real-time water status assessment in grapevines. We conducted four climate chamber experiments with potted grapevines under different irrigation regimes. Various morphological and physiological assessments were performed in parallel with electrophysiological measurements to correlate classic water status assessment methods with plant electrophysiological signals. Two machine learning approaches based on classification and regression were employed to train the prediction models. Results obtained from both models indicate significant differences in irrigation status between well-watered and water-deficit plants, with the latter showing reduced growth and physiological activity, confirming the water stress status of the plant. While the binary classification model successfully differentiates between well-watered and water-deficit plants, its practical use is limited. Therefore, a regression model was developed to directly predict predawn leaf water potential. To the best of our knowledge, this is the first time that electrical signals are correlated with vine water potential measurements. The findings presented here thus provide a promising new tool for future real-time and remote monitoring of vine water status to manage irrigation and adapt agronomic strategies. Nevertheless, validation and optimisation of the models are still necessary, particularly under field conditions.
Why it matches plant phenotyping methodsブドウの水分状態という植物生理形質を、植物電気生理信号と機械学習でリアルタイム推定する手法を開発・検証しており、表現型取得が研究の中心である。
abstractThis study proposes the use of plant electrophysiology as a novel approach for real-time water status assessment in grapevines.
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 7 Sept 2026
Modern viticulture faces significant challenges including climate change and increasing crop diseases, necessitating sustainable solutions to reduce fungicide use and mitigate soil health risks, particularly from copper accumulation. Advances in plant phenomics are essential for evaluating and tracking phenotypic traits under environmental stress, aiding in selecting resilient vine varieties. However, current methods are limited, hindering effective integration with genomic data for breeding purposes. Remote sensing technologies provide efficient, non-destructive methods for measuring biophysical and biochemical traits of plants, offering detailed insights into their physiological and nutritional state, surpassing traditional methods. Smart phenotyping is essential for selecting crop varieties with desired traits, such as pathogen-resilient vine varieties, tolerant to altered soil fertility including copper toxicity. Identifying plants with typical copper toxicity symptoms under high soil copper levels is straightforward, but it becomes complex with supra-optimal, already toxic, copper levels common in vineyard soils. This can induce multiple stress responses and interferes with nutrient acquisition, leading to ambiguous visual symptoms. Characterizing resilience to copper toxicity in vine plants via smart phenotyping is feasible by relating smart data with physiological assessments, supported by trained professionals who can identify primary stressors. However, complexities increase with more data sources and uncertainties in symptom interpretations. This suggests that artificial intelligence could be valuable in enhancing decision support in viticulture. While smart technologies, powered by artificial intelligence, provide significant benefits in evaluating traits and response times, the uncertainties in interpreting complex symptoms (e.g., copper toxicity) still highlight the need for human oversight in making final decisions.
Why it matches plant phenotyping methods植物フェノタイピングとリモートセンシングを中心に、環境ストレス下の形質評価手法と将来展望を論じるレビューであり、方法論的役割が明確です。
abstractAdvances in plant phenomics are essential for evaluating and tracking phenotypic traits under environmental stress
Diseases and pests in agriculture significantly impact crop yield and quality. Downy mildew (Plasmopara viticola) is a particular noteworthy example in grapevines. Traditional detection methods are laborious, subjective and time-consuming. Consequently, a technological solution based on artificial intelligence, would provide higher levels of reproducibility and sampling. The aim of this work was to develop an interpretable, automated method for detection and localisation of plant disease symptoms under field conditions. Images of the grapevine canopy were taken in 14 commercial vineyard plots under a range of lightning conditions, including both static and on-the-go settings. The images were processed using a sliding window, classifying sub-images into areas with and without downy mildew. Transfer learning, fine-tuning and data augmentation were employed to automate the classification, comparing convolutional neural networks (CNNs) and vision transformers (ViT). Subsequently, the trained model was integrated into the sliding window to localise regions within the canopy images exhibiting symptoms of downy mildew. Model predictions were interpreted using explainable artificial intelligence (XAI) methods. The EfficientNetV2S model achieved an accuracy of 91 % and an F1-score of 0.92 when classifying image areas and an Intersection over Union (IoU) of 0.83 when locating symptomatic areas. This method showed promising results, enabling automatic and explainable detection and localisation of plant diseases in complex conditions. The straightforward labelling process facilitated adaptation to new conditions, making it suitable for different crops and diseases. Integration into mobile platforms could enhance disease management and reduce the spread of pathogens, making a significant advance in agricultural technology.
Why it matches plant phenotyping methods植物病害症状の画像取得・深層学習による検出および局在化手法の開発が研究の中心であり、症状という植物状態を直接推定しているため。
abstractThe aim of this work was to develop an interpretable, automated method for detection and localisation of plant disease symptoms under field conditions.
Crop improvement by means of traditional or molecular breeding is a key strategy to accomplish the European Green Deal target of reducing pesticides by 50% by 2030. Regarding viticulture, this is exacerbated by the massive use of chemicals to control pathogen infections. Black rot is an emergent disease caused by the ascomycete Phyllosticta ampelicida, and its destructiveness is alarming vine growers. Implementing and improving effective phenotyping strategies are fundamental preliminary steps to breed disease resistant varieties and this work suggests good practices adopted for this purpose. Primarily, the pedigree of black rot resistance donors was reconstructed based on the collection of phenotypic historical data, highlighting unexplored sources of black rot resistance. Strains used for artificial infections were isolated, genetically characterized and mixed to avoid race-specific resistance selection. A new inoculation protocol based on the use of leaf mature lesions was developed. Ex vivo inoculation on detached leaves was effective for the evaluation of conidia germination and hyphal growth, but not for disease progression. Finally, the pedigree was used for the identification of 23 genotypes to be tested. Two breeding selections (NY39 and NY24) resulted symptomless in all assessments and a third one (F25P52) also showed very high resistance, although with a greater variability. Other two genotypes (F12P19 and ‘Charvir’) fell within the medium resistance category, making them good candidates in a regime of well-timed preventive treatments. In conclusion, this work was effective to a comprehensive parental line characterization and preparatory towards grapevine breeding programs for black rot resistance.
Why it matches plant phenotyping methodsブドウ黒腐病抵抗性の表現型評価を目的に、人工接種条件と新規接種プロトコルを開発・評価しており、病徴に基づく植物表現型取得が研究の中心である。
abstractImplementing and improving effective phenotyping strategies are fundamental preliminary steps to breed disease resistant varieties and this work suggests good practices adopted for this purpose.
Most of the work done in image processing-based crop disease detection focuses on images with plain background. This paper presents a technique for crop disease detection for complex real field background images. A segmentation technique is presented to extract leaf patches from the entire image. Transform domain cepstral analysis is proposed for obtaining cepstral coefficients, to attain two level classifications. The first level classifies the crop species while the second level classifies the species into healthy leaf or leaf with specific type of disease. The work is tested on three crops Banana, Soybean and Grape and is checked on plain background laboratory images and on complex real field images. Suggested technique give species level accuracy of 94.33 %, 94.11 % and 98.44 % and disease level average accuracy of 97.75 %, 96.66 % and 97.95 % for Banana, Soybean and Grape, respectively. Comparison with standard features like texture and shape indicate that the presented technique gives the best results for both plain and complex background images suggesting its utilization in crop disease detection to reduce the agricultural and economic losses.
Why it matches plant phenotyping methods複雑な野外画像から葉を抽出し、画像特徴により健全葉と病害葉を分類する手法が中心で、植物の病害状態を直接推定している。
abstractThis paper presents a technique for crop disease detection for complex real field background images.
The monitoring of grape quality parameters within viticulture using airborne remote sensing is an increasingly important aspect of precision viticulture. Airborne remote sensing allows high volumes of spatial consistent data to be collected with improved efficiency over ground-based surveys. Spectral data can be used to understand the characteristics of vineyards, including the characteristics and health of the vines. Within viticultural remote sensing, the use of cover-crop spectra for monitoring is often overlooked due to the perceived noise it generates within imagery. However, within viticulture, the cover crop is a widely used and important management tool. This study uses multispectral data acquired by a high-resolution uncrewed aerial vehicle (UAV) and Sentinel-2 MSI to explore the benefit that cover-crop pixels could have for grape yield and quality monitoring. This study was undertaken across three growing seasons in the southeast of England, at a large commercial wine producer. The site was split into a number of vineyards, with sub-blocks for different vine varieties and rootstocks. Pre-harvest multispectral UAV imagery was collected across three vineyard parcels. UAV imagery was radiometrically corrected and stitched to create orthomosaics (red, green, and near-infrared) for each vineyard and survey date. Orthomosaics were segmented into pure cover-cropuav and pure vineuav pixels, removing the impact that mixed pixels could have upon analysis, with three vegetation indices (VIs) constructed from the segmented imagery. Sentinel-2 Level 2a bottom of atmosphere scenes were also acquired as close to UAV surveys as possible. In parallel, the yield and quality surveys were undertaken one to two weeks prior to harvest. Laboratory refractometry was performed to determine the grape total acid, total soluble solids, alpha amino acids, and berry weight. Extreme gradient boosting (XGBoost v2.1.1) was used to determine the ability of remote sensing data to predict the grape yield and quality parameters. Results suggested that pure cover-cropuav was a successful predictor of grape yield and quality parameters (range of R2 = 0.37–0.45), with model evaluation results comparable to pure vineuav and Sentinel-2 models. The analysis also showed that, whilst the structural similarity between the both UAV and Sentinel-2 data was high, the cover crop is the most influential spectral component within the Sentinel-2 data. This research presents novel evidence for the ability of cover-cropuav to predict grape yield and quality. Moreover, this finding then provides a mechanism which explains the success of the Sentinel-2 modelling of grape yield and quality. For growers and wine producers, creating grape yield and quality prediction models through moderate-resolution satellite imagery would be a significant innovation. Proving more cost-effective than UAV monitoring for large vineyards, such methodologies could also act to bring substantial cost savings to vineyard management.
Why it matches plant phenotyping methodsUAV・衛星マルチスペクトル画像のセグメンテーションとXGBoostを用いて、ブドウ収量・品質という植物形質を推定する方法が研究の中心である。
abstractThis study uses multispectral data acquired by a high-resolution uncrewed aerial vehicle (UAV) and Sentinel-2 MSI to explore the benefit that cover-crop pixels could have for grape yield and quality monitoring.
Automating pruning tasks entails overcoming several challenges, encompassing not only robotic manipulation but also environment perception and detection. To achieve efficient pruning, robotic systems must accurately identify the correct cutting points. A possible method to define these points is to choose the cutting location based on the number of nodes present on the targeted cane. For this purpose, in grapevine pruning, it is required to correctly identify the nodes present on the primary canes of the grapevines. In this paper, a novel method of node detection in grapevines is proposed with four distinct state-of-the-art versions of the YOLO detection model: YOLOv7, YOLOv8, YOLOv9 and YOLOv10. These models were trained on a public dataset with images containing artificial backgrounds and afterwards validated on different cultivars of grapevines from two distinct Portuguese viticulture regions with cluttered backgrounds. This allowed us to evaluate the robustness of the algorithms on the detection of nodes in diverse environments, compare the performance of the YOLO models used, as well as create a publicly available dataset of grapevines obtained in Portuguese vineyards for node detection. Overall, all used models were capable of achieving correct node detection in images of grapevines from the three distinct datasets. Considering the trade-off between accuracy and inference speed, the YOLOv7 model demonstrated to be the most robust in detecting nodes in 2D images of grapevines, achieving F1-Score values between 70% and 86.5% with inference times of around 89 ms for an input size of 1280 × 1280 px. Considering these results, this work contributes with an efficient approach for real-time node detection for further implementation on an autonomous robotic pruning system.
Why it matches plant phenotyping methodsブドウの節という明示的な植物器官形質をYOLO画像解析で検出する手法を開発・比較検証し、異なる品種・環境で評価しているため、フェノタイピング手法が中心である。
abstractIn this paper, a novel method of node detection in grapevines is proposed with four distinct state-of-the-art versions of the YOLO detection model: YOLOv7, YOLOv8, YOLOv9 and YOLOv10.
Reproduction assets foundThe paper's authors created and openly released a paper-specific grapevine node-detection image dataset (Dão and Douro vineyard images) on Zenodo, cited in the Data Availability Statement.Dataset · publicThe data presented in this study are openly available in the digital repository Zenodo: Douro & Dão Grapevines Dataset for Node Detection— https://doi.org/10.5281/zenodo.10991688 .Open asset ↗Zenodo · 10.5281/zenodo.10991688lines:552-565Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Oct 2024Computers and Electronics in Agriculture.
Plant leaf diseases severely impair crop quality and productivity. Accurate segmentation of diseases facilitates the understanding of disease distribution and is a critical step in achieving precise diagnosis and identification of diseases. However, grape leaf diseases suffer from complex backgrounds, small diseases, and high similarity between diseases, leading to limited segmentation precision. To this end, we propose a novel Mamba segmentation model for the grape leaf disease, termed GMamba, which aims to efficiently extract both fine-grained and coarse-grained disease features. Specifically, we leverage the UNet hierarchical architecture to deliver multiscale hierarchical disease information from complex backgrounds. Subsequently, we design a Co-SSM module characterized by a formulation of SSM and convolution. The former efficiently models long-range dependencies, which enables the mining of spatial global features of small diseases via multi-directional scanning. The latter focuses on capturing the local detailed features of leaf diseases. Besides, we incorporated SAB and CAB modules to diminish the loss of detail information during decoder fusion and to enhance the richness of feature information scales. Different from Transformer, GMamba globally models the leaf disease without self-attention operation. Extensive experiments have demonstrated that GMamba has competitive segmentation performance over current CNN, Transformer and CNN-Transformer combination architectures on Field-PV, Syn-PV and Plant Village datasets. GMamba yields a 3.28% IoU and 1.65% Precision gain in Black rot over Topformer. To the best of our knowledge, this is the first segmented Mamba tailored for grape leaf diseases. This study can provide an efficient and accurate method for the task of disease segmentation in grapevine leaves, which forms the basis for precise disease analysis.
Why it matches plant phenotyping methodsブドウ葉の病徴・病害領域を画像からセグメンテーションするGMambaモデルを開発し、複数データセットで性能比較・検証しており、植物の病害状態の取得手法が中心である。
abstractwe propose a novel Mamba segmentation model for the grape leaf disease, termed GMamba
Remote sensing is now a valued solution for more accurately budgeting water supply by identifying spectral and spatial information. A study was put in place in a Vitis vinifera L. cv. Cabernet-Sauvignon vineyard in the San Joaquin Valley, CA, USA, where a variable rate automated irrigation system was installed to irrigate vines with twelve different water regimes in four randomized replicates, totaling 48 experimental zones. The purpose of this experimental design was to create variability in grapevine water status, in order to produce a robust dataset for modeling purposes. Throughout the growing season, spectral data within these zones was gathered using a Near InfraRed (NIR) - Short Wavelength Infrared (SWIR) hyperspectral camera (900 to 1700 nm) mounted on an Unmanned Aircraft Vehicle (UAV). Given the high water-absorption in this spectral domain, this sensor was deployed to assess grapevine stem water potential, Ψₛₜₑₘ, a standard reference for water status assessment in plants, from pure grapevine pixels in hyperspectral images. The Ψₛₜₑₘ was acquired simultaneously in the field from bunch closure to harvest and modeled via machine-learning methods using the remotely sensed NIR-SWIR data as predictors in regression and classification modes (classes consisted of physiologically different water stress levels). Hyperspectral images were converted to bottom of atmosphere reflectance using standard panels on the ground and through the Quick Atmospheric Correction Method (QUAC) and the results were compared. The best models used data obtained with standard panels on the ground and allowed predicting Ψₛₜₑₘ values with an R² of 0.54 and an RMSE of 0.11 MPa as estimated in cross-validation, and the best classification reached an accuracy of 74%. This project aims to develop new methods for precisely monitoring and managing irrigation in vineyards while providing useful information about plant physiology response to deficit irrigation.
Why it matches plant phenotyping methodsUAV搭載NIR/SWIRハイパースペクトル画像と機械学習により、ブドウの茎水ポテンシャルという植物生理形質を推定し、回帰・分類性能を検証している。表現型取得法と技術評価が研究の中心である。
abstractthis sensor was deployed to assess grapevine stem water potential, Ψₛₜₑₘ, a standard reference for water status assessment in plants, from pure grapevine pixels in hyperspectral images.
PURPOSE: High resolution imagery from unmanned aerial vehicles (UAVs) has been established as an important source of information to perform precise irrigation practices, notably relevant for high value crops often present in semi-arid regions such as vineyards. Many studies have shown the utility of thermal infrared (TIR) sensors to estimate canopy temperature to inform on vine physiological status, while visible-near infrared (VNIR) imagery and 3D point clouds derived from red–green–blue (RGB) photogrammetry have also shown great promise to better monitor within-field canopy traits to support agronomic practices. Indeed, grapevines react to water stress through a series of physiological and growth responses, which may occur at different spatio-temporal scales. As such, this study aimed to evaluate the application of TIR, VNIR and RGB sensors onboard UAVs to track vine water stress over various phenological periods in an experimental vineyard imposed with three different irrigation regimes. METHODS: A total of twelve UAV overpasses were performed in 2022 and 2023 where in situ physiological proxies, such as stomatal conductance (gₛ), leaf (Ψₗₑₐf) and stem (Ψₛₜₑₘ) water potential, and canopy traits, such as LAI, were collected during each UAV overpass. Linear and non-linear models were trained and evaluated against in-situ measurements. RESULTS: Results revealed the importance of TIR variables to estimate physiological proxies (gₛ, Ψₗₑₐf, Ψₛₜₑₘ) while VNIR and 3D variables were critical to estimate LAI. Both VNIR and 3D variables were largely uncorrelated to water stress proxies and demonstrated less importance in the trained empirical models. However, models using all three variable types (TIR, VNIR, 3D) were consistently the most effective to track water stress, highlighting the advantage of combining vine characteristics related to physiology, structure and growth to monitor vegetation water status throughout the vine growth period. CONCLUSION: This study highlights the utility of combining such UAV-based variables to establish empirical models that correlated well with field-level water stress proxies, demonstrating large potential to support agronomic practices or even to be ingested in physically-based models to estimate vine water demand and transpiration.
Why it matches plant phenotyping methodsUAV搭載の熱・マルチスペクトル・3D画像を組み合わせ、ブドウの水ストレス、生理指標、LAIを推定する手法を評価・検証しており、表現型取得とモデル性能評価が研究の中心である。
abstractthis study aimed to evaluate the application of TIR, VNIR and RGB sensors onboard UAVs to track vine water stress over various phenological periods
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 7 Sept 2026
Rootstocks are gaining importance in viticulture as a strategy to combat abiotic challenges, as well as enhancing scion physiology. Photosynthetic parameters such as maximum rate of carboxylation of RuBP (V cmax ) and the maximum rate of electron transport driving RuBP regeneration (J max ) have been identified as ideal targets for potential influence by rootstock and breeding. However, leaf specific direct measurement of these photosynthetic parameters is time consuming, limiting the information scope and the number of individuals that can be screened. This study aims to overcome these limitations by employing hyperspectral imaging combined with artificial intelligence (AI) to predict these key photosynthetic traits at the canopy level. Hyperspectral imaging captures detailed optical properties across a broad range of wavelengths (400 to 1000 nm), enabling use of all wavelengths in a comprehensive analysis of the entire vine’s photosynthetic performance (V cmax and J max ). Artificial intelligence-based prediction models that blend the strength of deep learning and machine learning were developed using two growing seasons data measured post-solstice at 15 h, 14 h, 13 h and 12 h daylengths for Vitis hybrid ‘Marquette’ grafted to five commercial rootstocks and ‘Marquette’ grafted to ‘Marquette’. Significant differences in photosynthetic efficiency (V cmax and J max ) were noted for both direct and indirect measurements for the six rootstocks, indicating that rootstock genotype and daylength have a significant influence on scion photosynthesis. Evaluation of multiple feature-extraction algorithms indicated the proposed Vitis base model incorporating a 1D-Convolutional neural Network (CNN) had the best prediction performance with a R 2 of 0.60 for V cmax and J max . Inclusion of weather and chlorophyll parameters slightly improved model performance for both photosynthetic parameters. Integrating AI with hyperspectral remote phenotyping provides potential for high-throughput whole vine assessment of photosynthetic performance and selection of rootstock genotypes that confer improved photosynthetic performance potential in the scion.
Why it matches plant phenotyping methodsハイパースペクトル画像とAIにより、ブドウ樹キャノピーから光合成形質を推定するリモート・フェノタイピング手法を開発・評価しており、手法が研究の中心である。
abstractThis study aims to overcome these limitations by employing hyperspectral imaging combined with artificial intelligence (AI) to predict these key photosynthetic traits at the canopy level.
GrapevineAerial / UAVField / plotLiDAR / point cloudMultispectral / hyperspectralThermalLeafPhysiological trait estimationWater status / transpiration
Abstract. Grapevine water status exhibits substantial variability even within a single vineyard. Understanding how edaphic, topographic and climatic conditions impact grapevine water status heterogeneity at the field scale, in non-irrigated vineyards, is essential for winemakers as it significantly influences wine quality. This study aimed to quantify the spatial distribution of grapevine leaf water potential (Ψleaf) within vineyards and to assess the influence of soil properties heterogeneity, topography and weather on this intra-field variability, in two non-irrigated vineyards during two viticultural seasons. By combining multilinearly vegetation indices from very-high spatial resolution multispectral, thermal and LiDAR imageries collected with unmanned aerial systems, we efficiently and robustly captured the spatial distribution of Ψleaf across both vineyards, at different dates. Our results demonstrated that in non-irrigated vineyards, the spatial distribution of Ψleaf was mainly governed by the within-vineyard soil hydraulic conductivity heterogeneity (R² up to 0.81), and was particularly marked when the evaporative demand and the soil water deficit increased, since the range of Ψleaf was greater, up to 0.73 MPa, in these conditions. However, topographic attributes (elevation and slope) were less related to grapevine Ψleaf variability. These findings show that soil properties within-field spatial distribution and weather conditions are the primary factors governing Ψleaf heterogeneity observed in non-irrigated vineyards, and their effects are concomitants.
Why it matches plant phenotyping methodsUAVのマルチスペクトル・熱・LiDAR画像を統合し、ブドウの葉水ポテンシャルという生理形質の空間分布を推定する手法を実質的に適用しているため、植物フェノタイピング手法の応用として含める。
abstractBy combining multilinearly vegetation indices from very-high spatial resolution multispectral, thermal and LiDAR imageries collected with unmanned aerial systems, we efficiently and robustly captured the spatial distribution of Ψleaf across both vineyards, at different dates.
Abstract Grape disease image recognition is a crucial part of agricultural disease detection, and accurate identification of grape leaves plays a vital role in agricultural production. This study proposes a deep learning-based method for grape disease classification and recognition to address issues such as the complexity of grape disease features, uneven distribution of disease features, and data imbalance. First, the adversarial generative network FastGAN is used to generate grape disease images to enrich the sample information for different categories in the dataset and address the data imbalance problem. Then, a novel Transformer structure called LVT Block and a CNN structure called MARI Block are proposed to process the global and local information of images, respectively. Dense connections between these structures result in the DLVT Block, which leads to the lightweight neural network model DLVTNet. Additionally, a lightweight self-attention mechanism combined with CNN called CLSHSA is introduced , which maintains high recognition performance while reducing the model size. Moreover, a multi-scale attention mechanism (MELA) is proposed, which combines positional and multi-scale information to obtain attention weights. Experimental results show that this method achieves an average recognition accuracy of 98.48% in grape leaf disease detection, outperforming mainstream CNN and Transformer models, and effectively focuses on disease areas in leaf images. The method also demonstrates high recognition accuracy in tomato disease detection, indicating good generalization ability and suitability for detecting and recognizing various leaf diseases. The proposed method provides an effective solution for detecting and recognizing grape and other plant leaf diseases, offering a new research approach that combines CNN and Transformer structures.
Why it matches plant phenotyping methods植物葉の病害領域・病害状態を画像から認識する深層学習法の開発が中心であり、植物病害フェノタイピング手法に該当する。
abstractThis study proposes a deep learning-based method for grape disease classification and recognition
Reproduction assets foundThe paper's grape (and tomato) leaf disease image inputs come directly from the public New Plant Diseases Dataset on Kaggle, which is the image dataset used for the paper's phenotyping measurements. No author code, trained models, or generated FastGAN dataset is publicly deposited; the data availability statement only'Dataset · publicformation to obtain attention weights in images, aiding
the model in effectively extracting diseased areas.
2. Materials and Methods
2.1. Image datasets and preprocessing
The grape leaf disease dataset used in this study comes from the publicly available plant disease classification dataset New
Plant Diseases Dataset on Kaggle (https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset). Grape leaf images
from this dataset were selected as the image dataset for this study. The dataset includes images of three types of grape leaf diseases
as well as healthy leaves, totaling 7,222 images, divided into four categories. The images have been resized to 256×256 pixels and
processed usingOpen asset ↗Kaggle · new-plant-diseases-datasetpdf-raw-page:4 lines:1-36Plant phenotyping relevance match · UnverifiedEurope PMC · checked 7 Sept 2026
Abstract Grapes and potato, both a crucial component of the global agricultural economy, are susceptible to various diseases that can adversely affect crop quality and yield. Recently, the use of Deep Learning (DL) techniques in agriculture has shown potential for predicting and detecting diseases early. This study explores the effectiveness of Convolutional Neural Networks (CNN), Efficient Net, and Residual Networks (ResNet) in identifying diseases in grape and potata leavess. It employs a database containing high-resolution images of healthy and diseased grape leaves, including conditions like leaf blight, and grape and potata leaves browny mildew. Data pre-processing methods are used to standardize and enhance the datasets for model training and evaluation. The study implements and fine-tunes three DL classifiers—CNN, Efficient Net, and ResNet—using transfer learning. To assess the models' performance in disease classification, the dataset is divided into training and validation subsets. Metrics such as accuracy, recall, precision, and F1-score are used to evaluate the models' predictive capabilities. The experimental results show that CNN achieved 94% accuracy, ResNet attained the highest efficiency with 96% accuracy, and Efficient Net reached 97% accuracy.
Why it matches plant phenotyping methodsブドウ・ジャガイモ葉の病害状態を画像から分類する深層学習手法を比較・評価しており、植物病害フェノタイピングが中心的な技術課題である。
titleComparative Analysis of CNN, EFFICIENTNET and RESNET for Grape and Potato leaves Disease Prediction: A Deep Learning Approach.
In an agricultural field, plant phenotyping using object detection models is gaining attention. However, collecting the training data necessary to create generic and high-precision models is extremely challenging due to the difficulty of annotation and the diversity of domains. Furthermore, it is difficult to transfer training data across different crops, and although machine learning models effective for specific environments, conditions, or crops have been developed, they cannot be widely applied in actual fields. In this study, we propose a generative data augmentation method (D4) for vineyard shoot detection. D4 uses a pre-trained text-guided diffusion model based on a large number of original images culled from video data collected by unmanned ground vehicles or other means, and a small number of annotated datasets. The proposed method generates new annotated images with background information adapted to the target domain while retaining annotation information necessary for object detection. In addition, D4 overcomes the lack of training data in agriculture, including the difficulty of annotation and diversity of domains. We confirmed that this generative data augmentation method improved the mean average precision by up to 28.65% for the BBox detection task and the average precision by up to 13.73% for the keypoint detection task for vineyard shoot detection. Our generative data augmentation method D4 is expected to simultaneously solve the cost and domain diversity issues of training data generation in agriculture and improve the generalization performance of detection models.
Why it matches plant phenotyping methodsブドウのシュート検出を対象に、ドメイン適応型の生成データ拡張法を開発・評価しており、植物器官の画像ベース表現型取得における方法的貢献が中心である。
abstractIn this study, we propose a generative data augmentation method (D4) for vineyard shoot detection.
Automating pruning tasks entails overcoming several challenges, encompassing not only robotic manipulation but also environment perception and detection. To achieve efficient pruning, robotic systems must accurately identify the correct cutting points. A possible method to define these points is to choose the cutting location based on the number of nodes present on the targeted cane. For this purpose, in grapevine pruning, it is required to correctly identify the nodes present on the primary canes of the grapevines. In this paper, a novel method of node detection in grapevines is proposed with four distinct state-of-the-art versions of the YOLO detection model: YOLOv7, YOLOv8, YOLOv9 and YOLOv10. These models were trained on a public dataset with images containing artificial backgrounds and afterwards validated on different cultivars of grapevines from two distinct Portuguese viticulture regions with cluttered backgrounds. This allowed to evaluate the robustness of the algorithms on the detection of nodes in diverse environments, compare the performance of the YOLO models used, as well as create a publicly available dataset of grapevines obtained in Portuguese vineyards for node detection. Overall, all used models were capable of achieving correct node detection on images of grapevines from the three distinct datasets. Considering the trade-off between accuracy and inference speed, the YOLOv7 model demonstrated to be the most robust in detecting nodes in 2D images of grapevines, achieving F1-Score values between 70 % and 86.5 % with inference times of around 89 ms for an input size of 1280×1280 px. Considering these results, this work contributes with an efficient approach for real-time node detection for further implementation on an autonomous robotic pruning system.
Why it matches plant phenotyping methodsブドウの節という植物器官の形態を画像から検出する手法を開発し、異なる品種・環境で検証するとともに、データセットも作成しており、単なる収穫対象の位置検出を超えた植物表現型取得が中心である。
abstractThese models were trained on a public dataset with images containing artificial backgrounds and afterwards validated on different cultivars of grapevines from two distinct Portuguese viticulture regions with cluttered backgrounds.
Understanding geometric and biophysical characteristics is essential for determining grapevine vigor and improving input management and automation in viticulture. This study compares point cloud data obtained from a Terrestrial Laser Scanner (TLS) and various UAV sensors including multispectral, panchromatic, Thermal Infrared (TIR), RGB, and LiDAR data, to estimate geometric parameters of grapevines. Descriptive statistics, linear correlations, significance using the F-test of overall significance, and box plots were used for analysis. The results indicate that 3D point clouds from these sensors can accurately estimate maximum grapevine height, projected area, and volume, though with varying degrees of accuracy. The TLS data showed the highest correlation with grapevine height ( r = 0.95, p R 2 = 0.90; RMSE = 0.027 m), while point cloud data from panchromatic, RGB, and multispectral sensors also performed well, closely matching TLS and measured values ( r > 0.83, p R 2 > 0.70; RMSE r = 0.76, p R 2 = 0.58; RMSE = 0.147 m) and projected area ( r = 0.82, p R 2 = 0.66; RMSE = 0.165 m). The greater variability observed in projected area and volume from UAV sensors is related to the low point density associated with spatial resolution. These findings are valuable for both researchers and winegrowers, as they support the optimization of TLS and UAV sensors for precision viticulture, providing a basis for further research and helping farmers select appropriate technologies for crop monitoring.
Why it matches plant phenotyping methodsTLSおよびUAVセンサーによるブドウ樹の高さ・投影面積・体積推定を比較検証しており、植物形質の取得手法の技術評価が中心である。
abstractThis study compares point cloud data obtained from a Terrestrial Laser Scanner (TLS) and various UAV sensors including multispectral, panchromatic, Thermal Infrared (TIR), RGB, and LiDAR data, to estimate geometric parameters of grapevines.
Plant diseases are a critical threat to global food security and agricultural sustainability because of the crop losses they cause. 195 million tons of crops are lost to fungal diseases each year and alone in India, more than 5 million metric tonnes go waste annually from it[1]. Also, from FAO "Globally up to 16% of harvests worth about US$220 billion are lost due to plant pests every year" [2] The urgency of the situation is clear, as wrapped up in these figures are reasons why long-term growth requires early bite detection services to prevent plant disease. In this paper, deep learning algorithms were used to detect diseases in plants by taking image of the leaves as input. Our research is limited to the detection of these 6 plant diseases, Aphid infestation, Bacterial leaf spot disease Black apple scab Early blight Septoria leaf spot on tomato Grape powdery mildew. In order to do this, we construct and train a machine learning model using two different raw image datasets. These were meticulously curated and enriched datasets, geared towards improving the model's generalisability across extensive variety of conditions. Our approach not only facilitates the early detection of these diseases but also demonstrates the potential for scalable, real-time applications in agricultural settings. The results highlight the effectiveness of deep learning in identifying and classifying plant diseases, offering a promising solution for reducing crop losses and improving agricultural productivity.
Why it matches plant phenotyping methods葉画像から植物病害を検出・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するため、植物フェノタイピング手法として含める。
abstractdeep learning algorithms were used to detect diseases in plants by taking image of the leaves as input.
Abstract Precision agriculture is evolving toward a contemporary approach that involves multiple sensing techniques to monitor and enhance crop quality while minimizing losses and waste of no longer considered inexhaustible resources, such as soil and water supplies. To understand crop status, it is necessary to integrate data from heterogeneous sensors and employ advanced sensing devices that can assess crop and water status. This study presents a smart monitoring approach in agriculture, involving sensors that can be both stationary (such as soil moisture sensors) and mobile (such as sensor-equipped unmanned aerial vehicles). These sensors collect information from visual maps of crop production and water conditions, to comprehensively understand the crop area and spot any potential vegetation problems. A modular fuzzy control scheme has been designed to interpret spectral indices and vegetative parameters and, by applying fuzzy rules, return status maps about vegetation status. The rules are applied incrementally per a hierarchical design to correlate lower-level data (e.g., temperature, vegetation indices) with higher-level data (e.g., vapor pressure deficit) to robustly determine the vegetation status and the main parameters that have led to it. A case study was conducted, involving the collection of satellite images from artichoke crops in Salerno, Italy, to demonstrate the potential of incremental design and information integration in crop health monitoring. Subsequently, tests were conducted on vineyard regions of interest in Teano, Italy, to assess the efficacy of the framework in the assessment of plant status and water stress. Indeed, comparing the outcomes of our maps with those of cutting-edge machine learning (ML) semantic segmentation has indeed revealed a promising level of accuracy. Specifically, classification performance was compared to the output of conventional ML methods, demonstrating that our approach is consistent and achieves an accuracy of over 90% throughout various seasons of the year.
Why it matches plant phenotyping methods植物の状態・水ストレスを衛星画像やセンサー情報から推定し、階層型ファジールールで植物状態マップを生成・検証する手法が研究の中心であるため。
abstractA modular fuzzy control scheme has been designed to interpret spectral indices and vegetative parameters and, by applying fuzzy rules, return status maps about vegetation status.
The main purpose of this study was to create a prototype of an unmanned aerial system equipped with intelligent hardware and software technologies necessary for monitoring the health and growth of crops in orchards. Another important objective was to use low-cost sensors that accurately measure ultraviolet solar radiation. The device, which needs to be attached to the commercial DJI Mini 4 Pro drone, should be small in size, portable, and have very low energy consumption. For this purpose, the widely used Vishay VEML6075 digital optical sensor was selected and implemented in a prototype, alongside a Raspberry Pi Zero 2W minicomputer. To collect data from these sensors, a program written in Python was used, containing specific blocks for data acquisition from each sensor, to facilitate the monitoring of ultraviolet (UV) radiation, or battery current. By analyzing the data obtained from the sensors, several important conclusions are drawn that may provide valuable pathways for the further development of mobile or modular equipment. Furthermore, the results of the plant condition analysis with proposed models in the Geographic Information System (GIS) environment were also presented. The visualization of maps indicating variations in vegetation condition led to the identification of problem areas like hydric stress.
Why it matches plant phenotyping methods果樹園・ブドウ園の植物健全性・生育状態を監視するUAVセンサー／ソフトウェアの試作と、植生状態・水ストレスの解析を中心とするため、植物フェノタイピング基盤の開発に該当します。
abstractThe main purpose of this study was to create a prototype of an unmanned aerial system equipped with intelligent hardware and software technologies necessary for monitoring the health and growth of crops in orchards.
Abstract Thermal remote sensing indicators of crop water status can help to optimize irrigation across time and space. The Crop Water Stress Index (CWSI), calculated from thermal data, has been widely used in irrigation management as it has a proven association with evapotranspiration ratios. However, different approaches can be used to calculate the CWSI. The aim of this study is to identify the most robust method for estimating the CWSI in a commercial Merlot vineyard using high-resolution thermal imaging from Unoccupied Aerial Systems (UAS). To that end, three different methods were used to estimate the CWSI: Jackson’s model (CWSIj), Wet Artificial Reference Surface (WARS) method (CWSIw), and the Bellvert approach (CWSIb). A simpler indicator calculated as the difference between canopy and air temperature (Tc–Ta) was the benchmark to beat. The water status of a vine cultivar with anisohydric behavior (Merlot) in a vineyard in central Spain was assessed for two years with different agroclimatic conditions. Canopy temperature (Tc) was obtained from UAS flights at 9:00 h and 12:00 h solar hour over eight days during the irrigation period (June–August), and from vines under five different irrigation treatments. Stem water potential (SWP), stomatal conductance (gs), and leaf temperature (TL) were recorded at the time of the flights and compared with the thermal indices (CWSIj, CWSIw, CWSIb) and the benchmark indicator (Tc–Ta). Results show that the simpler indicator of water stress, Tc–Ta, performed better at identifying varying levels of crop hydration than CWSIb or CWSIw at 12:00 h. Under conditions of extreme aridity, the latter indices were less accurate than the physically-based CWSIj at 12:00 h, which had the highest correlation with SWP (r = 0.84), followed by the benchmark index Tc–Ta (r = 0.70 at 12:00). Considering the current climatic trends towards aridification, the CWSIj emerges as a useful operational tool, with robust performance across days and times of day. These results are important for irrigation management and could contribute to improving water use efficiency in agriculture.
Why it matches plant phenotyping methodsUAS熱画像からブドウの水分状態を推定する複数の熱指標を比較・検証しており、植物生理状態の取得方法が研究の中心である。
abstractThe aim of this study is to identify the most robust method for estimating the CWSI in a commercial Merlot vineyard using high-resolution thermal imaging from Unoccupied Aerial Systems (UAS).
Crop diseases can significantly affect various aspects of crop cultivation, including crop yield, quality, production costs, and crop loss. The utilization of modern technologies such as image analysis via machine learning techniques enables early and precise detection of crop diseases, hence empowering farmers to effectively manage and avoid the occurrence of crop diseases. The proposed methodology involves the use of modified MobileNetV3Large model deployed on edge device for real-time monitoring of grape leaf disease while reducing computational memory demands and ensuring satisfactory classification performance. To enhance applicability of MobileNetV3Large, custom layers consisting of two dense layers were added, each followed by a dropout layer, helped mitigate overfitting and ensured that the model remains efficient. Comparisons among other models showed that the proposed model outperformed those with an average train and test accuracy of 99.66% and 99.42%, with a precision, recall, and F1 score of approximately 99.42%. The model was deployed on an edge device (Nvidia Jetson Nano) using a custom developed GUI app and predicted from both saved and real-time data with high confidence values. Grad-CAM visualization was used to identify and represent image areas that affect the convolutional neural network (CNN) classification decision-making process with high accuracy. This research contributes to the development of plant disease classification technologies for edge devices, which have the potential to enhance the ability of autonomous farming for farmers, agronomists, and researchers to monitor and mitigate plant diseases efficiently and effectively, with a positive impact on global food security.
Why it matches plant phenotyping methodsブドウ葉の病徴を画像から分類するCNNを開発し、エッジデバイスへ実装・評価した研究であり、植物病害状態の取得・推定手法が中心である。
abstractThe proposed methodology involves the use of modified MobileNetV3Large model deployed on edge device for real-time monitoring of grape leaf disease while reducing computational memory demands and ensuring satisfactory classification performance.
Vine disease detection is considered one of the most crucial components in precision viticulture. It serves as an input for several further modules, including mapping, automatic treatment, and spraying devices. In the last few years, several approaches have been proposed for detecting vine disease based on indoor laboratory conditions or large-scale satellite images integrated with machine learning tools. However, these methods have several limitations, including laboratory-specific conditions or limited visibility into plant-related diseases. To overcome these limitations, this work proposes a low-altitude drone flight approach through which a comprehensive dataset about various vine diseases from a large-scale European dataset is generated. The dataset contains typical diseases such as downy mildew or black rot affecting the large variety of grapes including Muscat of Hamburg, Alphonse Lavallée, Grasă de Cotnari, Rkatsiteli, Napoca, Pinot blanc, Pinot gris, Chambourcin, Fetească regală, Sauvignon blanc, Muscat Ottonel, Merlot, and Seyve-Villard 18402. The dataset contains 10,000 images and more than 100,000 annotated leaves, verified by viticulture specialists. Grape bunches are also annotated for yield estimation. Further, tests were made against state-of-the-art detection methods on this dataset, focusing also on viable solutions on embedded devices, including Android-based phones or Nvidia Jetson boards with GPU. The datasets, as well as the customized embedded models, are available on the project webpage.
Why it matches plant phenotyping methodsブドウ葉の病徴を低高度ドローン画像で検出する大規模データセットを構築し、葉アノテーションと手法比較・組込み機器での評価を行っており、植物状態の画像ベース計測が中心です。
abstractthis work proposes a low-altitude drone flight approach through which a comprehensive dataset about various vine diseases from a large-scale European dataset is generated.
Reproduction assets foundThe authors state their UAV vine-disease dataset (10,000 images, 100,000+ annotated leaves), preprocessing scripts, and pre-trained embedded models are publicly available on the project website, whose URL (github.com/tamaslevente/vineye) appears in the supplied blocks. Other URLs (ultralytics, CVAT, labelImg, Zenodo, kDataset · publicThe dataset and useful preprocessing scripts and pre-trained models are available on the project website.Open asset ↗lines:43-81Plant phenotyping relevance match · UnverifiedEurope PMC · checked 7 Sept 2026
It is crucial for winegrowers to make informed decisions about the optimum time to harvest the grapes to ensure the production of premium wines. Global warming contributes to decreasing acidity and increasing sugar levels in grapes, resulting in bland wines with high contents of alcohol. Predicting quality in viticulture is thus pivotal. To assess the average ripeness, typically a sample of one hundred berries representative for the entire vineyard is collected. However, this process, along with the subsequent detailed must analysis, is time consuming and expensive. This study focusses on predicting essential quality parameters like sugar and acid content in Vitis vinifera (L.) varieties 'Chardonnay', 'Riesling', 'Dornfelder', and 'Pinot Noir'. A small near-infrared spectrometer was used measuring non-destructively in the wavelength range from 1 100 nm to 1 350 nm while the reference contents were measured using high-performance liquid chromatography. Chemometric models were developed employing partial least squares regression and using spectra of all four grapevine varieties, spectra gained from berries of the same colour, or from the individual varieties. The models exhibited high accuracy in predicting main quality-determining parameters in independent test sets. On average, the model regression coefficients exceeded 93% for the sugars fructose and glucose, 86% for malic acid, and 73% for tartaric acid. Using these models, prediction accuracies revealed the ability to forecast individual sugar contents within an range of ± 6.97 g/L to ± 10.08 g/L, and malic acid within ± 2.01 g/L to ± 3.69 g/L. This approach indicates the potential to develop robust models by incorporating spectra from diverse grape varieties and berries of different colours. Such insight is crucial for the potential widespread adoption of a handheld near-infrared sensor, possibly integrated into devices used in everyday life, like smartphones. A server-side and cloud-based solution for pre-processing and modelling could thus avoid pitfalls of using near-infrared sensors on unknown varieties and in diverse wine-producing regions.
Why it matches plant phenotyping methodsブドウ果実の糖・酸含量という植物器官形質を、近赤外分光とケモメトリックモデルで非破壊推定し、独立テストセットで精度検証しており、表現型取得法が研究の中心である。
abstractThis study focusses on predicting essential quality parameters like sugar and acid content in Vitis vinifera (L.) varieties 'Chardonnay', 'Riesling', 'Dornfelder', and 'Pinot Noir'.
The Normalized Difference Vegetation Index (NDVI) is a valuable indicator of plant vigor that is frequently used in agronomic practices to make timely and targeted decisions with the aim of increasing the productivity of the system. NDVI measurements of large-scale fields are typically performed using remote sensing from satellite and aerial imaging devices. However, due to their low spatial and temporal resolution, these technologies may have limitations in precision viticulture. This paper investigates the potential of a proximal sensing system to characterize the vine foliage that makes use of data collected by a farmer robot equipped with an Intel RealSense D435 camera. The camera includes two infrared (IR) sensors in stereoscopic configuration and one RGB sensor, which provide, for each observation, both infrared and visible red channel information, thus making possible pixel-per-pixel NDVI calculation. Solutions to IR filtering and radiometric calibration issues are proposed that significantly improve measurement accuracy and reliability. Since the camera also provides stereo-based 3D scene reconstruction, depth information can be used to separate the plant canopy from the background before NDVI measurement. At the same time, range data can be employed to extract geometric properties of the crop, such as plant height and/or volume. The system is validated in the field in a commercial vineyard at different phenological stages, from the formation of the berries to leaf discoloration and fall. Experimental results show good agreement compared with ground truth provided by a GreenSeeker, with an average percentage error in the NDVI estimation of 4.6% and a R2 of 0.87 tested over the whole grapevine cycle. Therefore, the proposed sensing system could be a feasible solution to automated NDVI estimation at plant-scale.
Why it matches plant phenotyping methods植物スケールのNDVI・樹冠形状を取得する近接センシングシステムを開発し、放射補正やIRフィルタリングを提案して圃場検証しているため、植物フェノタイピング手法が中心である。
abstractThis paper investigates the potential of a proximal sensing system to characterize the vine foliage that makes use of data collected by a farmer robot equipped with an Intel RealSense D435 camera.
This study investigated a quick way to discriminate grape varieties based on their composition in volatile compounds through a SIFT-MS scan coupled with simple chemometrics approaches such as analysis of variance (ANOVA), principal component analysis (PCA) and hierarchical ascendant classification (HAC). The 23 studied grape varieties were distinguishable using O₂⁺, H₃O⁺ and NO⁺ as reagent ions, and the combination of these three ions. For its ability to ionize most compounds, to efficiently fragment them to generate ions with distinct m/z ratio, and to enhance the differentiation of compounds of similar masses, O₂⁺ reagent ion should be preferentially considered. The use of one single ion rather than three enables to limit the time of analysis and the number of variables to be treated. The technique allowed the distinction of high and low aroma compounds producers as confirmed by headspace solid-phase microextraction followed by gas chromatography-mass spectrometry (HS-SPME/GC-MS) analyses. SIFT-MS is a quick and interesting tool with potential application in various fields of viticulture such as phenotyping of grape varieties or non-targeted studies on the impact of environmental factors or viticultural practices on grape aroma composition.
Why it matches plant phenotyping methodsSIFT-MSとケモメトリクスによるブドウ果実の揮発性成分・香気特性の識別手法が研究の中心であり、品種フェノタイピングへの応用も明示されている。
abstractThis study investigated a quick way to discriminate grape varieties based on their composition in volatile compounds through a SIFT-MS scan coupled with simple chemometrics approaches
Grape cluster architecture and compactness are complex traits influencing disease susceptibility, fruit quality, and yield. Evaluation methods for these traits include visual scoring, manual methodologies, and computer vision, with the latter being the most scalable approach. Most of the existing computer vision approaches for processing cluster images often rely on conventional segmentation or machine learning with extensive training and limited generalization. The Segment Anything Model (SAM), a novel foundation model trained on a massive image dataset, enables automated object segmentation without additional training. This study demonstrates out-of-the-box SAM's high accuracy in identifying individual berries in 2-dimensional (2D) cluster images. Using this model, we managed to segment approximately 3,500 cluster images, generating over 150,000 berry masks, each linked with spatial coordinates within their clusters. The correlation between human-identified berries and SAM predictions was very strong (Pearson's r 2 = 0.96). Although the visible berry count in images typically underestimates the actual cluster berry count due to visibility issues, we demonstrated that this discrepancy could be adjusted using a linear regression model (adjusted R 2 = 0.87). We emphasized the critical importance of the angle at which the cluster is imaged, noting its substantial effect on berry counts and architecture. We proposed different approaches in which berry location information facilitated the calculation of complex features related to cluster architecture and compactness. Finally, we discussed SAM's potential integration into currently available pipelines for image generation and processing in vineyard conditions.
Why it matches plant phenotyping methodsSAMを用いたブドウ房画像からの個別果粒セグメンテーション、検証、補正、および房構造・コンパクトネス形質の算出が研究の中心であるため。
abstractThis study demonstrates out-of-the-box SAM's high accuracy in identifying individual berries in 2-dimensional (2D) cluster images.
Reproduction assets foundThe authors state that all data and code to reproduce the study's grapevine cluster segmentation and architecture analysis are publicly available in their GitHub repository. Other URLs (SAM checkpoint, pycocotools, RMBG, arXiv refs) are generic third-party resources, not paper-specific assets.Code · publicAll the data and code to reproduce the results of this study are available at https://github.com/diazgarcialab/SAM-cluster-segmentation .Open asset ↗diazgarcialab/SAM-cluster-segmentationlines:105-230Plant phenotyping relevance match · UnverifiedEurope PMC · bioRxiv · checked 13 Sept 2026
Soil salinity affects major viticultural areas worldwide with chloride ions being the primary source of salt toxicity in grapevines. This toxicity impacts vine health and reduces fruit yield and quality. Current breeding efforts to improve grapevine salinity tolerance are limited by the low throughput of available phenotyping methods, which are time-consuming, labor-intensive, and destructive. This study demonstrated that hyperspectral proximal sensing can be utilized as a high-throughput, non-destructive screening technique to identify salinity-tolerant grapevine germplasm. The predictive abilities of two different hyperspectral devices, which varied in price, resolution, and sensitivity, were compared across 23 Vitis accessions spanning eight species. Prediction models were built using hyperspectral reflectance and leaf chloride content measured with a lab chloridometer. Three distinct approaches were studied: 1) analyzing the correlation between individual wavelengths and chloride content; 2) employing machine learning models, including Partial Least Squares Regression (PLSR), Random Forest (RF), and Support Vector Machine (SVM), utilizing all wavelengths; and 3) classification-based prediction using Partial Least Squares Discriminant Analysis (PLSDA). Multiple regions in the spectrum, including 613-660 nm, 689-696 nm, and 1357-1358 nm, showed a medium correlation (0.30-0.50) with chloride content in the leaves. PLSR was the most effective machine learning approach, demonstrating moderate predictive capability for chloride content (maximum R² = 0.67), though performance varied between the two devices tested. With PLSDA, predictions increased considerably, up to an accuracy of 0.97, depending on the instrument used and the spectral data transformation. Overall, the more expensive and sensitive device with a wider spectral range outperformed the more affordable, shorter-range device. However, when the prediction model was based on classes (chloride excluders vs. non-excluders) rather than chloride content, the differences in prediction abilities were minimal, with both instruments performing very well. This is promising for identifying breeding materials with chloride exclusion capabilities at low cost and high throughput.
Why it matches plant phenotyping methodsブドウ葉の塩化物含量・排除能力を対象に、ハイパースペクトルセンシングと予測モデルを開発・比較評価した、中心的な植物フェノタイピング研究である。
abstractThis study demonstrated that hyperspectral proximal sensing can be utilized as a high-throughput, non-destructive screening technique to identify salinity-tolerant grapevine germplasm.
Detecting crop diseases before they spread poses a significant challenge for farmers. While both deep learning (DL) and computer vision are valuable for image classification, DL necessitates larger datasets and more extensive training periods. To overcome the limitations of working with constrained datasets, this paper proposes an ensemble model to enhance overall performance. The proposed ensemble model combines the convolution neural network (CNN)-based models as feature extractors with random forest (RF) as the output classifier. Our method is built on popular CNN-based models such as VGG16, InceptionV3, Xception, and ResNet50. Traditionally, these CNN-based architectures are referred to as one-way models, but in our approach, they are connected in parallel to form a two-way configuration, enabling the extraction of more diverse features and reducing the risk of underfitting, particularly with limited datasets. To demonstrate the effectiveness of our ensemble approach, we train models using the grape leaf dataset, which is divided into two subsets: original and modified. In the original set, background removal is applied to the images, while the modified set includes preprocessing techniques such as intensity averaging and bilateral filtering for noise reduction and image smoothing. Our findings reveal that ensemble models trained on modified images outperform those trained on the original dataset. We observe improvements of up to 5.6 % in accuracy, precision, and sensitivity, thus validating the effectiveness of our approach in enhancing disease pattern recognition within limited datasets.
Why it matches plant phenotyping methodsブドウ葉の画像から病徴・病害状態を推定する画像解析モデルを提案し、前処理条件とアンサンブル構成を比較検証しているため、植物フェノタイピング手法が中心である。
abstractOur findings reveal that ensemble models trained on modified images outperform those trained on the original dataset.
Reproduction assets foundThe paper's grape leaf disease image dataset is publicly available on Kaggle; no author analysis code is released.Dataset · publicarious alterations affect these models' functionality in future study. Moreover, exploring how these results translate to different datasets and domains can provide new perspectives on how well enhanced images improve model performance.
Data availability
The datasets analyzed in the paper are available in the Kaggle repository, https://www.kaggle.com/datasets/piyushmishra1999/plantvillage-grape .
CRediT authorship contribution statement
Farian S. Ishengoma: Writing – review & editing, Writing – original draft, Visualization, Validation, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Neema N. Lyimo: Writing – review & editing,Open asset ↗Kaggle · plantvillage-grapelines:710-828Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 7 Sept 2026
Abstract Background Downy mildew is a plant disease that affects all cultivated European grapevine varieties. The disease is caused by the oomycete Plasmopara viticola . The current strategy to control this threat relies on repeated applications of fungicides. The most eco-friendly and sustainable alternative solution would be to use bred-resistant varieties. During breeding programs, some wild Vitis species have been used as resistance sources to introduce resistance loci in Vitis vinifera varieties. To ensure the durability of resistance, resistant varieties are built on combinations of these loci, some of which are unfortunately already overcome by virulent pathogen strains. The development of a high-throughput machine learning phenotyping method is now essential for identifying new resistance loci. Results Images of grapevine leaf discs infected with P. viticola were annotated with OIV 452–1 values, a standard scale, traditionally used by experts to assess resistance visually. This descriptor takes two variables into account the complete phenotype of the symptom: sporulation and necrosis. This annotated dataset was used to train neural networks. Various encoders were used to incorporate prior knowledge of the scale’s ordinality. The best results were obtained with the Swin transformer encoder which achieved an accuracy of 81.7%. Finally, from a biological point of view, the model described the studied trait and identified differences between genotypes in agreement with human observers, with an accuracy of 97% but at a high-throughput 650% faster than that of humans. Conclusion This work provides a fast, full pipeline for image processing, including machine learning, to describe the symptoms of grapevine leaf discs infected with P. viticola using the OIV 452–1, a two-symptom standard scale that considers sporulation and necrosis. If symptoms are frequently assessed by visual observation, which is time-consuming, low-throughput, tedious, and expert dependent, the method developed sweeps away all these constraints. This method could be extended to other pathosystems studied on leaf discs where disease symptoms are scored with ordinal scales.
Why it matches plant phenotyping methodsブドウ葉の病徴（胞子形成・壊死）を画像と深層学習で高スループット評価する手法を開発しており、植物表現型取得が研究の中心である。
abstractThe development of a high-throughput machine learning phenotyping method is now essential for identifying new resistance loci.
This study aimed to develop a new approach for predicting budburst and flowering with no dependency on GDD coupled with a GDD model to predict veraison with the capability of application for site-specific prediction of phenphases using geospatial functions. The budburst and flowering phases were predicted using probability statistical models based on threshold temperature occurrence and the chill requirement supplement was considered as a Boolean function. The field phenological data were recorded from irrigated vineyards with the same management systems. Based on the climatic conditions, four growing patterns of grapevine were defined over the study area, and phenological models were separately fitted for each growth pattern. The spatial analysis was performed by ArcGIS9x using linear models that fitted between the predicted phenophase and digital elevation model (DEM, grid cell 75 m). The results of the model indicated normalized RMSE and model efficiency of 12.5 %–0.93, 6.3 %–0.95, and 18.9 %–0.79, respectively for budburst, flowering, and veraison phases, with the absolute error of 1.12–1.75, which indicated high accuracy in the phenology estimation. The validated models of phenophase occurrence probability used historical weather data under different climatic conditions. The changes in probability during the days of the year were analyzed using different regression models. The best-fitted model for the budburst probability followed the sigmoid model under climatic conditions with mild (R²=0.98) or cool (R²=0.95) winter whereas the quadratic models (R²=0.98) were best-fitted under cold winter. The flowering model followed the quadratic model (R²=0.97) under all climatic conditions. The maps of phenophase characterized the site-specific time of budburst, flowering, and veraison.
Why it matches plant phenotyping methodsブドウの発芽・開花・着色期という植物フェノタイプを、確率モデル、GDD、地理空間解析で推定する手法を開発・検証し、精度評価と地図化まで行っており、フェノタイピング手法が中心である。
abstractThis study aimed to develop a new approach for predicting budburst and flowering with no dependency on GDD coupled with a GDD model to predict veraison with the capability of application for site-specific prediction of phenphases using geospatial functions.
With its ability to estimate yield, winemakers may better manage their vineyards and obtain important insights into the possible crop. The proper estimation of grape output is contingent upon an accurate evaluation of the morphology of the vine canopy, as this has a substantial impact on the final product. This study’s main goals were to gather canopy morphology data using a sophisticated 3D model and assess how well different morphology characteristics predicted yield results. An unmanned aerial vehicle (UAV) with an RGB camera was used in the vineyards of Topoľčianky, Slovakia, to obtain precise orthophotos of individual vine rows. Following the creation of an extensive three-dimensional (3D) model of the assigned region, a thorough examination was carried out to determine many canopy characteristics, including thickness, side section dimensions, volume, and surface area. According to the study, the best combination for predicting grape production was the side section and thickness. Using more than one morphological parameter is advised for a more precise yield estimate as opposed to depending on only one.
Why it matches plant phenotyping methodsUAV画像から3Dモデルを作成し、ブドウ樹冠の形態形質（厚さ、断面寸法、体積、表面積）を抽出・評価する手法が研究の中心であり、収量予測へ応用している。
abstractThis study’s main goals were to gather canopy morphology data using a sophisticated 3D model and assess how well different morphology characteristics predicted yield results.
Grape leaf diseases significantly threaten global viticulture, leading to substantial declines in grape yield. These diseases, attributed to various factors such as bacteria, viruses, fungi, and pests like mealybugs, can propagate through diverse means, including wind, rain, and human activities. They identify symptoms such as leaf spots and yellowing for effective disease management. Current strategies for disease control involve biological agents and chemical pesticides, with challenges arising from the lack of treatment options for certain viral infections like grapevine fan leaf and rupestris stem pitting. This review explores the myriad methods employed for grape leaf disease detection, emphasizing the use of emerging technologies to address these challenges. Integrating IoT (Internet of Things) and image analysis has demonstrated efficacy in disease detection, enabling farmers to make informed decisions promptly. While fungicides prove effective against diseases like downy mildew and powdery mildew, viral infections present persistent challenges. Technological interventions, particularly IoT and image processing, offer promising avenues for identifying grape leaf diseases, providing farmers with timely and valuable information to enhance disease management strategies. The review underscores the importance of continued research and technological innovation to address the complexities of grape leaf diseases and ensure sustainable grape cultivation.
Why it matches plant phenotyping methodsブドウ葉の病徴を対象に、画像解析やIoTを用いた病害検出手法をレビューしており、植物の病害状態を観測・推定する方法が中心である。
titleA Comparative Analysis of Grape Plant Leaf Disease Detection - Methods and Challenges
The discovery of well-preserved fossil Vitis L. seeds from the Gelasian stage in Italy has provided a unique opportunity to investigate the systematics of fossilized Vitis species. Through seed image analyses and elliptical Fourier transforms of fossil Vitis seeds from the sites Buronzo-Gifflenga and Castelletto Cervo II, we pointed out a strong relationship to the group of extant Eurasian Vitis species. However, classification analyses highlighted challenges in accurately assigning the fossil grape seeds to specific modern species. Morphological comparisons with modern Vitis species revealed striking similarities between the fossil seeds and V. vinifera subsp. sylvestris , as well as several other wild species from Asia. This close morphological resemblance suggests the existence of a population of V. vinifera sensu lato in Northen Italy during the Gelasian. These findings contributed to our understanding of the evolution and the complex interplay between ancient and modern Vitis species.
Why it matches plant phenotyping methods化石ブドウ種子の形態を画像解析と楕円フーリエ変換で定量・分類しており、種子器官の形質抽出が研究の中心であるため。
titleMorphological Characterization of Fossil Vitis L. Seeds from the Gelasian of Italy by Seed Image Analysis.
The "EscaYard" dataset comprises multimodal data collected from vineyards to support agricultural research, specifically focusing on vine health and productivity. Data collection involved two primary methods: (1) unmanned aerial vehicle (UAV) for capturing multispectral images and 3D point clouds, and (2) smartphones for detailed ground-level photography. The UAV used was DJI Matrice 210 V2 RTK, equipped with a Micasense Altum sensor, flying at 30 m above ground level to ensure detailed coverage. Ground-level data were collected using smartphones (iPhone X and Xiaomi Poco X3 Pro), which provided high-resolution images of individual plants. These images were geotagged, enabling location mapping, and included data on the phytosanitary status and number of grape clusters per plant. Additionally, the dataset contains RTK GNSS data, offering high-precision location information for each vine, enhancing the dataset's value for spatial analysis. Moreover, the dataset is structured to support various research applications, including agronomy, remote sensing, and machine learning. It is particularly suited for studying disease detection, yield estimation, and vineyard management strategies. The high-resolution and multispectral nature of the data allows for a detailed analysis of vineyard conditions. Potential reuse of the dataset spans multiple disciplines, enabling studies on environmental monitoring, geographic information systems (GIS), and precision agriculture. Its comprehensive nature makes it a valuable resource for developing and testing algorithms for disease classification, yield prediction, and plant phenotyping. For instance, the images of bunches and grape leaves can be used to train object detection algorithms for accurate disease detection and consequent precise spraying. Moreover, yield prediction algorithms can be trained by extracting the phenotypic traits of the grape bunches. The "EscaYard" dataset provides a foundation for advancing research in sustainable farming practices, optimising crop health, and improving productivity through precise agricultural technologies.
Why it matches plant phenotyping methodsブドウの病徴・生産性・房形質を対象とするマルチモーダル画像／UAVデータセットであり、植物フェノタイピングや病害・収量推定アルゴリズムの開発と評価を主目的としているため。
abstractThe "EscaYard" dataset provides a foundation for advancing research in sustainable farming practices, optimising crop health, and improving productivity through precise agricultural technologies.
Reproduction assets foundThe paper is a Data in Brief article describing the EscaYard dataset, publicly deposited on Zenodo with explicit DOI and direct URL. The dataset contains the paper's own phenotyping measurements (geotagged smartphone images, phytosanitary status, grape cluster counts, UAV orthomosaics, 3D point clouds, RTK GNSS trunk-Dataset · publics
City/Town/Region: Tomiño, Pontevedra, Galicia
Country: Spain
Coordinates: Vineyard B7, X: 517183.8, Y: 4645072.8; Vineyard B9, X: 516987.8, Y: 4644823.7 (ETRS89 / UTM zone 29N, EPSG:25829).
Data accessibility
Repository name: Zenodo
Data identification number: https://zenodo.org/doi/10.5281/zenodo.10362567
Direct URL to data: https://zenodo.org/records/10362567
1.
Value of the Data
•
The dataset offers a unique combination of multimodal data, including geotagged smartphone images, UAV orthomosaics, 3D point clouds, and precise geolocation data, enabling a multifaceted analysis of vineyard health and productivity.
•Open asset ↗Zenodo · 10.5281/zenodo.10362567lines:1-49Plant phenotyping relevance match · UnverifiedEurope PMC · checked 7 Sept 2026
In the framework of precision viticulture, satellite data have been demonstrated to significantly support many tasks. Specifically, they enable the rapid, large-scale estimation of some viticultural parameters like vine stem water potential (Ψstem) and intercepted solar radiation (ISR) that traditionally require time-consuming ground surveys. The practice of covering table grape vineyards with plastic films introduces an additional challenge for estimation, potentially affecting vine spectral responses and, consequently, the accuracy of estimations from satellites. This study aimed to address these challenges with a special focus on the exploitation of Sentinel-2 Level 2A and meteorological data to monitor a plastic-covered vineyard in Southern Italy. Estimates of Ψstem and ISR were obtained using different algorithms, namely, Ordinary Least Square (OLS), Multivariate Linear Regression (MLR), and machine learning (ML) techniques, which rely on Random Forest Regression, Support Vector Regression, and Partial Least Squares. The results proved that, despite the potential spectral interference from the plastic coverings, ISR and Ψstem can be locally estimated with a satisfying accuracy. In particular, (i) the OLS regression-based approach showed a good performance in providing accurate ISR estimates using the near-infrared spectral bands (RMSE < 8%), and (ii) the MLR and ML algorithms could estimate both the ISR and vine water status with a higher accuracy (RMSE < 7 for ISR and RMSE < 0.14 MPa for Ψstem). These results encourage the adoption of medium-high resolution multispectral satellite imagery for deriving satisfying estimates of key crop parameters even in anomalous situations like the ones where plastic films cover the monitored vineyard, thus marking a significant advancement in precision viticulture.
Why it matches plant phenotyping methodsSentinel-2データと複数の回帰・機械学習手法により、ブドウの茎水ポテンシャルと遮光放射を推定し、精度比較・検証しているため、植物形質取得手法が中心である。
abstractEstimates of Ψstem and ISR were obtained using different algorithms, namely, Ordinary Least Square (OLS), Multivariate Linear Regression (MLR), and machine learning (ML) techniques
The cultivation of grapes encounters various challenges, such as the presence of pests and diseases, which have the potential to considerably diminish agricultural productivity. Plant diseases pose a significant impediment, resulting in diminished agricultural productivity and economic setbacks, thereby affecting the quality of crop yields. Hence, the precise and timely identification of plant diseases holds significant importance. This study employs a Convolutional neural network (CNN) with and without data augmentation, in addition to a DCNN Classifier model based on VGG16, to classify grape leaf diseases. A publicly available dataset is utilized for the purpose of investigating diseases affecting grape leaves. The DCNN Classifier Model successfully utilizes the strengths of the VGG16 model and modifies it by incorporating supplementary layers to enhance its performance and ability to generalize. Systematic evaluation of metrics, such as accuracy and F1-score, is performed. With training and test accuracy rates of 99.18 and 99.06%, respectively, the DCNN Classifier model does a better job than the CNN models used in this investigation. The findings demonstrate that the DCNN Classifier model, utilizing the VGG16 architecture and incorporating three supplementary CNN layers, exhibits superior performance. Also, the fact that the DCNN Classifier model works well as a decision support system for farmers is shown by the fact that it can quickly and accurately identify grape diseases, making it easier to take steps to stop them. The results of this study provide support for the reliability of the DCNN classifier model and its potential utility in the field of agriculture.
Why it matches plant phenotyping methodsブドウ葉の画像から病害状態を推定するCNN/DCNN手法の開発・比較評価が研究の中心であり、植物病害フェノタイピングに該当する。
abstractThis study employs a Convolutional neural network (CNN) with and without data augmentation, in addition to a DCNN Classifier model based on VGG16, to classify grape leaf diseases.
This study delves into the analysis of a vineyard in Carinthia, Austria, focusing on the automated derivation of ecosystem structures of individual vine parameters, including vine heights, leaf area index (LAI), leaf surface area (LSA), and the geographic positioning of single plants. For the derivation of these parameters, intricate segmentation processes and nuanced UAS-based data acquisition techniques are necessary. The detection of single vines was based on 3D point cloud data, generated at a phenological stage in which the plants were in the absence of foliage. The mean distance from derived vine locations to reference measurements taken with a GNSS device was 10.7 cm, with a root mean square error (RMSE) of 1.07. Vine height derivation from a normalized digital surface model (nDSM) using photogrammetric data showcased a strong correlation (R2 = 0.83) with real-world measurements. Vines underwent automated classification through an object-based image analysis (OBIA) framework. This process enabled the computation of ecosystem structures at the individual plant level post-segmentation. Consequently, it delivered comprehensive canopy characteristics rapidly, surpassing the speed of manual measurements. With the use of uncrewed aerial systems (UAS) equipped with optical sensors, dense 3D point clouds were computed for the derivation of canopy-related ecosystem structures of vines. While LAI and LSA computations await validation, they underscore the technical feasibility of obtaining precise geometric and morphological datasets from UAS-collected data paired with 3D point cloud analysis and object-based image analysis.
Why it matches plant phenotyping methodsUAS画像・3D点群・OBIAを用いて個体レベルのブドウ形態形質を自動抽出し、実測値で検証しており、フェノタイピング手法が中心である。
abstractthe automated derivation of ecosystem structures of individual vine parameters, including vine heights, leaf area index (LAI), leaf surface area (LSA), and the geographic positioning of single plants
Image and video tools for analyzing crop scenes and plants are essential for applying precision agriculture to crop maintenance, harvesting, or pruning. In this paper, we are interested in vine pruning, a task that requires a precise understanding of the vine structure with branch type identification, orientations, and node locations. However, estimating such a structure is highly challenging, given the large variety in grapevine appearances, lighting conditions, viewpoint, the interweaving of branches, occlusions, and the level of details needed. To address these challenges, we propose ViNet: a deep-learning approach for estimating the structure of grapevine, which comprises two main steps: The first one detects nodes and identifies the branch types of the plant, as well as the spatial relation between them, whilst the second one uses the extracted nodes and branches to build a graph, out of which the structure of the grapevine is inferred. In doing so, we make four main contributions: (i) we put forward for the first time a method for automatic segmentation and extraction of the grapevine structure from images; (ii) we propose a novel approach leveraging the powerful stacked hourglass network to infer node location, branch types and the spatial relationships between them; (iii) we propose a novel shortest path weighted graph optimization step to extract connections between nodes and infer the structure, allowing to address the problem of having an unknown number of branches in the tree; (iv) we publicly release a dataset of more than 1500 grapevine images fully annotated with the structure information. Extensive experiments on this dataset demonstrate the efficiency of our approach at predicting the structure of a grapevine, achieving a precision and recall for node prediction of 95% and 90%, respectively, as well as ablation studies validating our design choices.
Why it matches plant phenotyping methods画像からブドウ樹の節・枝型・空間関係を抽出し、植物構造を推定する手法を開発・検証しており、注釈付き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 ViNet: a deep-learning approach for estimating the structure of grapevine
This study explores spectroscopy in the 350 to 2500 nm range for detecting powdery mildew (Erysiphe necator) in grapevine leaves, crucial for precision agriculture and sustainable vineyard management. In a controlled experimental vineyard setting, the spectral reflectance on leaves with varying infestation levels was measured using a FieldSpec 4 spectroradiometer during July and September. A detailed assessment was conducted following the guidelines recommended by the European and Mediterranean Plant Protection Organization (EPPO) to quantify the level of infestation; categorising leaves into five distinct grades based on the percentage of leaf surface area affected. Subsequently, spectral data were collected using a contact probe with a tungsten halogen bulb connected to the spectroradiometer, taking three measurements across different areas of each leaf. Partial Least Squares Regression (PLSR) analysis yielded coefficients of determination R2 = 0.74 and 0.71, and Root Mean Square Errors (RMSEs) of 12.1% and 12.9% for calibration and validation datasets, indicating high accuracy for early disease detection. Significant spectral differences were noted between healthy and infected leaves, especially around 450 nm and 700 nm for visible light, and 1050 nm, 1425 nm, 1650 nm, and 2250 nm for the near-infrared spectrum, likely due to tissue damage, chlorophyll degradation and water loss. Finally, the Powdery Mildew Vegetation Index (PMVI) was introduced, calculated as PMVI = (R755 − R675)/(R755 + R675), where R755 and R675 are the reflectances at 755 nm (NIR) and 675 nm (red), effectively estimating disease severity (R2 = 0.7). The study demonstrates that spectroscopy, combined with PMVI, provides a reliable, non-invasive method for managing powdery mildew and promoting healthier vineyards through precision agriculture practices.
Why it matches plant phenotyping methodsブドウ葉の病害状態・重症度を分光反射から推定する手法を開発・検証しており、植物表現型の取得と定量化が研究の中心である。
abstractThis study explores spectroscopy in the 350 to 2500 nm range for detecting powdery mildew (Erysiphe necator) in grapevine leaves
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 7 Sept 2026
Abstract Crop pests have profoundly deleterious effects on crop yield and food security. However, conventional pest control depends heavily on the utilization of insecticides, which develops strong pesticide resistance and concerns of food safety. Crop and their wild relatives display diverse levels of pest resistance, indicating the feasibility for breeding of pest-resistant crop varieties. In this study, we integrate deep learning (DL)/machine learning (ML) algorithms, plant phenomics and whole genome sequencing (WGS) data to conduct genomic selection (GS) of pest-resistance in grapevine. We employ deep convolutional neural networks (DCNN) to accurately calculate the severity of damage by pests on grape leaves, which achieves a classification accuracy of 95.3% (Visual Geometry Group 16, VGG16, for binary trait) and a correlation coefficient of 0.94 in regression analysis (DCNN with Pest Damage Score, DCNN-PDS, for continuous trait). We apply DL models to predict and integrate phenotype (both binary and continuous) along with WGS data from 231 grape accessions, conducting Genome-Wide Association Studies (GWAS). This analysis detects a total of 69 QTLs, encompassing 139 candidate genes involved in pathways associated with pest resistance, including jasmonic acid (JA), salicylic acid (SA), ethylene, and other related pathways. Furthermore, through the combination with transcriptome data, we identify specific pest-resistant genes, such as ACA12 and CRK3 , which play distinct roles in resisting herbivore attacks. Machine learning-based GS demonstrates a high accuracy (95.7%) and a strong correlation (0.90) in predicting the leaf area damaged by pests as binary and continuous traits in grapevine, respectively. In general, our study highlights the power of DL/ML in plant phenomics and GS, facilitating genomic breeding of pest-resistant grapevine.
Why it matches plant phenotyping methodsブドウ葉の害虫被害面積・重症度を深層学習で画像から定量推定し、精度検証と育種への応用を行っており、表現型取得・抽出手法が中心的です。
abstractWe employ deep convolutional neural networks (DCNN) to accurately calculate the severity of damage by pests on grape leaves, which achieves a classification accuracy of 95.3% (Visual Geometry Group 16, VGG16, for binary trait) and a correlation coefficient of 0.94 in regression analysis (DCNN with Pest Damage Score, DCNN-PDS, for continuous trait).
O_LIGrapevine leaves are a model morphometric system. Sampling over ten thousand leaves using dozens of landmarks, the genetic, developmental, and environmental basis of leaf shape has been studied and a morphospace for the genus Vitis predicted. Yet, these representations of leaf shape fail to capture the exquisite features of leaves at high resolution. C_LIO_LIWe measure the shapes of 139 grapevine leaves using 1672 pseudo-landmarks derived from 90 homologous landmarks with Procrustean approaches. From hand traces of the vasculature and blade, we have derived a method to automatically detect landmarks and place pseudo-landmarks that results in a high-resolution representation of grapevine leaf shape. Using polynomial models, we create continuous representations of leaf development in 10 Vitis spp. C_LIO_LIWe visualize a high-resolution morphospace in which genetic and developmental sources of leaf shape variance are orthogonal to each other. Using classifiers, V. vinifera, Vitis spp., rootstock and dissected leaf varieties as well as developmental stages are accurately predicted. Theoretical eigenleaf representations sampled from across the morphospace that we call synthetic leaves can be classified using models. C_LIO_LIBy predicting a high-resolution morphospace and delimiting the boundaries of leaf shapes that can plausibly be produced within the genus Vitis, we can sample synthetic leaves with realistic qualities. From an ampelographic perspective, larger numbers of leaves sampled at lower resolution can be projected onto this high-resolution space; or, synthetic leaves can be used to increase the robustness and accuracy of machine learning classifiers. C_LI Societal Impact StatementGrapevine leaves are emblematic of the strong visual associations people make with plants. At a glance, leaf shape is immediately recognizable, and it is because of this reason it is used to distinguish grape varieties. In an era of computationally-enabled, machine learning-derived representations of reality, we can revisit how we view and use the shapes and forms that plants display to understand our relationship with them. Using computational approaches combined with time-honored methods, we can predict theoretical leaves that are possible to understand the genetics, development, and environmental responses of plants in new ways.
Why it matches plant phenotyping methodsブドウ葉の形状を高解像度に取得・表現する自動ランドマーク検出と擬似ランドマーク配置法を開発しており、葉形状フェノタイピング手法が研究の中心である。
abstractwe have derived a method to automatically detect landmarks and place pseudo-landmarks that results in a high-resolution representation of grapevine leaf shape
The scientific progress in artificial intelligence and robotics has enabled precision viticulture to pursue sustainability and improve the final yield. For instance, monitoring the canopy volume of each plant can allow the correct ripening of the bunches. In this context, this paper proposes a novel approach for the characterization of biomass volume using images acquired in a vineyard with the low-cost Azure Kinect RGB-D camera. Semantic image segmentation is implemented using three encoder–decoder deep architectures (U-Net, DeepLabV3+, and MANet) to produce accurate masks of the vine leaf structure. In a transfer learning approach, a public dataset acquired with the Intel RealSense D435 depth camera is used to train the segmentation networks. Then, a complete pipeline to estimate possible changes in biomass volume is presented. Experiments are run to analyze the biomass removed during the trimming process of grapevine plants. The best segmentation result is obtained by the U-Net architecture with ResNet50 backbone, showing an accuracy of 92.10%, although the training and test sets consist of images acquired by different cameras. However, the DeepLabV3+ network with ResNeXt50 backbone, which scores an accuracy of 90.25% on the test set, gives the best estimate of the removed biomass, requiring the shortest time for training. These outcomes prove the potential capability of this automatic approach for controlling leaf growth and ensuring sustainable viticulture practices.
Why it matches plant phenotyping methodsRGB-D画像のセマンティックセグメンテーションと点群解析により、ブドウ樹の葉構造およびバイオマス体積を推定する手法を開発・評価しており、植物形質取得が中心です。
abstractthis paper proposes a novel approach for the characterization of biomass volume using images acquired in a vineyard with the low-cost Azure Kinect RGB-D camera.
High-throughput phenotyping of grapevine leafroll disease (GLD) at the canopy scale helps develop fast and effective management in viticulture. However, detecting GLD efficiently in a vineyard is challenging owing to the limited adaptation of prior art. Therefore, we propose a novel convolutional neural network called GLDCNet to improve GLD recognition using unmanned aerial vehicle–based imagery. The effectiveness of the GLDCNet is attributed to the four new network designs used and is validated through ablation experiments. The GLDCNet achieves a classification accuracy of 99.57% using the RGB dataset and obtains more efficient and accurate results than nine other state-of-the-art methods. Furthermore, we systematically evaluated the impacts of image spatial resolution and vegetation indexes on the classification performance of the model. Experimental results suggest that improving image spatial resolution is more cost-effective than enhancing multispectral information for improving GLD recognition. Our proposed method offers a rapid, scalable, and accurate diagnostic protocol for detecting GLD in vineyards.
Why it matches plant phenotyping methodsUAV画像からブドウ樹の葉巻病状態を推定するCNNを開発し、アブレーション実験と他手法比較で検証しており、植物表現型取得・判定手法が中心です。
abstractHigh-throughput phenotyping of grapevine leafroll disease (GLD) at the canopy scale
Fresh grapes are characterized by a short shelf life and are often subjected to quality losses during post-harvest storage. The quality assessment of grapes using image analysis may be a useful approach using non-destructive methods. This study aimed to compare the effect of different storage methods on the grape image texture parameters of the fruit outer structure. Grape bunches were stored for 4 weeks using 3 storage methods (– 18 °C, + 4 °C, and room temperature) and then were subjected subsequently to image acquisition using a flatbed scanner and image processing. The models for the classification of fresh and stored grapes were built based on selected image textures using traditional machine learning algorithms. The fresh grapes and stored fruit samples (for 4 weeks) in the freezer, in the refrigerator and in the room were classified with an overall accuracy reaching 96% for a model based on selected texture parameters from images in color channels R, G, B, L, a, and b built using Random Forest algorithm. Among the individual color channels, the carried-out classification for the R color channel produced the highest overall accuracies of up to 92.5% for Random Forest. As a result, this study proposed an innovative approach combining image analysis and traditional machine learning to assess changes in the outer structure of grape berries caused by different storage conditions.
Why it matches plant phenotyping methodsブドウ果実の外表構造変化を、スキャナ画像のテクスチャ解析と機械学習で評価する手法が研究の中心であり、植物器官の状態を画像から推定しているため。
abstractThe quality assessment of grapes using image analysis may be a useful approach using non-destructive methods.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 7 Sept 2026
Abstract Quantifying healthy and degraded inner tissues in plants is of great interest in agronomy, for example, to assess plant health and quality and monitor physiological traits or diseases. However, detecting functional and degraded plant tissues in-vivo without harming the plant is extremely challenging. New solutions are needed in ligneous and perennial species, for which the sustainability of plantations is crucial. To tackle this challenge, we developed a novel approach based on multimodal 3D imaging and artificial intelligence-based image processing that allowed a non-destructive diagnosis of inner tissues in living plants. The method was successfully applied to the grapevine ( Vitis vinifera L.). Vineyard’s sustainability is threatened by trunk diseases, while the sanitary status of vines cannot be ascertained without injuring the plants. By combining MRI and X-ray CT 3D imaging with an automatic voxel classification, we could discriminate intact, degraded, and white rot tissues with a mean global accuracy of over 91%. Each imaging modality contribution to tissue detection was evaluated, and we identified quantitative structural and physiological markers characterizing wood degradation steps. The combined study of inner tissue distribution versus external foliar symptom history demonstrated that white rot and intact tissue contents are key-measurements in evaluating vines’ sanitary status. We finally proposed a model for an accurate trunk disease diagnosis in grapevine. This work opens new routes for precision agriculture and in-situ monitoring of tissue quality and plant health across plant species.
Why it matches plant phenotyping methodsブドウ樹内部組織の非破壊的な表現型取得・診断を目的に、MRI・X線CT・自動ボクセル分類を開発し、精度評価と組織状態の定量化を行っており、フェノタイピング手法が中心である。
abstractwe developed a novel approach based on multimodal 3D imaging and artificial intelligence-based image processing that allowed a non-destructive diagnosis of inner tissues in living plants.
Novel monitoring architecture approaches are required to detect viticulture diseases early. Existing micro-climate decision support systems can only cope with late detection from empirical and semi-empirical models that provide less accurate results. Such models cannot alleviate precision viticulture planning and pesticide control actions, providing early reconnaissances that may trigger interventions. This paper presents a new plant-level monitoring architecture called thingsAI. The proposed system utilizes low-cost, autonomous, easy-to-install IoT sensors for vine-level monitoring, utilizing the low-power LoRaWAN protocol for sensory measurement acquisition. Facilitated by a distributed cloud architecture and open-source user interfaces, it provides state-of-the-art deep learning inference services and decision support interfaces. This paper also presents a new deep learning detection algorithm based on supervised fuzzy annotation processes, targeting downy mildew disease detection and, therefore, planning early interventions. The authors tested their proposed system and deep learning model on the grape variety of protected designation of origin called debina, cultivated in Zitsa, Greece. From their experimental results, the authors show that their proposed model can detect vine locations and timely breakpoints of mildew occurrences, which farmers can use as input for targeted intervention efforts.
Why it matches plant phenotyping methodsブドウ個体レベルのIoT監視と深層学習により、うどんこ病ではなくべと病の発生を植物状態として検出するシステムとアルゴリズムを開発・評価しており、フェノタイピング手法が中心です。
abstractThis paper presents a new plant-level monitoring architecture called thingsAI.
Viticulture and associated products are an important part of the economy in many countries. However, biotic and abiotic stresses impact negatively the production of grapes and wine. Climate change is in many aspects increasing both these stresses. Routine sample retrievals and analysis tend to be time-consuming and require expensive equipment and skilled personnel to operate. These challenges could be overcome through the development of a miniaturized analytic device for early detection of grapevine stresses in the field. Abscisic acid is involved in several plant processes, including the onset of fruit ripening and tolerance mechanisms against drought stress. This hormone can be detected through a competitive immunoassay and is found in plants in concentrations up to 10 -1 mg/mL. A microfluidic platform is developed in this work which can detect a minimum of 10 -11 mg/mL of abscisic acid in buffer. Grape samples were tested using the microfluidic system alongside benchmark techniques such as high-performance liquid chromatography. The microfluidic system could detect the increase to 10 -5 mg/mL of abscisic acid present in real berry samples at the veraison stage of ripening.
Why it matches plant phenotyping methodsブドウのストレスや成熟状態に関連するアブシジン酸を現場で測定するマイクロ流体免疫測定プラットフォームの開発・ベンチマークが中心であり、植物の生理状態を推定するフェノタイピング手法に該当する。
abstractGrape samples were tested using the microfluidic system alongside benchmark techniques such as high-performance liquid chromatography.
This research revolutionizes grapevine security worldwide and sustains premium wine production by using CNN-driven algorithms and different datasets to pioneer multimodal detection for early Esca disease in grapevines. Also give a model with more accuracy so that we can predict this plant disease early. Global grapevine output is being threatened by the complicated fungal illness known as esca disease, which also threatens the stability of the economy and the quality of premium wines. The capacity of current detection techniques to detect Esca to detect the disease early is restricted and frequently imprecise. By utilizing the capabilities of Convolutional Neural Networks (CNNs) and a variety of datasets, this study offers a novel method that develops multimodal detection for early Esca diagnosis. Compared to traditional methods, our model achieves higher accuracy by combining spectral and temporal information with visual imaging of leaves and stems. Key Words: CNN, ESCA, Machine Learning, Plant Disease
Why it matches plant phenotyping methodsブドウ樹の葉・茎の画像とスペクトル・時系列情報からEsca病を早期推定するCNNベースのマルチモーダル手法が研究の中心であり、植物の病徴・病害状態を直接推定するため、植物フェノタイピング手法に該当する。
abstractusing CNN-driven algorithms and different datasets to pioneer multimodal detection for early Esca disease in grapevines
Current infrared spectroscopy applications in the field of viticulture are moving toward direct in-field measuring techniques. However, limited research is available on quantitative applications using direct measurement of fresh tissue. The few studies conducted have combined the spectral data from various cultivars, growing regions, grapevine organs, and phenological stages during model development. The spectral data from these heterogeneous samples are combined into a single data set and analyzed jointly during quantitative analysis. Combining the spectral information of these diverse samples into a global data set could be an unsuitable approach and could yield less accurate prediction results. Spectral differences among samples could be overlooked during model development and quantitative analysis. The development of specialized calibrations should be considered and could lead to more accurate quantitative analyses. This study explored a model optimization strategy attempting global and specialized calibrations. Global calibrations, containing data from multiple organs, berry phenological, and shoot lignification stages, were compared to specialized calibrations per organ or stage. The global calibration for organs contained data from shoots, leaves, and berries and produced moderately accurate prediction results for nitrogen, carbon, and hydrogen. The specialized calibrations per organ yielded more accurate calibrations with a coefficient of determination in validation (R 2 val) at 90.65% and a root mean square error of prediction (RMSEP) at 0.32% dry matter (DM) for the berries' carbon calibrations. The leaves and shoots carbon calibrations had R 2 val and RMSEP at 84.99%, 0.34% DM, and 90.06%, 0.37% DM, respectively. The specialized calibrations for nitrogen and hydrogen showed similar improvements in prediction accuracy per organ. Specialized calibrations per phenological and lignification stage were also explored. Not all stages showed improvement, however, most stages had comparable or improved results for the specialized calibrations compared to the global calibrations containing all phenological or lignification stages. The results indicated that both global and specialized calibrations should be considered during model development to optimize prediction accuracy.
Why it matches plant phenotyping methods近赤外分光によるブドウ器官の栄養成分・生育段階・木化状態の定量推定について、グローバル／特化キャリブレーションを開発・検証しており、植物形質取得法が研究の中心である。
abstractThis study explored a model optimization strategy attempting global and specialized calibrations.
Introduction Grapes are prone to various diseases throughout their growth cycle, and the failure to promptly control these diseases can result in reduced production and even complete crop failure. Therefore, effective disease control is essential for maximizing grape yield. Accurate disease identification plays a crucial role in this process. In this paper, we proposed a real-time and lightweight detection model called Fusion Transformer YOLO for 4 grape diseases detection. The primary source of the dataset comprises RGB images acquired from plantations situated in North China. Methods Firstly, we introduce a lightweight high-performance VoVNet, which utilizes ghost convolutions and learnable downsampling layer. This backbone is further improved by integrating effective squeeze and excitation blocks and residual connections to the OSA module. These enhancements contribute to improved detection accuracy while maintaining a lightweight network. Secondly, an improved dual-flow PAN+FPN structure with Real-time Transformer is adopted in the neck component, by incorporating 2D position embedding and a single-scale Transformer Encoder into the last feature map. This modification enables real-time performance and improved accuracy in detecting small targets. Finally, we adopt the Decoupled Head based on the improved Task Aligned Predictor in the head component, which balances accuracy and speed. Results Experimental results demonstrate that FTR-YOLO achieves the high performance across various evaluation metrics, with a mean Average Precision (mAP) of 90.67%, a Frames Per Second (FPS) of 44, and a parameter size of 24.5M. Conclusion The FTR-YOLO presented in this paper provides a real-time and lightweight solution for the detection of grape diseases. This model effectively assists farmers in detecting grape diseases.
Why it matches plant phenotyping methodsブドウ葉・植物画像から病害状態を推定するリアルタイム画像解析モデルを開発し、精度と速度を評価しており、病害表現型の取得手法が中心である。
abstractwe proposed a real-time and lightweight detection model called Fusion Transformer YOLO for 4 grape diseases detection.
Why it matches plant phenotyping methodsブドウ葉の薬剤効果・病害状態を非破壊的に推定するハイパースペクトルセンシング手法を評価・検証しており、植物状態の取得が研究の中心です。
abstractThe goal of this study was to evaluate if hyperspectral sensing in the visible to shortwave infrared range (400 to 2,400 nm) can quantify foliar fungicide efficacy on grape leaves.
The scientific progress in artificial intelligence and robotics has enabled precision viticulture to pursue sustainability and improve the final yield. For instance, monitoring the canopy volume of each plant can allow the correct ripening of the bunches. In this context, this paper proposes a novel approach for the characterization of biomass volume using images acquired in a vineyard with the low-cost Azure Kinect RGB-D camera. Semantic image segmentation is implemented using three encoder–decoder deep architectures (U-Net, DeepLabV3+, and MANet) to produce accurate masks of the vine leaf structure. In a transfer learning approach, a public dataset acquired with the Intel RealSense D435 depth camera is used to train the segmentation networks. Then, a complete pipeline to estimate possible changes in biomass volume is presented. Experiments are run to analyze the biomass removed during the trimming process of grapevine plants. The best segmentation result is obtained by the U-Net architecture with ResNet50 backbone, showing an accuracy of 92.10%, although the training and test sets consist of images acquired by different cameras. However, the DeepLabV3+ network with ResNeXt50 backbone, which scores an accuracy of 90.25% on the test set, gives the best estimate of the removed biomass, requiring the shortest time for training. These outcomes prove the potential capability of this automatic approach for controlling leaf growth and ensuring sustainable viticulture practices.
Why it matches plant phenotyping methodsブドウ樹の葉構造を画像セグメンテーションし、バイオマス体積と剪定による変化を推定する手法・パイプラインが研究の中心であるため。
abstractthis paper proposes a novel approach for the characterization of biomass volume using images acquired in a vineyard with the low-cost Azure Kinect RGB-D camera.
Our approach includes picture preprocessing, feature extraction utilizing the SqueezeNet model, hyperparameter optimisation utilising the Equilibrium Optimizer (EO) algorithm, and classification utilising a Stacked Autoencoder (SAE) model. Each of these processes is carried out in a series of separate steps. During the image preprocessing stage, contrast limited adaptive histogram equalisations (CLAHE) is utilized to improve the contrasts, and Adaptive Bilateral Filtering (ABF) to get rid of any noise that may be present. The SqueezeNet paradigm is utilized to obtain relevant characteristics from the pictures that have been preprocessed, and the EO technique is utilized to fine-tune the hyperparameters. Finally, the SAE model categorises the diseases that affect the grape leaf. The simulation analysis of the EODTL-GLDC technique tested New Plant Diseases Datasets and the results were inspected in many prospects. The results demonstrate that this model outperforms other deep learning techniques and methods that are more often related to machine learning. Specifically, this technique was able to attain a precision of 96.31% on the testing datasets and 96.88% on the training data set that was split 80:20. These results offer more proof that the suggested strategy is successful in automating the detection and categorization of grape leaf diseases.
Why it matches plant phenotyping methodsブドウ葉の画像から病害状態を推定・分類する画像解析手法が研究の中心であり、植物の病害表現型を直接評価しているため。
abstractOur approach includes picture preprocessing, feature extraction utilizing the SqueezeNet model, hyperparameter optimisation utilising the Equilibrium Optimizer (EO) algorithm, and classification utilising a Stacked Autoencoder (SAE) model.
GrapevineLeafPhysiological trait estimationWater status / transpiration
Abstract Quantifying drought tolerance in crops is critical for agricultural management under environmental change, and drought response traits in wine grapes have long been the focus of viticultural research. Turgor loss point ( π tlp ) is gaining attention as an indicator of drought tolerance in plants, though estimating π tlp often requires the construction and analysis of pressure-volume (P-V) curves which is time consuming. While P-V curves remain a valuable tool for assessing π tlp and related traits, there is considerable interest in developing high-throughput methods for rapidly estimating π tlp , especially in the context of crop screening. We tested the ability of a dewpoint hygrometer to quantify variation in π tlp across and within 12 varieties of wine grapes ( Vitis vinifera ) and one wild relative ( Vitis riparia ) and compared these results to those derived from P-V curves. At the leaf-level, methodology explained only 4–5% of the variation in π tlp while variety/species identity accounted for 39% of the variation, indicating that both methods are sensitive to detecting intraspecific π tlp variation in wine grapes. Also at the leaf level, π tlp measured using a dewpoint hygrometer significantly approximated π tlp values ( r 2 = 0.254) and conserved π tlp rankings from P-V curves (Spearman’s ρ = 0.459). While the leaf-level datasets differed statistically from one another (paired t -test p = 0.01), average difference in π tlp for a given pair of leaves was small (0.1 ± 0.2 MPa (s.d.)). At the species/variety level, estimates of π tlp measured by the two methods were also statistically correlated ( r 2 = 0.304), did not deviate statistically from a 1:1 relationship, and conserved π tlp rankings across varieties (Spearman’s ρ = 0.692). The dewpoint hygrometer (taking ~ 10–15 minutes on average per measurement) captures fine-scale intraspecific variation in π tlp , with results that approximate those from P-V curves (taking 2–3 hours on average per measurement). The dewpoint hygrometer represents a viable method for rapidly estimating intraspecific variation in π tlp , and potentially greatly increasing replication when estimating this drought tolerance trait in wine grapes and other crops.
Why it matches plant phenotyping methodsブドウ葉の乾燥耐性形質（膨圧損失点）を高スループットに推定する露点湿度計法を開発・圧力容積曲線法と比較検証しており、フェノタイピング手法が中心である。
titleA high-throughput approach for quantifying turgor loss point in wine grapes
Grapevine is among the most economically important crops suffering environmental constraints, including drought and salt stress. Although imaging is increasingly used to detect abiotic stress in agriculture, image-based phenotyping in grapevine still needs optimisation. This study presents the RGB-(red, green, blue)-based phenotyping of the early stage of salt stress response in potted grapevine (Aleatico/SO4) irrigated with saline water (100 mM NaCl) for 9 days in contrast with vines irrigated with fresh water. The response was measured using stomatal conductance (gs), net photosynthetic rate (A), transpiration (E), maximum potential photosynthetic efficiency (Fv/Fm), stem water potential (SWP) concurrently with RGB imaging via a robotised platform.The image-based phenotyping of salt-stressed vines employed two sets of measurements: (i) the pixel fraction of specific colour bands (Yellow, Green, Brown and Dark Green) and (ii) the mean pixel value of R, G and B and other RGB-based colorimetric indexes. Results show that the responses of gs, A, E, Fv/Fm were closely related to increasing soil electrical conductivity (EC) and that imaging could detect the EC threshold of approx. 4 dS m-1 causing a ~60 % decrease in these physiological traits compared to the pre-stress level. The SWP declined to about –0.7 MPa at the end of the experiment. The change of the relative pixel fraction of Dark Green to increasing EC has been analysed within a dose-response context, showing that a decrease of 1 % of the Dark Green colour band corresponded to the 4 dS m-1 EC threshold. This study also examined the use of the mean pixel value of the R, G and B channels as proxies of EC along with new RGB-based indexes resulting from the rearrangement of original R, G and B mean pixel values. Results show the suitability of the mean pixel value of R and Coloration Index [(R-B)/R] to serve as predictors of EC (R2 >= 0.80).
Why it matches plant phenotyping methodsRGB画像とロボット化プラットフォームを用いてブドウの塩ストレス表現型を抽出し、画素特徴量を生理指標・土壌ECと関連付けて評価することが中心であるため。
abstractThis study presents the RGB-(red, green, blue)-based phenotyping of the early stage of salt stress response in potted grapevine
High-throughput phenotyping of grapevine leafroll disease (GLD) at the canopy scale helps develop fast and effective management in viticulture. However, detecting GLD efficiently in a vineyard is challenging owing to the limited adaptation of prior art. Therefore, we propose a novel convolutional neural network called GLDCNet to improve GLD recognition using unmanned aerial vehicle–based imagery. The effectiveness of the GLDCNet is attributed to the four new network designs used and is validated through ablation experiments. The GLDCNet achieves a classification accuracy of 99.57% using the RGB dataset and obtains more efficient and accurate results than nine other state-of-the-art methods. Furthermore, we systematically evaluated the impacts of image spatial resolution and vegetation indexes on the classification performance of the model. Experimental results suggest that improving image spatial resolution is more cost-effective than enhancing multispectral information for improving GLD recognition. Our proposed method offers a rapid, scalable, and accurate diagnostic protocol for detecting GLD in vineyards.
Why it matches plant phenotyping methodsUAV画像からブドウ樹の葉巻病状態を推定するCNNを開発し、アブレーション実験、既存手法との比較、解像度・植生指数の評価で検証しており、植物病害表現型の取得手法が中心である。
abstractwe propose a novel convolutional neural network called GLDCNet to improve GLD recognition using unmanned aerial vehicle–based imagery.
Close-range remote sensing techniques employing multispectral sensors on unoccupied aerial vehicles (UAVs) offer both advantages and drawbacks in comparison to traditional remote sensing using satellite-mounted sensors. Close-range remote sensing techniques have been increasingly used in the field of precision agriculture. Planning the flight, including optimal flight altitudes, can enhance both geometric and temporal resolution, facilitating on-demand flights and the selection of the most suitable time of day for various applications. However, the main drawbacks stem from the lower quality of the sensors being used compared to satellites. Close-range sensors can capture spectral responses of plants from multiple viewpoints, mitigating satellite remote sensing challenges, such as atmospheric interference, while intensifying issues such as bidirectional reflectance distribution function (BRDF) effects due to diverse observation angles and morphological variances associated with flight altitude. This paper introduces a methodology for achieving high-quality vegetation indices under varied observation conditions, enhancing reflectance by selectively utilizing well-geometry vegetation pixels, while considering factors such as hotspot, occultation, and BRDF effects. A non-parametric ANOVA analysis demonstrates significant statistical differences between the proposed methodology and the commercial photogrammetric software AgiSoft Metashape, in a case study of a vineyard in Fuente-Alamo (Albacete, Spain). The BRDF model is expected to substantially improve vegetation index calculations in comparison to the methodologies used in satellite remote sensing and those used in close-range remote sensing.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像から植物の反射特性・植生指数を抽出する方法を開発し、既存ソフトウェアと比較検証しており、植物表現型取得が中心である。
abstractThis paper introduces a methodology for achieving high-quality vegetation indices under varied observation conditions
Deep learning plays a vital role in precise grapevine disease detection, yet practical applications for farmer assistance are scarce despite promising results. The objective of this research is to develop an intelligent approach, supported by user-friendly, open-source software named AI GrapeCare (Version 1, created by Osama Elsherbiny). This approach utilizes RGB imagery and hybrid deep networks for the detection and prevention of grapevine diseases. Exploring the optimal deep learning architecture involved combining convolutional neural networks (CNNs), long short-term memory (LSTM), deep neural networks (DNNs), and transfer learning networks (including VGG16, VGG19, ResNet50, and ResNet101V2). A gray level co-occurrence matrix (GLCM) was employed to measure the textural characteristics. The plant disease detection platform (PDD) created a dataset of real-life grape leaf images from vineyards to improve plant disease identification. A data augmentation technique was applied to address the issue of limited images. Subsequently, the augmented dataset was used to train the models and enhance their capability to accurately identify and classify plant diseases in real-world scenarios. The analyzed outcomes indicated that the combined CNN RGB -LSTM GLCM deep network, based on the VGG16 pretrained network and data augmentation, outperformed the separate deep network and nonaugmented version features. Its validation accuracy, classification precision, recall, and F-measure are all 96.6%, with a 93.4% intersection over union and a loss of 0.123. Furthermore, the software developed through the proposed approach holds great promise as a rapid tool for diagnosing grapevine diseases in less than one minute. The framework of the study shows potential for future expansion to include various types of trees. This capability can assist farmers in early detection of tree diseases, enabling them to implement preventive measures.
Why it matches plant phenotyping methodsブドウ葉のRGB画像から病害状態を推定する深層学習手法とデータセット、診断ソフトウェアを開発・評価しており、植物表現型取得が中心である。
abstractThe objective of this research is to develop an intelligent approach, supported by user-friendly, open-source software named AI GrapeCare
Reproduction assets foundThe authors publicly deposited the Python script, the trained hybrid deep network model, real-world grape disease sample images, and the standalone AI GrapeCare software on Google Drive. The PDD grape leaf image dataset (295 images) is also public at pdd.jinr.ru, but that URL is not in the allowed list, so only the verCode · publicface [ 32 ]. To ensure cross-platform compatibility, including Windows, Linux, and Mac OS, PyInstaller [ 33 ] was applied. The Python script, the hybrid deep network that was generated, grape disease samples from real-world conditions, and the stand-alone version of this software are all available for download on Google Drive ( https://drive.google.com/file/d/1uOVAMiFDWBZsm8U9alzSdSk2c-A8zWBN , accessed on 10 December 2023), packaged in a RAR file with a size of 1.03 GB. As depicted in the overarching flowchart ( Figure 7 ), the pseudo-code explains the establishment of the AI GrapeCare software and its associated functions. The software workflow is organized into five primary stages: (1) loOpen asset ↗Google Drivelines:148-246Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Jan 2024Computers and Electronics in Agriculture.
The monitoring of grapes for ripeness estimation is a practice that enables fruit harvesting at the optimal time. Hyperspectral Imaging (HSI) represents a non-destructive and high-throughput alternative to traditional laboratory analyses. Current literature approaches perform hyperspectral measurements using line scan sensors or low-resolution static snapshot cameras, which hinder a fast per-bunch ripeness characterization. We propose a framework for on-the-go collection and processing of proximal snapshot hyperspectral images to estimate single bunch ripeness parameters. Focusing on table grapes (Vitis vinifera L. cv. Red Globe), we collected images under natural illumination with a hyperspectral camera (500-900 nm) mounted on a moving vehicle in an experimental block sited in Piacenza, Italy. We investigated images collected in August and September 2021 representing two ripening stages. The composition of the imaged grape bunches was determined through laboratory chemical analyses to predict Total Soluble Solids (TSS) and anthocyanin concentration. The images were pre-processed via multimodal image registration to correct the unalignment of bands due to the vehicle motion, and the single bunches were automatically identified on false RGB images through a Mask Region-Convolutional Neural Network (Mask R-CNN) instance segmentation network. The mean spectra of the bunches were used as input features of a Partial Least Squares Regression (PLSR) model to predict the chemical parameters at single bunch and whole vine scales. The regression model of TSS had an R2 (10-fold nested cross-validation) of 0.75 and 0.85 on a per-bunch and per-vine basis, respectively. The regression model of anthocyanin had an R2 of 0.68 and 0.49 on a per-bunch and per-vine basis, respectively. The results suggest the potential of using snapshot hyperspectral images for high-throughput analysis of a per-bunch grape ripeness estimation. The method described in this study could give valuable information to improve grape ripening monitoring and management of harvest operations and even allow for precise and automated robotic harvesting.
Why it matches plant phenotyping methods近接ハイパースペクトル画像、画像登録、果房セグメンテーション、回帰モデルを統合し、果房単位の成熟度形質（TSS・アントシアニン）を推定・検証する手法が研究の中心である。
abstractWe propose a framework for on-the-go collection and processing of proximal snapshot hyperspectral images to estimate single bunch ripeness parameters.
Botrytis cinerea is one of the most destructive diseases for Vitis vinifera, and grape bunch morphology plays a crucial role in grey mould infection. However, the common visual evaluation technique for assessing bunch compactness suffers from a lack of sensitivity and objectivity. This study proposes a standardised digital twin shape analysis to evaluate the morphology of grape bunches. Seventeen Pinot Gris and six Pinot Noir clones were considered. The grey mould severity was evaluated in the field. Fully ripened bunches (138) were gathered and then photographed at different angulations. Digital twin reconstruction was carried out using the photogrammetry technique. Several measures and indices were extracted from each digital twin. Principal component analysis and multiple linear regression models were applied to identify the descriptors most related to grey mould symptoms. The results revealed that the most significant factors include the berries density, the estimated empty volume, and the bunch width. These results show that digital twins are a suitable tool for estimating grey mould infection risks. Two linear models, divided into 2D and 3D descriptor models, were proposed. The R-squared value and the root mean square error were compared between the models. For Pinot Gris, from the 2D to the 3D models, the R-squared value rose from 0.656 to 0.838, while the error decreased from 1.713 to 1.175. In Pinot Noir, the 2D model did not provide sufficient robustness, while the 3D model had an R-squared value of 0.936 and an error of 0.290.
Why it matches plant phenotyping methodsブドウ房形態をデジタルツインとフォトグラメトリで再構成し、形態記述子を抽出・検証して灰色かび感染リスクを推定する手法が研究の中心である。
abstractThis study proposes a standardised digital twin shape analysis to evaluate the morphology of grape bunches.
Near-Infrared spectroscopy (NIR) returns full spectra in the region between 750-2500 nm. Although a full spectrum provides extremely informative data, sometimes this enormous amount of detail is redundant and does not bring any additional information. In this work, different attribute selection methods for the development of vineyard water status predictive models are presented. Spectra from grapevine leaves were collected on-the-go (from a moving vehicle) along nine dates during the 2015 season in a commercial vineyard using a NIR spectrometer (1200-2100 nm). Contemporarily, the stem water potential (Ψₛₜₑₘ) was also measured in the monitored vines. A manual selection, based on Variable Importance in Projection scores (VIP scores) to choose the spectrum intervals including the most important wavelengths (interval selection), the locally most important wavelengths in the spectrum (peak selection), as well as the Interval Partial Least Squares (IPLS) were tested as attribute selection methods. The results obtained for the estimation of Ψₛₜₑₘ using the whole spectrum (R²P=0.84, RMSEP=0.167MPa) were comparable to those yielded by the three attribute selection methods: the interval selection method (R²P=0.80, RMSEP=0.186 MPa), the peak selection method (R²P=0.77, RMSEP=0.201MPa) and the IPLS (R²P ∼ 0.62-0.79, RMSEP ∼ 0.186-0.252 MPa). The highest simplification was provided by two IPLS models with three wavelengths and bandwidths of 20 and 4 nm that yielded R²P∼0.78 and RMSEP∼ 0.190 MPa. These results corroborate the suitability of a highly reduced selection of NIR wavelengths for the prediction of grapevine water status, and its utility to develop simpler multispectral devices for vineyard water status estimation.
Why it matches plant phenotyping methodsNIRスペクトルからブドウの水分状態を推定する波長選択法を比較・検証し、簡易マルチスペクトル装置への応用可能性まで示しており、植物表現型取得法が研究の中心である。
abstractdifferent attribute selection methods for the development of vineyard water status predictive models are presented
Various nitrogen (N) prediction approaches were tested using aerial hyperspectral imagery and ground truth data (leaf tissue analysis) collected from a table grape vineyard in Shafter, California, at various phenological stages, namely pre-bloom, bloom, fruit set, and veraison. The best results were achieved by chemometrics, machine learning, and physically based modeling with the coefficient of determination (R²) values ranging between 0.68 and 0.69. A significant finding was the high correlation between the VIS-NIR spectrum and canopy N at bloom, a pattern not replicated in other phenological stages. This suggests that measurement timing may be critical for remote sensing of N, possibly due, in part, to known interactive effects of leaf age on the relationship between Chl and N. The results offer insights into using aerial spectral imagery and Radiative Transfer Modeling (RTM) for more accurate N prediction in grapevines, also suggesting that using the full VIS-NIR spectrum can potentially improve N prediction accuracy by incorporating both chlorophyll influence and canopy structure effects, thereby surpassing the traditional reliance solely on the chlorophyll-nitrogen correlation. Our study highlights the need for a better understanding of the factors that affect the efficacy of remote sensing.
Why it matches plant phenotyping methodsブドウ樹冠の窒素量という植物形質を対象に、航空ハイパースペクトル画像、機械学習、ケモメトリクス、RTMによる推定手法を比較・評価しており、フェノタイピング手法が中心です。
abstractVarious nitrogen (N) prediction approaches were tested using aerial hyperspectral imagery and ground truth data (leaf tissue analysis)
The fungus Botrytis cinerea causes severe diseases in many crops. In grapevines, it causes Botrytis bunch rot (BBR), one of the most reported diseases worldwide. It affects all herbaceous organs of the vine, especially the ripe berries, causing significant reductions in yield and wine quality. Botrytis detection models traditionally focus on temporal analysis at a specific spatial location, ignoring the study of the spatial variability of the crop. Unmanned aerial vehicles (UAVs) equipped with multispectral cameras can provide high-resolution images that can be valuable information to develop a tool for aerial pest detection. This paper proposes an algorithm to assess the risk of Botrytis development in a vineyard in Spain, using as input products generated by UAV imagery: DTM (Digital Terrain Model), NDVI (Normalised Difference Vegetation Index), CHM (Canopy Height Model) and LAI (Leaf Area Index). They represent the height and architecture of the canopy, the topography and the plant status. Healthy vines were significantly different from vines affected by Botrytis (p 0.7) that may support vineyard managers in understanding the spatial variability of the disease, allowing the spatial 2D visualisation of the risk of BBR disease development and, potentially, resulting in higher operational efficiency and reducing phytosanitary treatments, as well as economic costs. Furthermore, the present work takes advantage of imaging technologies that provide information about any location in the field, not only about specific points in the vineyard, suggesting that UAV imagery is appropriate to measure the likelihood of BBR development within the vineyard, highlighting the importance of efficient disease management based on spatial variability.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像からキャノピー形状・植物状態を抽出し、ブドウの灰色かび病リスクを空間推定するアルゴリズムが研究の中心であり、植物の病害状態を測定する実質的なフェノタイピング手法である。
abstractThis paper proposes an algorithm to assess the risk of Botrytis development in a vineyard in Spain, using as input products generated by UAV imagery: DTM (Digital Terrain Model), NDVI (Normalised Difference Vegetation Index), CHM (Canopy Height Model) and LAI (Leaf Area Index).
Work carried out as part of the Casdar RT TechnoDoseViti project (2018-2021) identified the vegetation descriptors needed to predict the statistical distribution of deposits resulting from plant protection products sprayed within the plant canopy (Cheraiet et al., 2021). This experimental modelling work (Codis et al., 2018) has made it possible to build, calibrate and validate multivariate models for predicting the different deciles of the quantity of surface deposits in the plant canopy as a function of vegetation descriptors (height, thickness, and porosity of the canopy) measured using a mobile 2D LiDAR sensor enabling a 3D reconstruction of the vegetation (Cheraiet et al., 2019 & 2020). Models were developed and calibrated for different types of sprayers and different viticultural contexts (wide vines and narrow vines). The data was used to compare different technological scenarios consisting of different mechanisation strategies and different levels of equipment technology, in order to assess the potential contribution of precision application techniques in terms of phytosanitary inputs reduction. As part of the project, a web simulator has been developed: https://technodoseviti.hdigitag.fr
Why it matches plant phenotyping methods移動式2D LiDARによる植物キャノピーの高さ・厚さ・空隙率の測定と3D再構成を用い、散布付着量予測モデルを構築・較正・検証しており、植物形態の取得・推定手法が中心である。
abstractbuild, calibrate and validate multivariate models for predicting the different deciles of the quantity of surface deposits in the plant canopy as a function of vegetation descriptors (height, thickness, and porosity of the canopy) measured using a mobile 2D LiDAR sensor enabling a 3D reconstruction of the vegetation
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 7 Sept 2026
Abstract Reflectance spectroscopy has become a powerful tool for non-destructive and high- throughput phenotyping in crops. Emerging evidence indicates that this technique allows for estimation of multiple leaf traits across large numbers of samples, while alleviating the constraints associated with traditional field- or lab-based approaches. While the ability of reflectance spectroscopy to predict leaf traits across species and ecosystems has received considerable attention, whether or not this technique can be applied to quantify within species trait variation have not been extensively explored. Employing reflectance spectroscopy to quantify intraspecific variation in functional traits is especially appealing in the field of agroecology, where it may present an approach for better understanding crop performance, fitness, and trait-based responses to managed and unmanaged environmental conditions. We tested if reflectance spectroscopy coupled with Partial Least Square Regression (PLSR) predicts rates of photosynthetic carbon assimilation ( A max ), Rubisco carboxylation ( V cmax ), electron transport ( J max ), leaf mass per area (LMA), and leaf nitrogen (N), across six wine grape ( Vitis vinifera ) varieties (Cabernet Franc, Cabernet Sauvignon, Merlot, Pinot Noir, Viognier, Sauvignon Blanc). Our PLSR models showed strong capability in predicting intraspecific trait variation, explaining 55%, 58%, 62%, and 64% of the variation in observed J max , V cmax , leaf N, and LMA values, respectively. However, predictions of A max were less strong, with reflectance spectra explaining only 29% of the variation in this trait. Our results indicate that trait variation within species and crops is less well-predicted by reflectance spectroscopy, than trait variation that exists among species. However, our results indicate that reflectance spectroscopy still presents a viable technique for quantifying trait variation and plant responses to environmental change in agroecosystems.
Why it matches plant phenotyping methods反射分光法とPLSRによる葉の生理・機能形質推定を検証しており、フェノタイピング手法の評価が研究の中心である。
abstractEmploying reflectance spectroscopy to quantify intraspecific variation in functional traits
The Scholander-type pressure chamber to measure midday stem water potential (MSWP) has been widely used to schedule irrigation in commercial vineyards. However, the limited number of sites that can be evaluated using the pressure chamber makes it difficult to evaluate the spatial variability of vineyard water status. As an alternative, several authors have suggested using the crop water stress index (CWSI) based on low-cost thermal infrared (TIR) sensors to estimate the MSWP. Therefore, this study aimed to develop a low-cost wireless infrared sensor network (WISN) to monitor the spatial variability of MSWPs in a drip-irrigated Cabernet Sauvignon vineyard under two levels of water stress. For this study, the MLX90614 sensor was used to measure canopy temperature (Tc), and thus compute the CWSI. The results indicated that good performance of the MLX90614 infrared thermometers was observed under laboratory and vineyard conditions with root mean square error (RMSE) and mean absolute error (MAE) values being less than 1.0 °C. Finally, a good nonlinear correlation between the MSWP and CWSI (R2 = 0.72) was observed, allowing the development of intra-vineyard spatial variability maps of MSWP using the low-cost wireless infrared sensor network.
Why it matches plant phenotyping methods低コスト熱赤外センサーネットワークを開発・検証し、ブドウ樹の水分状態（MSWP）をCWSIから推定する手法が研究の中心であるため。
abstractthis study aimed to develop a low-cost wireless infrared sensor network (WISN) to monitor the spatial variability of MSWPs
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 7 Sept 2026
BACKGROUND: Grapevine berries undergo asynchronous growth and ripening dynamics within the same bunch. Due to the lack of efficient methods to perform sequential non-destructive measurements on a representative number of individual berries, the genetic and environmental origins of this heterogeneity, remain nearly unknown. To address these limitations, we propose a method to track the growth and coloration kinetics of individual berries on time-lapse images of grapevine bunches. RESULTS: First, a deep-learning approach is used to detect berries with at least 50 ± 10% of visible contours, and infer the shape they would have in the absence of occlusions. Second, a tracking algorithm was developed to assign a common label to shapes representing the same berry along the time-series. Training and validation of the methods were performed on challenging image datasets acquired in a robotised high-throughput phenotyping platform. Berries were detected on various genotypes with a F1-score of 91.8%, and segmented with a mean absolute error of 4.1% on their area. Tracking allowed to label and retrieve the temporal identity of more than half of the segmented berries, with an accuracy of 98.1%. This method was used to extract individual growth and colour kinetics of various berries from the same bunch, allowing us to propose the first statistically relevant analysis of berry ripening kinetics, with a time resolution lower than one day. CONCLUSIONS: We successfully developed a fully-automated open-source method to detect, segment and track overlapping berries in time-series of grapevine bunch images acquired in laboratory conditions. This makes it possible to quantify fine aspects of individual berry development, and to characterise the asynchrony within the bunch. The interest of such analysis was illustrated here for one cultivar, but the method has the potential to be applied in a high throughput phenotyping context. This opens the way for revisiting the genetic and environmental variations of the ripening dynamics. Such variations could be considered both from the point of view of fruit development and the phenological structure of the population, which would constitute a paradigm shift.
Why it matches plant phenotyping methodsブドウ果実の画像から個々のベリーを検出・セグメント化・追跡し、成長および色彩という植物形質を自動抽出する手法の開発と検証が中心である。
abstractwe propose a method to track the growth and coloration kinetics of individual berries on time-lapse images of grapevine bunches.
Botrytis cinerea is one of the most destructive diseases for Vitis vinifera, and grape bunch morphology plays a crucial role in grey mould infection. However, the common visual evaluation technique for assessing bunch compactness suffers from a lack of sensitivity and objectivity. This study proposes a standardised digital twin shape analysis to evaluate the morphology of grape bunches. Seventeen Pinot Gris and six Pinot Noir clones were considered. The grey mould severity was evaluated in the field. Fully ripened bunches (138) were gathered and then photographed at different angles. Digital twin reconstruction was carried out using the photogrammetry technique. Several measures and indices were extracted from each digital twin. Principal component analysis and multiple linear regression models were applied to identify the descriptors most related to grey mould symptoms. The results revealed that the most significant factors include the berries density, the estimated empty volume, and the bunch width. These results show that digital twins are a suitable tool for estimating grey mould infection risks. Two linear models, divided into 2D and 3D descriptor models, were proposed. The R-squared value and the root mean square error were compared between the models. For Pinot Gris, from the 2D to the 3D models, the R-squared value rose from 0.656 to 0.838, while the error decreased from 1.713 to 1.175. In Pinot Noir, the 2D model did not provide sufficient robustness, while the 3D model had an R-squared value of 0.936 and an error of 0.290.
Why it matches plant phenotyping methodsブドウ果房の形態をフォトグラメトリとデジタルツインで再構築し、形態形質を抽出して灰色かび症状との関連を検証する手法が研究の中心である。
abstractThis study proposes a standardised digital twin shape analysis to evaluate the morphology of grape bunches.
This paper introduces a tomography-like method for assessing grape maturation. It analyses inner tissue spectra through point-of-measurement (POM) sensing. A multi-block hierarchical principal component analysis (MHPCA) algorithm was used for the spectral reconstruction of total grapes (skin, pulp, and seed). Two grape cultivars, Loureiro (white; n = 216) and Vinhão (red; n = 205) were measured at 12 dates after veraison (DAV). The reconstructed spectra showed no significant differences (p < 0.001) from the originals for both grapes. Loureiro had better statistical metrics (Person's correlation coefficient (r) values for: total grape: 0.99, skin: 1; pulp: 1, seed: 0.94) than Vinhão (r values for: total grape: 0.92, skin: 0.92; pulp: 0.95, seed: 0.95). Using self-learning artificial intelligence (SL-AI), the following parameters were predicted for both grapes: soluble solids content (%; MAPE <13%), puncture force (N; MAPE <29%), chlorophyll content (a.u.; MAPE <29%), and anthocyanin content (a.u.; MAPE <17%, Vinhão only). When comparing observed values with predicted skin, pulp, and seed spectra, Vinhão showed no statistical differences for most parameters, except pulp chlorophyll on one DAV in the final maturation stage. The same was done with the Loureiro cultivar. Although Loureiro mostly showed no statistical differences in assessed parameters across tissues and dates, variations were found in pulp and skin chlorophyll content and puncture force. This tomography-like approach based on tissue maturation can help viticulturists to access instant data on grape maturation, supporting informed decision-making and promoting more sustainable agricultural practices.
Why it matches plant phenotyping methodsブドウ組織の成熟に関するスペクトルを再構成し、糖度・硬度・クロロフィル・アントシアニンなどの植物器官形質を予測する新規センシング／計算手法を開発・検証しており、フェノタイピング手法が中心である。
abstractThis paper introduces a tomography-like method for assessing grape maturation.
Published23 Nov 20232023 International Conference on Advances in Computation, Communication and Information Technology (ICAICCIT)Cited by 4 · OpenAlex ↗
India, being an agriculture-centric nation, historically relied on traditional farming methods that often resulted in crop losses and significant financial setbacks for farmers. In the present era, however, the incorporation of technology in agriculture has led to a rise in crop value, still, there is a great scope of research. One of the key reasons of the production of low-quality crops is the presence of infections or diseases. This research paper focuses on the application of machine learning algorithms, specifically Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and MobileNetV2, for plant disease detection in apple, strawberry, and grape leaves. The dataset used comprises 29,327 images, and various performance metrics were employed to evaluate the models' accuracies. According to the findings, the SVM model performs best with 82%, 98% and 72% accuracy for apple, strawberry, and grape crop respectively. Also, the precision, recall and F-score values are significantly high for CNN. So, among three considered model in the experiment for identification of disease in apple while for grapes and strawberry CNN works better than other two models.
Why it matches plant phenotyping methods葉画像から植物病害を推定する機械学習手法を比較・評価しており、植物の病徴状態の取得・判定が研究の中心です。
abstractThis research paper focuses on the application of machine learning algorithms, specifically Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and MobileNetV2, for plant disease detection in apple, strawberry, and grape leaves.
The strong societal demand to reduce pesticide use and adaptation to climate change challenges the capacities of phenotyping new varieties in the vineyard. High-throughput phenotyping is a way to obtain meaningful and reliable information on hundreds of genotypes in a limited period. We evaluated traits related to growth in 209 genotypes from an interspecific grapevine biparental cross, between IJ119, a local genitor, and Divona, both in summer and in winter, using several methods: fresh pruning wood weight, exposed leaf area calculated from digital images, leaf chlorophyll concentration, and LiDAR-derived apparent volumes. Using high-density genetic information obtained by the genotyping by sequencing technology (GBS), we detected 6 regions of the grapevine genome [quantitative trait loci (QTL)] associated with the variations of the traits in the progeny. The detection of statistically significant QTLs, as well as correlations ( R 2 ) with traditional methods above 0.46, shows that LiDAR technology is effective in characterizing the growth features of the grapevine. Heritabilities calculated with LiDAR-derived total canopy and pruning wood volumes were high, above 0.66, and stable between growing seasons. These variables provided genetic models explaining up to 47% of the phenotypic variance, which were better than models obtained with the exposed leaf area estimated from images and the destructive pruning weight measurements. Our results highlight the relevance of LiDAR-derived traits for characterizing genetically induced differences in grapevine growth and open new perspectives for high-throughput phenotyping of grapevines in the vineyard.
Why it matches plant phenotyping methodsLiDARによるブドウ樹冠・剪定木体積の取得を、従来法との相関、遺伝率、QTL解析で評価しており、植物形質の高スループット計測法が研究の中心です。
abstractThe detection of statistically significant QTLs, as well as correlations ( R 2 ) with traditional methods above 0.46, shows that LiDAR technology is effective in characterizing the growth features of the grapevine.
Reproduction assets foundThe paper's Data availability statement provides a public repository deposit (DOI 10.57745/PETTGY) for the study data and an authors' public ImageJ script for estimating foliage coverage used in the RGB-image phenotyping analysis.Dataset · publictyping but also his expertise and helped with the manuscript review. D.M. supervised the program and helped with manuscript writing. É.D. supervised the whole study and wrote the first draft of the manuscript.
Competing interests: The authors declare that they have no competing interests.
Data availability
Data are available at https://doi.org/10.57745/PETTGY . ImageJ script for estimating foliage coverage: https://forgemia.inra.fr/eric.duchene/image-analysis-scripts/-/blob/main/FoliageCoverage_PC_EN.txt
Supplementary Materials
Supplementary 1
Fig. S1
Tables S1 to S5
Click here for additional data file.
References
1.
Carvalho
LC , Goncalves
EF ,
da
Silva
JM , Costa
JM
.
Potential phOpen asset ↗10.57745/PETTGY · 10.57745/PETTGYlines:825-1015Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 7 Sept 2026
Accurately characterizing vineyard parameters is crucial for precise vineyard management and breeding purposes. Various macroscopic vineyard parameters are required to make informed management decisions, such as pesticide application, defoliation strategies, and determining optimal sugar content in each berry by assessing biomass. In this paper, we present a novel approach that utilizes point cloud data to detect trunk positions and extract macroscopic vineyard characteristics, including plant height, canopy width, and canopy volume. Our approach relies solely on geometric features and is compatible with different training systems and data collected using various 3D sensors. To evaluate the effectiveness and robustness of our proposed approach, we conducted extensive experiments on multiple grapevine rows trained in two different systems. Our method provides more comprehensive canopy characteristics than traditional manual measurements, which are not representative throughout the row. The experimental results demonstrate the accuracy and efficiency of our method in extracting vital macroscopic vineyard characteristics, providing valuable insights for yield monitoring, grape quality optimization, and strategic interventions to enhance vineyard productivity and sustainability.
Why it matches plant phenotyping methodsブドウ個体を検出し、点群から樹高・樹冠幅・樹冠体積を抽出する手法を開発・評価しており、植物形質の取得が研究の中心である。
abstractwe present a novel approach that utilizes point cloud data to detect trunk positions and extract macroscopic vineyard characteristics, including plant height, canopy width, and canopy volume.
Grapevine downy mildew (GDM), caused by the oomycete Plasmopara viticola, can cause 100% yield loss and vine death under conducive conditions. Growers currently rely on frequent fungicide applications for control, but this practice has led to widespread resistance. Rapid remote detection and surveillance of GDM outbreaks would enable precision pesticide applications to target effective but resistance-prone fungicides where and when most needed, while relying on less resistance-prone protectants elsewhere. High resolution commercial satellite platforms offer the opportunity to track rapidly spreading diseases like GDM over large, heterogeneous fields. Here, we investigate the capacity of PlanetScope (3 m) and SkySat (50 cm) imagery for season-long GDM detection and surveillance. A team of trained scouts rated GDM severity and incidence in two acres of Chardonnay grapevines in Geneva, NY, USA in June-August of 2020, 2021, and 2022. Satellite imagery acquired within 72 hours of scouting was processed to extract single-band reflectance and vegetation indices (VIs). Random forest models trained on spectral bands and VIs derived from both image datasets could classify areas of high and low GDM incidence and severity with maximum accuracies of 0.88 (SkySat) and 0.94 (PlanetScope). However, we do not observe significant differences between VIs of high and low damage classes until late July-early August. We identify cloud cover, image co-registration, and low spectral resolution as key challenges to operationalizing satellite-based GDM surveillance. This work establishes the capacity of spaceborne multispectral sensors to detect late-stage GDM and outlines steps towards incorporating satellite remote sensing in grapevine disease surveillance systems.
Why it matches plant phenotyping methods衛星画像と機械学習を用いてブドウのべと病の発生・重症度を直接推定し、精度評価と運用上の課題を検証しているため、植物病害フェノタイピング手法が中心である。
abstractHere, we investigate the capacity of PlanetScope (3 m) and SkySat (50 cm) imagery for season-long GDM detection and surveillance.
India ranks among the top ten nations in the world for grape production. Fungal pathogens inflict damage to crop plants in turn making cultivators bear huge economical losses. With an output of 1.21 million tons (about 2% of 57.40 million tons produced globally). 1.2% of the nation’s total fruit cropland is covered by grapes. But due to fungal diseases the effect of the yield produced ranges from 5-80% depending on the severity of diseases which will affect the yield of grape vineyard. In precision agriculture, new sensing technologies and artificial intelligence could be used to automatically identify grapevine and disease pest symptoms. Traditional manual disease-monitoring methods are inefficient, labor-intensive, and ineffective. Timely effective and precise evaluation of grape diseases is admitted as a critical step in the field management. In this paper, we are explaining about different optical sensing methods applied for RGB, Multispectral and Thermal cameras. Section-wise we will be describing environmental set up for image-aquation, data-preprocessing, different modelling methods, evaluation matrix, result, and reviewer’s comment.
Why it matches plant phenotyping methods植物の真菌病徴をRGB・マルチスペクトル・熱画像で検出する方法を体系的に扱うレビューであり、植物状態の取得・推定手法が中心です。
titleA Review on Plant Fungal Disease Detection based on RGB, Multispectral and Thermal Camera
Introduction The challenges associated with data availability, class imbalance, and the need for data augmentation are well-recognized in the field of plant disease detection. The collection of large-scale datasets for plant diseases is particularly demanding due to seasonal and geographical constraints, leading to significant cost and time investments. Traditional data augmentation techniques, such as cropping, resizing, and rotation, have been largely supplanted by more advanced methods. In particular, the utilization of Generative Adversarial Networks (GANs) for the creation of realistic synthetic images has become a focal point of contemporary research, addressing issues related to data scarcity and class imbalance in the training of deep learning models. Recently, the emergence of diffusion models has captivated the scientific community, offering superior and realistic output compared to GANs. Despite these advancements, the application of diffusion models in the domain of plant science remains an unexplored frontier, presenting an opportunity for groundbreaking contributions. Methods In this study, we delve into the principles of diffusion technology, contrasting its methodology and performance with state-of-the-art GAN solutions, specifically examining the guided inference model of GANs, named InstaGAN, and a diffusion-based model, RePaint. Both models utilize segmentation masks to guide the generation process, albeit with distinct principles. For a fair comparison, a subset of the PlantVillage dataset is used, containing two disease classes of tomato leaves and three disease classes of grape leaf diseases, as results on these classes have been published in other publications. Results Quantitatively, RePaint demonstrated superior performance over InstaGAN, with average Fréchet Inception Distance (FID) score of 138.28 and Kernel Inception Distance (KID) score of 0.089 ± (0.002), compared to InstaGAN’s average FID and KID scores of 206.02 and 0.159 ± (0.004) respectively. Additionally, RePaint’s FID scores for grape leaf diseases were 69.05, outperforming other published methods such as DCGAN (309.376), LeafGAN (178.256), and InstaGAN (114.28). For tomato leaf diseases, RePaint achieved an FID score of 161.35, surpassing other methods like WGAN (226.08), SAGAN (229.7233), and InstaGAN (236.61). Discussion This study offers valuable insights into the potential of diffusion models for data augmentation in plant disease detection, paving the way for future research in this promising field.
Why it matches plant phenotyping methods植物病害画像を対象に拡散モデルとGANを比較し、病害画像データ拡張の性能をFID・KIDで検証する研究であり、植物病害状態の画像ベース評価を支える方法が中心である。
titleHarnessing the power of diffusion models for plant disease image augmentation
Reproduction assets foundThe paper's experiments use a subset of the public PlantVillage image dataset, which the authors explicitly link in the data availability statement. No author analysis code, trained models, or generated-image deposits are mentioned.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://www.kaggle.com/datasets/mohitsingh1804/plantvillage .Open asset ↗Kaggle · mohitsingh1804/plantvillagelines:841-873Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Nov 2023Computers and Electronics in Agriculture.
In order to realize the acquisition of plant leaf images by agricultural drones and perform regional segmentation and disease recognition on leaf cluster images over a wide area, this paper takes grape leaves as an example and designs a set of processing procedures combining the improved U-net and VGG-19 networks. This paper aims to address the difficulty of obtaining target leaves for recognition in complex environments due to the presence of multiple invalid leaves. First, the feature extraction process is performed on the training and validation datasets to reduce the input parameters of the network and increase the training speed. Second, the test data is divided into three datasets: low-light, jitter interference, and two-factor combination. After recovery processing, the Multi-fusion U-net network is fed to locate diseased leaves. Finally, the improved VGG-19 network was used again to locate and identify the disease. Experimental results show that the proposed procedure achieves satisfactory performance in UAV image processing. The average accuracy of segmentation reaches 71.91%, and the identification rate of disease location is increased by 12.33% after segmentation, which provides a strong practical basis for the implementation of unmanned smart ecological farms.
Why it matches plant phenotyping methodsUAV画像からブドウ葉の病変状態を分割・定位・認識するCNNベースの画像解析手法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。
abstractdesigns a set of processing procedures combining the improved U-net and VGG-19 networks
Anthocyanin composition is responsible for the red colour of grape berries and wines, and contributes to their organoleptic quality. However, anthocyanin biosynthesis is under genetic, developmental and environmental regulation, making its targeted fine-tuning challenging. We constructed a mechanistic model to simulate the dynamics of anthocyanin composition throughout grape ripening in Vitis vinifera, employing a consensus anthocyanin biosynthesis pathway. The model was calibrated and validated using six datasets from eight cultivars and 37 growth conditions. Tuning the transformation and degradation parameters allowed us to accurately simulate the accumulation process of each individual anthocyanin under different environmental conditions. The model parameters were robust across environments for each genotype. The coefficients of determination (R2) for the simulated versus observed values for the six datasets ranged from 0.92 to 0.99, while the relative root mean square errors (RRMSEs) were between 16.8 and 42.1 %. The leave-one-out cross-validation for three datasets showed R2 values of 0.99, 0.96 and 0.91, and RRMSE values of 28.8, 32.9 and 26.4 %, respectively, suggesting a high prediction quality of the model. Model analysis showed that the anthocyanin profiles of diverse genotypes are relatively stable in response to parameter perturbations. Virtual experiments further suggested that targeted anthocyanin profiles may be reached by manipulating a minimum of three parameters, in a genotype-dependent manner. This model presents a promising methodology for characterizing the temporal progression of anthocyanin composition, while also offering a logical foundation for bioengineering endeavours focused on precisely adjusting the anthocyanin composition of grapes.
Why it matches plant phenotyping methodsブドウ果実のアントシアニン組成という植物器官形質を対象に、動態モデルを構築し、多数のデータセット・品種・環境で較正、検証、交差検証している。組成推定モデルの開発と技術評価が中心であり、単なる生物学的測定ではない。
abstractWe constructed a mechanistic model to simulate the dynamics of anthocyanin composition throughout grape ripening in Vitis vinifera
Unmanned Aerial Vehicle (UAV) thermal imagery is rapidly becoming an essential tool in precision agriculture. Its ability to enable widespread crop status assessment is increasingly critical, given escalating water demands and limited resources, which drive the need for optimizing water use and crop yield through well-planned irrigation and vegetation management. Despite advancements in crop assessment methodologies, including the use of vegetation indices, 2D mapping, and 3D point cloud technologies, some aspects remain less understood. For instance, mission plans often capture nadir and oblique images simultaneously, which can be time- and resource-intensive, without a clear understanding of each image type's impact. This issue is particularly critical for crops with specific growth patterns, such as woody crops, which grow vertically. This research aims to investigate the role of nadir and oblique images in the generation of CWSI (Crop Water Stress Index) maps and CWSI point clouds, that is 2D and 3D products, in woody crops for precision agriculture. To this end, products were generated using Agisoft Metashape, ArcGIS Pro, and CloudCompare to explore the effects of various flight configurations on the final outcome, seeking to identify the most efficient workflow for each remote sensing product. A linear regression analysis reveals that, for generating 2D products (orthomosaics), combining flight angles is redundant, while 3D products (point clouds) are generated equally from nadir and oblique images. Volume calculations show that combining nadir and oblique flights yields the most accurate results for CWSI point clouds compared to LiDAR in terms of geometric representation (R 2 = 0.72), followed by the nadir flight (R 2 = 0.68), and, finally, the oblique flight (R 2 = 0.54). Thus, point clouds offer a fuller perspective of the canopy. To our knowledge, this is the first time that CWSI point clouds have been used for precision viticulture, and this knowledge can aid farm managers, technicians, or UAV pilots in optimizing the capture of UAV image datasets in line with their specific goals.
Why it matches plant phenotyping methodsUAV熱画像からブドウ樹冠の水ストレス指標を2D・3Dで抽出し、飛行条件とLiDARを比較検証する方法論的研究であり、植物表現型取得が中心。
abstractThis research aims to investigate the role of nadir and oblique images in the generation of CWSI (Crop Water Stress Index) maps and CWSI point clouds, that is 2D and 3D products, in woody crops for precision agriculture.
In this paper, we propose a novel approach for analyzing the effects of water regime on grapevine canopy status using robotics as an aid for monitoring and mapping. Data from an unmanned aerial vehicle (UAV) and a ground mobile robot are used to obtain multispectral images and multiple vegetation indexes, and the 3D reconstruction of the canopy, respectively. Unlike previous works, sixty vegetation indexes are computed precisely by using the projected area of the vineyard point cloud as a mask. Extensive experimental tests on repeated plots of Pinot gris vines show that the GDVI, PVI, and TGI vegetation indexes are positively correlated with the water potential: GDVI (R2=0.90 and 0.57 for the stem and pre-dawn water potential, respectively), PVI (R2=0.90 and 0.57), TGI (R2=0.87 and 0.77). Furthermore, the canopy volume and the canopy area projected on the ground are impacted by the water status, as well as stem and pre-dawn water potential measurements. The results obtained in this work demonstrate the feasibility of the proposed approach and the potential of robotic technologies, supporting precision viticulture.
Why it matches plant phenotyping methodsUAVと移動ロボットによる画像・3D再構成を用いて、ブドウ樹冠の水分状態関連形質を抽出・検証する方法が研究の中心である。
abstractData from an unmanned aerial vehicle (UAV) and a ground mobile robot are used to obtain multispectral images and multiple vegetation indexes, and the 3D reconstruction of the canopy, respectively.
GrapevineField / plotFruitObject detectionGrowth / development / phenology
In the viticulture sector, robots are being employed more frequently to increase productivity and accuracy in operations such as vineyard mapping, pruning, and harvesting, especially in locations where human labor is in short supply or expensive. This paper presents the development of an algorithm for grape maturity estimation in the framework of vineyard management. An object detection algorithm is proposed based on You Only Look Once (YOLO) v7 and its extensions in order to detect grape maturity in a white variety of grape (Assyrtiko grape variety). The proposed algorithm was trained using images received over a period of six weeks from grapevines in Drama, Greece. Tests on high-quality images have demonstrated that the detection of five grape maturity stages is possible. Furthermore, the proposed approach has been compared against alternative object detection algorithms. The results showed that YOLO v7 outperforms other architectures both in precision and accuracy. This work paves the way for the development of an autonomous robot for grapevine management.
Why it matches plant phenotyping methodsブドウ果実の成熟段階という植物状態を画像から推定する物体検出法を開発し、複数アルゴリズムと比較検証しており、表現型取得手法が中心である。
abstractThis paper presents the development of an algorithm for grape maturity estimation
Detailed and precise knowledge of production parameters (yield, quality, health status, etc.) in agriculture is the basis for analyzing the effect of any agricultural practice. Fine mapping of production parameters makes it possible to identify the origin of observed variability, whether associated with environmental factors or with agricultural practices. In viticulture, in real commercial context, these data are rare because monitoring systems embedded on harvesting machines for grape yield and quality are not yet available. As a result, they are costly and/or cumbersome to acquire manually. As an alternative, a research project has been proposed to test low-cost methods using GNSS tracking devices for yield and harvest quality mapping in viticulture. The data set was acquired as part of this research. The methodology was applied on a commercial vineyard of 30 ha during the whole 2022 harvest season. The method has identified harvest sectors (HS) associated to measured production parameters (grape mass and harvest quality parameters: sugar content, total acidity, pH, yeast assimilable nitrogen, organic nitrogen) and calculated production parameters (potential alcohol of grapes, yield, yield per plant, percentage of unproductive plants) over the entire vineyard. The grape mass was measured at the vineyard cellar or at the wine-growing cooperative by calibrated scales. The harvest quality parameters were measured from samples on grape must at a commercial laboratory specialized in oenological analysis (Institut Coopératif du Vin, Montpellier, France) with standardized protocols. The percentage of unproductive plants of a harvest sector was calculated from the manually geolocation of each unproductive plants (dead plants + missing plants) over the entire vineyard, the plantation density of blocks, and the geolocalization of the harvest sector. The mean area of these harvest sectors is 0.3 ha. The data set is supplemented by climatic data from a weather station deployed in the center of the vineyard. It provided three climatic parameters (relative humidity, rainfall, air temperature) every 15 min, for the 2020, 2021 and 2022 years. It was also supplemented by a complete description of the vineyard blocks (grape variety, plantation year, area, inter-row distance and vine distance). The proposed data set constitutes a unique and interesting resource for research in agronomy, vine ecophysiology and remote sensing. It can be used for any research in vine ecophysiology aimed at identifying potential relationships between yield and harvest quality parameters for different grape varieties. The data set only covers one year, which is a limitation for studying inter-annual variability of the parameters measured. Another limitation of the method concerns the footprint (0.3 ha on average) of the parameters measured.
Why it matches plant phenotyping methodsGNSSを用いた低コストのブドウ収量・収穫品質マッピング手法と、その大規模データセットが研究の中心であり、収量や不生産株割合などの植物・圃場形質を抽出している。
abstracta research project has been proposed to test low-cost methods using GNSS tracking devices for yield and harvest quality mapping in viticulture.
Reproduction assets foundThe article is a Data in Brief describing the authors' own public Zenodo deposit containing the vineyard phenotyping measurements (block, agronomic/harvest-sector, and weather data as .shp and .csv files), plus an example analysis script (Yield_vs_variety.py) and yield map, all hosted at the stated Zenodo DOI.Dataset · publicce), as well as the geolocation of unproductive wines were obtained from the vineyard Farm Management Information System.
Data source location
Institution: Institut Agro Montpellier
City: Montpellier
Country: France
Data accessibility
Repository name: Zenodo
Data identification number: 10.5281/zenodo.8328384
Direct URL to data: https://doi.org/10.5281/zenodo.8328384
Related research article
J-P. Gras, S. Moinard, T. Crestey and B. Tisseyre. Mapping grape yield with low-cost vehicle tracking devices, In Precision agriculture’23 , Wageningen Academic Publishers. (2023) 555-561. https://doi.org/10.3920/978-90-8686-947-3_70
1.
Value of the Data
The dataset presented in this paper is a particulOpen asset ↗Zenodo · 10.5281/zenodo.8328384lines:32-61Plant phenotyping relevance match · UnverifiedEurope PMC · Crossref · checked 7 Sept 2026
Abstract An innovative approach for automated plant disease identification has been proposed in this study. The main contribution of this study is the introduction of a bipartite graph-based clustering technique that has been used for image segmentation, a feature extraction methodology using Self Organizing Map (SOM), and a ray tracing method. Numerous research works have already been done in this area with their respective merits and demerits. But this bipartite graph-based clustering for image segmentation, feature extraction using SOM and ray tracing technique has not been used in any of these studies as far as we are aware. The core idea behind this clustering technique is to represent similar spatial data points using a bipartite graph and then Singular Value Decomposition has been used on that graph for clustering. It is common to use SOM for clustering. However, in this study, SOM has been used for feature extraction. First, a spatial dependency matrix based on the pixel value of the gray image has been constructed using SOM. Then some statistical features have been computed from this matrix. Using the ray tracing method, the length of the most extended cluster i.e. the length of the most extended disease-affected patch, and the distribution of the clusters in the image have been computed. The accuracy of our model has been greatly enhanced using these features only. It has been experimented on disease-affected Grape leaf images taken from the Plant Village Dataset. This model outperforms state-of-the-art models which is shown in the result section. Not only that, the proposed features produce better accuracy rather than using some existing features. This comparison has also been shown in the result section. The result has been validated using K-fold cross-validation. Last, but not the least, these features produce good accuracy using different classifiers also.
Why it matches plant phenotyping methods植物病害画像から病斑の長さ・分布を抽出する画像解析手法を開発し、交差検証と既存特徴量との比較で評価しており、病害表現型の取得が中心です。
abstractThe main contribution of this study is the introduction of a bipartite graph-based clustering technique that has been used for image segmentation, a feature extraction methodology using Self Organizing Map (SOM), and a ray tracing method.
Plant disease classification is the use of machine learning techniques for determining the type of disease from the input leaf images of the plants based on certain features. It is an important research area since early identification and treatment of plant disease is critical for saving crops, preventing agricultural disasters, and improving productivity in agriculture. This study proposes a new convolutional neural network model that accurately classifies the diseases on the plant leaves for the agriculture sectors. It especially works on the classification of plant diseases for grape leaves from images by designing a deep-learning architecture. A web application was also implemented to help the agricultural workers. The experiments carried out on real-world images showed that a significant improvement (8.7%) on average was achieved by the proposed model (98.53%) against the state-of-the-art models (89.84%) in terms of accuracy.
Why it matches plant phenotyping methodsブドウ葉画像から病害状態を分類するCNNモデルの開発が中心であり、植物病害表現型の画像ベース推定に該当する。
abstractThis study proposes a new convolutional neural network model that accurately classifies the diseases on the plant leaves for the agriculture sectors.
Plant health plays an important role in influencing agricultural yields and poor plant health can lead to significant economic losses. Grapes are an important and widely cultivated plant, especially in the southern regions of Russia. Grapes are subject to a number of diseases that require timely diagnosis and treatment. Incorrect identification of diseases can lead to large crop losses. A neural network deep learning dataset of 4845 grape disease images was created. Eight categories of common grape diseases typical of the Black Sea region were studied: Mildew, Oidium, Anthracnose, Esca, Gray rot, Black rot, White rot, and bacterial cancer of grapes. In addition, a set of healthy plants was included. In this paper, a new selective search algorithm for monitoring the state of plant development based on computer vision in viticulture, based on YOLOv5, was considered. The most difficult part of object detection is object localization. As a result, the fast and accurate detection of grape health status was realized. The test results showed that the accuracy was 97.5%, with a model size of 14.85 MB. An analysis of existing publications and patents found using the search “Computer vision in viticulture” showed that this technology is original and promising. The developed software package implements the best approaches to the control system in viticulture using computer vision technologies. A mobile application was developed for practical use by the farmer. The developed software and hardware complex can be installed in any vehicle. Such a mobile system will allow for real-time monitoring of the state of the vineyards and will display it on a map. The novelty of this study lies in the integration of software and hardware. Decision support system software can be adapted to solve other similar problems. The software product commercialization plan is focused on the automation and robotization of agriculture, and will form the basis for adding the next set of similar software.
Why it matches plant phenotyping methodsブドウの病害・健全状態を画像から推定する深層学習手法、データセット、ソフトウェア／ハードウェア統合システムが研究の中心であり、植物状態の取得・抽出方法を実質的に開発・評価している。
abstractA neural network deep learning dataset of 4845 grape disease images was created.
Characterizing crop canopies is especially important in the management of woody crops. In this article, two systems were compared to characterise a 50 m long vineyard row section. One of the systems was a mobile terrestrial laser scanner based on a light detection and ranging (LiDAR) sensor (MTLS-LiDAR). The other was an uncrewed aerial vehicle (UAV) based system using digital aerial photogrammetry (UAV-DAP). The resulting 3D point clouds were assessed qualitatively and quantitatively. Canopy heights, widths and volumes were obtained in 0.1 m long sections along the studied row. All the parameters derived from the two systems presented statistically significant differences. The coefficients of determination between systems were 0.619 for canopy maximum heights above ground level (agl), 0.686 for 90th percentile (P90) heights agl, and 0.283 and 0.274 for maximum and P90 vegetated heights, respectively. Coefficients of determination between averaged maximum canopy width and P90 canopy width were 0.328 and 0.317, respectively. Coefficients of determination between cross-sectional areas determined from maximum widths, P90 widths and from the occupancy grid method were 0.423, 0.409 and 0.334, respectively. Total canopy volume for the entire row obtained from the three cross section estimation methods differed between 19 m3 and 25 m3. The reasons found were that the MTLS-LiDAR-derived point cloud captured the canopy top and side variability but could be affected by occlusions, mixed pixels and tall grass-like weeds present in the surveyed area. For its part, the UAV-DAP-derived point cloud tended to miss top and side shoots and somewhat smoothed canopy variability. As neither of the systems is optimal, a balance needs to be found according to the specific requirements of the survey. For this purpose, a list of pros and cons is presented to support the selection of one of the two systems for canopy monitoring. The MTLS-LiDAR system should be chosen when high detail is required but small areas are to be scanned. Alternatively, the UAV-DAP system should be chosen when large areas are to be monitored and when canopy detail is not so important. Further results are presented in Part 2 for a larger area and including pear and peach orchards with different training systems. Future research is to be conducted on how the compared systems affect variability detection and support variable-rate prescriptions. .
Why it matches plant phenotyping methodsLiDARとUAV写真測量を用いてブドウ樹冠の高さ・幅・体積を抽出し、両手法を定量比較・評価しており、植物形質取得法が研究の中心である。
abstracttwo systems were compared to characterise a 50 m long vineyard row section
The Internet of Things (IoT) has gained significance in agriculture, using remote sensing and machine learning to help farmers make high-precision management decisions. This technology can be applied in viticulture, making it possible to monitor disease occurrence and prevent them automatically. The study aims to achieve an intelligent grapevine disease detection method, using an IoT sensor network that collects environmental and plant-related data. The focus of this study is the identification of the main parameters which provide early information regarding the grapevine's health. An overview of the sensor network, architecture, and components is provided in this paper. The IoT sensors system is deployed in the experimental plots located within the plantations of the Research Station for Viticulture and Enology (SDV) in Murfatlar, Romania. Classical methods for disease identification are applied in the field as well, in order to compare them with the sensor data, thus improving the algorithm for grapevine disease identification. The data from the sensors are analyzed using Machine Learning (ML) algorithms and correlated with the results obtained using classical methods in order to identify and predict grapevine diseases. The results of the disease occurrence are presented along with the corresponding environmental parameters. The error of the classification system, which uses a feedforward neural network, is 0.05. This study will be continued with the results obtained from the IoT sensors tested in vineyards located in other regions.
Why it matches plant phenotyping methodsIoTセンサーネットワークと機械学習によるブドウ樹の疾病状態推定が研究の中心であり、古典的な疾病同定との比較・アルゴリズム改善も行っているため、植物フェノタイピング手法として採用する。
abstractThe study aims to achieve an intelligent grapevine disease detection method, using an IoT sensor network that collects environmental and plant-related data.
O_LIAccurate and real-time monitoring of grapevine freezing tolerance is crucial for the sustainability of the grape industry in cool climate viticultural regions. However, on-site data is limited. Current prediction models underperform under diverse climate conditions, which limits the large-scale deployment of these methods. C_LIO_LIWe combined grapevine freezing tolerance data from multiple regions in North America and generated a predictive model based on hourly temperature-derived features and cultivar features using AutoGluon, an automatic machine learning engine. Feature importance was quantified by AutoGluon and SHAP value. The final model was evaluated and compared with previous models for its performance under different climate conditions. C_LIO_LIThe final model achieved an overall 1.36 {degrees}C root-mean-square error during model testing and outperformed two previous models using three test cultivars at all testing regions. Two feature importance quantification methods identified five shared essential features. Detailed analysis of the features indicates that the model might have adequately extracted some biological mechanisms during training. C_LIO_LIThe final model, named NYUS.2, was deployed along with two previous models as an R shiny-based application in the 2022-2023 dormancy season, enabling large-scale and real-time simulation of grapevine freezing tolerance in North America for the first time. C_LI
Why it matches plant phenotyping methodsブドウの凍結耐性という植物状態を大規模・リアルタイムに推定する自動機械学習モデルを開発し、既存モデルとの性能比較と実運用展開まで行っており、表現型推定手法が中心です。
abstractWe combined grapevine freezing tolerance data from multiple regions in North America and generated a predictive model based on hourly temperature-derived features and cultivar features using AutoGluon, an automatic machine learning engine.
Reproduction assets foundThe authors publicly released the original LT50 training data and the source code for feature extraction, model training, and deployment in a GitHub repository explicitly stated in the Data availability section. The ACIS URL is a generic external climate data service, not a paper-specific asset.Code · publiclly yielding with a more generalizable model to help understand the biology of grapevine
526
freezing tolerance and quantify the threat of freezing under climate change.
527
5. Data availability
528
All the original training data and source code for feature extraction, modeling training and model
529
deployment are available at https://github.com/imbaterry11/NYUS.2
530
6. Acknowledgements
531
The authors would like to thank Lynn Mills (WA), Beth Ann Workmaster (WI), Katherine
532
Benedict (NS), Alexander Campbell and Jessee Tinslay (QC), Don Smith and Meredith Persico
533
(PA) and Hanna Martins, Felex Pike, and Bill Wilsey (NY) for their help in LT50 data collection.
534
This work was parOpen asset ↗https://github.com/imbaterry11/NYUS.2 · NYUS.2pdf-raw-page:25 lines:1-64Plant phenotyping relevance match · UnverifiedEurope PMC · checked 7 Sept 2026
Irrigation has a strong impact in terms of yield regulation and grape and wine quality, so the implementation of precision watering systems would facilitate the decision-making process about the water use efficiency and the irrigation scheduling in viticulture. The objectives of this work were two-fold. On one hand, to compare and assess grapevine water status using two different spectral devices assembled in a mobile platform and to evaluate their capability to map the spatial variability of the plant water status in two commercial vineyards from July to early October in season 2021, and secondly to develop an algorithm capable of automate the spectral acquisition process using one of the two spectral sensors previously tested. Contemporarily to the spectral measurements collected from the ground vehicle at solar noon, stem water potential (Ψ s ) was used as the reference method to evaluate the grapevine water status. Calibration and prediction models for grapevine water status assessment were performed using the Partial least squares (PLS) regression and the Variable Importance in the Projection (VIP) method. The best regression models returned a determination coefficient for cross validation (R 2 cv ) and external validation (R 2 p ) of 0.70 and 0.75 respectively, and the standard error of cross validation (RMSECV) values were lower than 0.105 MPa and 0.128 MPa for Tempranillo and Graciano varieties using a more expensive and heavier near-infrared (NIR) spectrometer (spectral range 1200-2100 nm). Remarkable models were also built with the miniaturized, low-cost spectral sensor (operating between 900-1860 nm) ranging from 0.69 to 0.71 for R 2 cv , around 0.74 in both varieties for R 2 p and the RMSECV values were below 0.157 MPa, while the RMSEP values did not exceed 0.151 MPa in both commercial vineyards. This work also includes the development of a software which automates data acquisition and allows faster (up to 40% of time saving in the field) and more efficient deployment of the developed algorithm. The encouraging results presented in this work demonstrate the great potential of this methodology to assess the water status of the vineyard and estimate its spatial variability in different commercial vineyards, providing useful information for better irrigation scheduling.
Why it matches plant phenotyping methodsブドウ樹の水分状態という植物生理形質を、移動車両搭載NIRセンサーで推定・空間マッピングする手法を比較検証し、スペクトル取得自動化ソフトウェアも開発しており、フェノタイピング手法が中心である。
abstractto compare and assess grapevine water status using two different spectral devices assembled in a mobile platform and to evaluate their capability to map the spatial variability of the plant water status
Abstract Feature selection, reducing number of input variables to develop classification model, is an important process to reduce computational and modelling complexity and affects the performance of image process. In this paper, we have proposed new statistical approaches for feature selection based on sample selection. We have applied our new approaches to grapevine leaves data that possesses properties of shape, thickness, featheriness, and slickness are investigated in images. To analyze such kind of data by using image process, thousands of features are created and selection of features plays important role to predict the outcome properly. In our numerical study, Convolutional Neural Networks (CNNs) have been used as feature extractors and then obtained features from the last average pooling layer to detect the type of grapevine leaves from images. These features have been reduced by using our suggested four statistical methods: Simple random sampling (SRS), ranked set sampling (RSS), extreme ranked set sampling (ERSS), Moving extreme ranked set sampling (MERSS). Then selected features have been classified with Artificial Neural Network (ANN) and we have obtained the best accuracy of 97.33% with our proposed approaches. Based on our empirical analysis, it has been determined that the proposed approach exhibits efficacy in the classification of grapevine leaf types. Furthermore, it possesses the potential for integration into various computational devices.
Why it matches plant phenotyping methodsCNN特徴抽出と新規特徴選択法を用いてブドウ葉画像の形態的な葉タイプ分類を行う手法開発が中心であり、画像から植物器官の表現型を推定する研究と判断します。
abstractwe have proposed new statistical approaches for feature selection based on sample selection
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 · publicThis dataset was created by Koklu et al. ( 2022 ) and obtained from the website http://www.muratkoklu.com/datasets/Grapevine_Leaves_Image_Dataset.rar .Open asset ↗http://www.muratkoklu.com/datasets/Grapevine_Leaves_Image_Dataset.rar · Grapevine_Leaves_Image_Dataset.rarlines:151-284Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
The use of RGB cameras or multispectral imaging systems can provide a wide range of applications for crop monitoring, plant phenotyping and disease detection. Although several approaches have been proposed, they increasingly use convolutional neural network-based architectures, which have, however, become increasingly cumbersome for improving classification results and difficult to train with few labeled data. Other increasingly popular approaches consist of using an ensemble of convolutional neural networks, in which each model solves a different problem. Since the inference is time- and resource-consuming due to the execution of multiple models, recent works have focused on transferring knowledge from an ensemble of models to a compact model to obtain better performance. In this paper, we propose an original approach that improves both accuracy and speed by reusing feature maps extracted by heterogeneous models from different data. Linked to each model, a transformation block allows keeping the correct number of feature maps and changing their dimension if necessary. To generate the feature maps, we only need the first layers of the ensemble models, thus taking advantage of ensemble learning methods, while adding only a few layers of a second model dedicated to aggregation of features. This approach allows an ensemble of models to be combined with different architectures that can process different data, such as several representations of the same input image or multispectral images, while being fast enough at the inference stage. This approach is adapted to hierarchical classification tasks by re-exploiting the same feature maps with different transformation blocks, offering accuracy gains in tasks not handled by the ensemble model. The results are provided for the PlantVillage dataset, with RGB images converted to three different color spaces, and for a custom Grapevine Yellow dataset, with multispectral images acquired with two different multispectral cameras.
Why it matches plant phenotyping methods植物画像から病害状態を推定する分類手法の開発が中心であり、異種モデルの特徴マップ融合による精度・速度向上を検証している。
abstractIn this paper, we propose an original approach that improves both accuracy and speed by reusing feature maps extracted by heterogeneous models from different data.
While point clouds hold promise for measuring the geometrical features of 3D objects, their application to plants remains problematic. Plants are three dimensional (3D) organisms whose morphology is complex, varies from one individual to another and changes over time. Objective measurement of attributes in 3D point cloud domain is increasingly attractive as techniques improve the accuracy and reduce computational time. Analysis of point cloud data, however, is not straightforward, due to its discrete nature, imaging noise and cluttered background. In this paper, we introduce a robust method for the direct analysis of plants of point cloud data. To this end, we generalise the random sample consensus (RANSAC) algorithm for the analysis of 3D point cloud data and then use it to model different plant organs. Since 3D point clouds are obtained from multi-view stereo images, they are often contaminated with a considerable level of noise, distortions and out-of-distribution points. Key to our approach is the use of the RANSAC algorithm on 3D point cloud, making our technique more robust to undesirable outliers. We tested our proposed method on Brassica and grapevine by comparing the estimated measurements extracted from the models with manual ones taken from the actual plants. Our proposed method achieved R2>0.90 for measured diameters of branches and stems in Brassica while it yielded R2>0.91 for the measured leaf angles of grapevine and branch angles of Brassica. In all cases, the approach produced stable performance under imaging noise and cluttered background while the conventional methods often failed to work.
Why it matches plant phenotyping methods3D点群から植物器官の形態形質を抽出するRANSAC手法を開発し、実測値との比較で検証しており、植物フェノタイピング手法が研究の中心である。
abstractwe generalise the random sample consensus (RANSAC) algorithm for the analysis of 3D point cloud data and then use it to model different plant organs.
Grapevine phenotyping is the process of determining the physical properties (e.g., size, shape, and number) of grape bunches and berries. Grapevine phenotyping information provides valuable characteristics to monitor the sanitary status of the vine. Knowing the number and dimensions of bunches and berries at an early stage of development provides relevant information to the winegrowers about the yield to be harvested. However, the process of counting and measuring is usually done manually, which is laborious and time-consuming. Previous studies have attempted to implement bunch detection on red bunches in vineyards with leaf removal and surveys have been done using ground vehicles and handled cameras. However, Unmanned Aerial Vehicles (UAV) mounted with RGB cameras, along with computer vision techniques offer a cheap, robust, and timesaving alternative. Therefore, Multi-object tracking and segmentation (MOTS) is utilized in this study to determine the traits of individual white grape bunches and berries from RGB videos obtained from a UAV acquired over a commercial vineyard with a high density of leaves. To achieve this goal two datasets with labelled images and phenotyping measurements were created and made available in a public repository. PointTrack algorithm was used for detecting and tracking the grape bunches, and two instance segmentation algorithms - YOLACT and Spatial Embeddings - have been compared for finding the most suitable approach to detect berries. It was found that the detection performs adequately for cluster detection with a MODSA of 93.85. For tracking, the results were not sufficient when trained with 679 frames.This study provides an automated pipeline for the extraction of several grape phenotyping traits described by the International Organization of Vine and Wine (OIV) descriptors. The selected OIV descriptors are the bunch length, width, and shape (codes 202, 203, and 208, respectively) and the berry length, width, and shape (codes 220, 221, and 223, respectively). Lastly, the comparison regarding the number of detected berries per bunch indicated that Spatial Embeddings assessed berry counting more accurately (79.5%) than YOLACT (44.6%).
Why it matches plant phenotyping methodsUAV RGB動画と画像解析によりブドウ房・果粒の形状や寸法を自動抽出するパイプラインを開発・比較し、データセットも公開しており、植物表現型取得が中心である。
abstractMulti-object tracking and segmentation (MOTS) is utilized in this study to determine the traits of individual white grape bunches and berries from RGB videos obtained from a UAV
Infrared spectroscopy is widely used in viticulture. Spectroscopy correlates spectral properties with reference data to obtain calibrations later used to predict the analyte content in new samples with a single spectral measurement. However, the main limitation lies in generating the reference data required to build robust prediction calibrations. This study proposes a data generation strategy to obtain reference data for larger spectral datasets. A reduced sample set was used to develop initial calibrations. These initial calibrations were subsequently applied to predict the reference data in larger spectral datasets. Calibrations for nitrogen, carbon, and hydrogen content were then attempted using the larger generated datasets. The initial nitrogen calibrations per organ showed coefficients of determination in validation (R²val) between 80.08 and 89.93%. The root mean square errors of prediction (RMSEP) ranged from 0.10 to 0.18% dry matter, and the residual predictive deviations in validation (RPD) were between 2.27 and 3.19. The larger predicted datasets showed improved prediction accuracy with coefficients of determination in validation values above 91.79%, root mean square errors of prediction below 0.14% dry matter, and residual predictive deviations in validation above 3.49. The carbon calibrations showed, on average, a 20% increase in the coefficient of determination in validation decreased root mean square errors of prediction and increased residual predictive deviations in validation. The hydrogen calibrations showed a similar increase in prediction accuracy. The results showed the suitability of using reduced sample sets to generate the reference data of larger datasets capable of yielding more accurate prediction calibrations.
Why it matches plant phenotyping methodsブドウ樹器官の栄養成分を赤外分光で推定する校正法と、少数サンプルから大規模データの参照値を生成する戦略を開発・検証しており、表現型取得法が研究の中心である。
abstractThis study proposes a data generation strategy to obtain reference data for larger spectral datasets.
Monitoring the phenological development stages of grapes represents a challenge in viticulture. It includes the phenological distinction of the growth stages of grapevines and the continuous technological developments, especially in computer vision, enabling a detailed classification of economically relevant development stages of grapes. In the present work, we show that based on a cascading computer vision approach, the development stages of grapes can be classified and distinguished at the micro level. In a comparative experiment (ResNet, DenseNet, InceptionV3), it could be shown that a ResNet architecture provides the best classification results with an average accuracy of 88.1%.
Why it matches plant phenotyping methodsブドウの生育ステージという植物状態をコンピュータビジョンで分類し、複数の深層学習手法を比較評価しているため、表現型取得・推定法が中心である。
abstractbased on a cascading computer vision approach, the development stages of grapes can be classified and distinguished at the micro level
The measurement of geometric canopy parameters in woody crops is an important task in Precision Agriculture because of their correlation with crop condition and productivity. In recent years, several technological approaches have been developed as an alternative to manual measurements, which are time- and labour-consuming. Two of the most commonly used 3D canopy characterization technologies are mobile terrestrial laser scanning (MTLS) based on light detection and ranging (LiDAR) sensors, and digital aerial photogrammetry (DAP) using imagery from uncrewed aerial vehicles (UAVs). Although both are state-of-the-art and have been fully tested and validated, a complete comparison between their geometric canopy parameter estimations in different woody crops and training systems has not been carried out. For this reason, a set of geometric parameters (canopy height, projected area, and volume) of a vineyard, an intensive peach orchard, and an intensive pear orchard were measured using UAV-DAP and MTLS-LiDAR. A comparison between both kinds of measurements was performed, accounting for the length of the sections in which the crop hedgerows were divided to extract the geometric parameters. Measurements from the UAV and the MTLS were highly correlated (R2 from 0.82 to 0.94) when considering the data from the three crops together, and the correlations were higher when analysing longer row sections. The canopy geometric parameters estimated using the MTLS-LiDAR always had higher values than those from the UAV-DAP. The results presented in this work provide useful data for a more informed selection of technological approaches for 3D crop characterization in Precision Fruticulture and high-throughput phenotyping.
Why it matches plant phenotyping methodsUAV-DAPとMTLS-LiDARによる果樹キャノピー形状計測を比較・検証し、高スループット表現型解析への適用可能性を評価しているため、植物形質取得法が中心である。
abstractA comparison between both kinds of measurements was performed
Abstract Plant diseases are a major factor contributing to agricultural production losses, necessitating effective disease detection and classification methods. Traditional manual approaches heavily rely on expert knowledge, which can introduce biases. However, advancements in computing and image processing have opened up possibilities for leveraging these technologies to assist non-experts in managing plant diseases. Particularly, deep learning techniques have shown remarkable success in assessing and classifying plant health based on digital images. This paper focuses on fine-tuning state-of-the-art pre-trained convolutional neural network (CNN) models and vision transformer models for the detection and diagnosis of grape leaves and diseases using digital images.The experiments were conducted using two datasets: PlantVillage, which encompasses four classes of grape diseases (Black Rot, Leaf Blight, Healthy, and Esca leaves), and Grapevine, which includes five classes for leaf recognition (Ak, Alaidris, Buzgulu, Dimnit, and Nazli). The results of the experiments, involving a total of 14 models based on six well-known CNN architectures and 17 models based on five widely recognized vision transformer architectures, demonstrated the capability of deep learning techniques in accurately distinguishing between grape diseases and recognizing grape leaves. Notably, four CNN models and four vision transformer models achieved 100% accuracy on the test data from the PlantVillage dataset, while one CNN model and one vision transformer model achieved 100% accuracy on the Grapevine dataset. Among the models tested, the Swinv2-Base model stood out by achieving 100% accuracy on both the PlantVillage and Grapevine datasets. The proposed deep learning-based approach is believed to have the potential to enhance crop productivity through early detection of grape diseases. Additionally, it is expected to offer a fresh perspective to the agricultural sector by providing insights into the characterization of various grape varieties.
Why it matches plant phenotyping methodsブドウ葉の画像から葉の識別と病害状態を分類する深層学習手法を開発・比較しており、植物表現型の取得・推定が研究の中心である。
abstractThis paper focuses on fine-tuning state-of-the-art pre-trained convolutional neural network (CNN) models and vision transformer models for the detection and diagnosis of grape leaves and diseases using digital images.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 7 Sept 2026
Abstract Background Grapevine berries undergo asynchronous growth and ripening dynamics within the same bunch. Due to the lack of efficient methods to perform sequential non-destructive measurements on a representative number of individual berries, the genetic and environmental origins of this heterogeneity, as well as its impacts on both vine yield and wine quality, remain nearly unknown. To address these limitations, we propose to track the growth and coloration kinetics of individual berries on time-lapse images of grapevine bunches. Result First, a deep-learning approach is used to detect berries with at least 50±10% of visible contours, and infer the shape they would have in the absence of occlusions. Second, a tracking algorithm was developed to assign a common label to shapes representing the same berry along the time-series. Training and validation of the methods were performed on challenging image datasets acquired in a robotised high-throughput phenotyping platform. Berries were detected on various genotypes with a F1-score of 91.8%, and segmented with a mean absolute error of 4.1% on their area. Tracking allowed to label and retrieve the temporal identity of more than half of the segmented berries, with an accuracy of 98.1%. This method was used to extract individual growth and colour kinetics of various berries from the same bunch, allowing us to propose the first statistically relevant analysis of berry ripening kinetics, with a time resolution lower than one day. Conclusions We successfully developed a fully-automated open-source method to detect, segment and track overlapping berries in time-series of grapevine bunch images. This makes it possible to quantify fine aspects of individual berry development, and to characterise the asynchrony within the bunch. The interest of such analysis was illustrated here for one genotype, but the method has the potential to be applied in a high throughput phenotyping context. This opens the way for revisiting the genetic and environmental variations of the ripening dynamics. Such variations could be considered both from the point of view of fruit development and the phenological structure of the population, which would constitute a paradigm shift.
Why it matches plant phenotyping methodsブドウ果粒の検出・セグメンテーション・時系列追跡を開発し、ロボット型高スループット表現型解析基盤で検証して、個別果粒の成長・着色形質を抽出する方法が中心である。
abstractwe propose to track the growth and coloration kinetics of individual berries on time-lapse images of grapevine bunches
Fungal infection of grape berries (Vitis vinifera) by Botrytis cinerea frequently coincides with harvest, impacting both the yield and quality of grape and wine products. A rapid and non-destructive method for identifying B. cinerea infection in grapes at an early stage prior to harvest is critical to manage loss. In this study, zeolitic imidazolate framework-8 (ZIF-8) crystal was applied as an absorbent material for volatile extraction from B. cinerea infected and healthy grapes in a vineyard, followed by thermal desorption gas chromatography–mass spectrometry. The performance of ZIF-8 in regard to absorbing and trapping the targeted volatiles was evaluated with a standard solution of compounds and with a whole bunch of grapes enclosed in a glass container to maintain standard sampling conditions. The results from the sampling methods were then correlated to B. cinerea infection in grapes, as measured and determined by genus-specific antigen quantification. Trace levels of targeted compounds reported as markers of grape B. cinerea infection were successfully detected with in-field sampling. The peak area counts for volatiles 3-octanone, 1-octen-3-one, 3-octanol, and 1-octen-3-ol extracted using ZIF-8 were significantly higher than values achieved using Tenax®-TA from field testing and demonstrated good correlation with B. cinerea infection severities determined by B. cinerea antigen detection.
Why it matches plant phenotyping methodsブドウの感染状態を非破壊的に推定する揮発性物質サンプリング法を開発・性能評価しており、植物病害状態の取得が中心である。
abstractA rapid and non-destructive method for identifying B. cinerea infection in grapes at an early stage prior to harvest is critical to manage loss.
Several variables, including a rising human population, varying weather patterns in the context of ongoing climate change, and the rapid worldwide spread of epidemics, all contribute to boosting agricultural demand. To assure food availability, quality, and safety while increasing yields and profitability, precision agriculture must progress swiftly. Precision viticulture aims to optimize vineyard management in this setting by reducing resource consumption and environmental impact while simultaneously enhancing the yield, product quality, and oenological potential of vineyards. This comprehensive review article offers an overview of the real-world and laboratory applications of optical and non-optical sensors in precision viticulture for 3D modelling. Hence, there is a pressing need to track the development of crops at a wide range of spatial and temporal scales, in a wide variety of environments, and for a wide range of objectives in a non-destructive manner. Due to the intrinsic spatial heterogeneity of vineyards, the adoption of precision viticulture necessitates crop monitoring using contactless and non-invasive sensors such as ultrasonic, LiDAR (Light Detection and Ranging), depth, or RGB cameras to prevent low accuracy and sparse sampling. This study aims to assist researchers in gaining a broad understanding of the sensing technologies for precision viticulture, the present problems, and the advancement of the state of the art. The study focuses on sensors used for Proximal Sensing to geometrically characterize vines using statically or dynamically ground-based measurements through a wide range of mobile sensing platforms. The employed sensors, data extraction, and analysis procedures are described. Moreover, the present and future potential of Proximal Sensing and Remote Sensing in vineyards is discussed.
Why it matches plant phenotyping methodsブドウ樹の形状を光学・非光学センサーで計測・3Dモデル化する近接センシング手法を中心に扱うレビューであり、植物フェノタイピング手法レビューに該当する。
abstractThis comprehensive review article offers an overview of the real-world and laboratory applications of optical and non-optical sensors in precision viticulture for 3D modelling.
The concentration of magnetic particulate matter (PM) on the leaf surface (an indicator of current pollution) and topsoil (an indicator of magnetic PMs which have geogenic natural signal or historical pollution origin) was assessed in agricultural areas (conventional and organic vineyards). The main aim of this study was to explore whether magnetic parameters such as saturation isothermal remanent magnetization (SIRM) and mass-specific magnetic susceptibility (χ) can be a proxy for magnetic particulate matter (PM) pollution and associated potentially toxic elements (PTEs) in agricultural areas. Besides, wavelength dispersive X-ray fluorescence spectroscopy (WD-XRF) was investigated as a screening method for total PTE content in soil and leaf samples. Both magnetic parameters (SIRM and χ) pinpoint soil pollution, while SIRM was more suitable for evaluating magnetic PM accumulated on leaves. The values of both magnetic parameters were significantly (p < 0.01) correlated within the same type of sample (soil-soil or leaf-leaf), but not between different matrixes (soil-leaf). Differences between magnetic particles' grain sizes among vegetation seasons in vineyards were obtained by observing the SIRM/χ ratio. WD-XRF was revealed to be an appropriate screening method for soil and leaf total element contents in agricultural ambient. For a more precise application of WD-XRF leaf measurements, specific calibration using a similar matrix to plant material is required. In parallel, measurements of SIRM, χ, and element content (by WD-XRF) can be recommended as user-friendly, fast, and eco-sustainable techniques for determining magnetic PM and PTE pollution hotspots in agricultural ambient.
Why it matches plant phenotyping methods植物葉に蓄積した磁性粒子・元素含量という植物状態を対象に、磁気パラメータとWD-XRFをスクリーニング／評価手法として検証しており、単なる環境測定や routine な植物測定ではなく、葉試料の汚染状態を取得する方法が中心です。
abstractThe main aim of this study was to explore whether magnetic parameters such as saturation isothermal remanent magnetization (SIRM) and mass-specific magnetic susceptibility (χ) can be a proxy for magnetic particulate matter (PM) pollution and associated potentially toxic elements (PTEs) in agricultural areas.
SegmentatFinding plant diseases early on is essential for reducing damage while improving the quality of the yield. This paper describes an intelligent cloud-edge-based system for identifying and categorizing diseases in grape plants. The system is based on the support vector machine (SVM), a supervised machine learning technique to classify data. The traits of healthy and diseased plants are identified using digital photographs of grape plants. To determine color and texture information, we retrieved global features from the grape images. Patterns or structures (such as corners or edges) are discovered using the speeded-up robust features (SURF) method. The K-means clustering approach is used to quantify feature space, which lowers the number of feature descriptors. The training set for the SVM classifier is made up of feature descriptors. The system classified the unlabeled grape images using the trained SVM classifier during testing. 1600 RGB pictures from four classes: Black-rot, Black-measles, Leaf-blight, and Healthy-leaf make up the original data set. To evaluate the system and provide accuracy and confusion matrices, simulations are run in four different color spaces (grayscale, RGB, YCbCr, and L*a*b*). In the L*a*b* color space, the system attained a maximum average accuracy of up to 90.63% at a ratio of 70:30 training to testing data.
Why it matches plant phenotyping methodsブドウ葉の画像から健全・病害状態を抽出し、画像特徴量、セグメンテーション／分類、複数色空間で性能評価するシステム開発が中心であるため。
abstractThis paper describes an intelligent cloud-edge-based system for identifying and categorizing diseases in grape plants.
Reproduction assets foundThe paper's grape disease classification uses the public PlantVillage grape leaf image dataset from Kaggle, and the authors also deposited their selected 1600-image dataset on Figshare with an explicit availability statement. No author analysis code or trained model is shared.Dataset · publicAvailability of data and materials
The data set is available at the following share repository:
https://figshare.com/ndownloader/files/37001836Open asset ↗figsharepdf-page:15 lines:1-59Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Jun 2023Computers and Electronics in Agriculture.
In recent years, climate fluctuations have been increasingly extreme, affecting agricultural production. The development of digital agriculture driven by new intelligent sensors is one of the privileged paths to improve farm management. Assessing transpiration E and stomatal conductance gs in real time with optical instruments is a real challenge to detect water stress. In this study, the objective is to evaluate VIS–NIR spectroscopy to predict transpiration E and stomatal conductance gs of grapevine plants (Vitis vinifiera L.). For this purpose, a water stress gradient was obtained using vine pots of three varieties (Syrah, Merlot, Riesling) tested under two water conditions where precise monitoring of physiological variables has been carried out. Hyperspectral images were acquired to form a spectral database and a weather station provided radiation (Rg), relative humidity (RH), temperature (Ta) and wind speed (Ws). First, Partial Least Squares (PLS) models were established to relate spectral data to physiological variables. Then, Sequential Orthogonalized-Partial Least Squares (SO-PLS) was used to predict these physiological variables with two blocks: spectral and climate data. PLS models are obtained for gs (R²= 0.656, bias = 8.76 mmol.m⁻².s⁻¹, RMSE = 64.7 mmol.m⁻².s⁻¹) and E (R²= 0.625, bias=-0.02 mmol.m⁻².s⁻¹, RMSE = 0.67 mmol.m⁻².s⁻¹). For E, improved results (R²= 0.699, bias = 0.055 mmol.m⁻².s⁻¹, RMSE = 0.614 mmol.m⁻².s⁻¹) are obtained by using climate data with SO-PLS. Generic PLS models achieved good predictive quality despite different coloured berry varieties. Quality of these prediction models could be improved by defining varietal models on a larger data set. Merging spectral data with climate data improves prediction quality of transpiration variable providing insights by adding further information with the aim of improving predictive qualities.
Why it matches plant phenotyping methodsVIS–NIRハイパースペクトル画像と気候データからブドウの蒸散・気孔コンダクタンスを推定する手法を構築・評価しており、植物生理形質の取得・予測が研究の中心である。
abstractFirst, Partial Least Squares (PLS) models were established to relate spectral data to physiological variables. Then, Sequential Orthogonalized-Partial Least Squares (SO-PLS) was used to predict these physiological variables with two blocks: spectral and climate data.
Since 2010, more and more farmers have been using remote sensing data from unmanned aerial vehicles, which have a high spatial–temporal resolution, to determine the status of their crops and how their fields change. Imaging sensors, such as multispectral and RGB cameras, are the most widely used tool in vineyards to characterize the vegetative development of the canopy and detect the presence of missing vines along the rows. In this study, the authors propose different approaches to identify and locate each vine within a commercial vineyard using angled RGB images acquired during winter in the dormant period (without canopy leaves), thus minimizing any disturbance to the agronomic practices commonly conducted in the vegetative period. Using a combination of photogrammetric techniques and spatial analysis tools, a workflow was developed to extract each post and vine trunk from a dense point cloud and then assess the number and position of missing vines with high precision. In order to correctly identify the vines and missing vines, the performance of four methods was evaluated, and the best performing one achieved 95.10% precision and 92.72% overall accuracy. The results confirm that the methodology developed represents an effective support in the decision-making processes for the correct management of missing vines, which is essential for preserving a vineyard’s productive capacity and, more importantly, to ensure the farmer’s economic return.
Why it matches plant phenotyping methodsUAV画像とフォトグラメトリを用いてブドウ樹の位置・欠株を抽出する手法を開発し、複数手法の性能評価も行っており、植物状態の測定法が研究の中心である。
abstractthe authors propose different approaches to identify and locate each vine within a commercial vineyard
Over the last decade, there have been renewed efforts into wine grape breeding within the research community. Quick characterisation by phenotyping of quality traits, including aroma composition, remains challenging. Selected Ion Flow Tube Mass Spectrometry (SIFT-MS), a high throughput soft ionisation technology first available in 2008, could be particularly useful for this purpose. In light of this, this technical article summarises recent results obtained with SIFT-MS.
Why it matches plant phenotyping methodsSIFT-MSを用いたブドウ果実の揮発性指紋のハイスループット表現型解析が記事の中心であり、品質形質（香気組成）の取得技術を扱っている。
titleSelected Ion Flow Tube Mass Spectrometry (SIFT-MS): a promising technology for the high throughput phenotyping of grape berry volatile fingerprints
Abstract Developing actionable early detection and warning systems for agricultural stakeholders is crucial to reduce the annual $200B USD losses and environmental impacts associated with crop diseases. Agricultural stakeholders primarily rely on labor‐intensive, expensive scouting and molecular testing to detect disease. Spectroscopic imagery (SI) can improve plant disease management by offering decision‐makers accurate risk maps derived from Machine Learning (ML) models. However, training and deploying ML requires significant computation and storage capabilities. This challenge will become even greater as global‐scale data from the forthcoming Surface Biology & Geology satellite becomes available. This work presents a cloud‐hosted architecture to streamline plant disease detection with SI from NASA’s AVIRIS‐NG platform, using grapevine leafroll‐associated virus complex 3 (GLRaV‐3) as a model system. Here, we showcase a pipeline for processing SI to produce plant disease detection models and demonstrate that the underlying principles of a cloud‐based disease detection system easily accommodate model improvements and shifting data modalities. Our goal is to make the insights derived from SI available to agricultural stakeholders via a platform designed with their needs and values in mind. The key outcome of this work is an innovative, responsive system foundation that can empower agricultural stakeholders to make data‐driven plant disease management decisions while serving as a framework for others pursuing use‐inspired application development for agriculture to follow that ensures social impact and reproducibility while preserving stakeholder privacy.
Why it matches plant phenotyping methods画像分光と機械学習によるブドウ病害状態の検出パイプラインおよびクラウド基盤が研究の中心であり、植物病害表現型の取得・推定手法として適格。
abstractHere, we showcase a pipeline for processing SI to produce plant disease detection models
Grapes are a globally popular fruit, with grape cultivation worldwide being second only to citrus. This article focuses on the low efficiency and accuracy of traditional manual grading of red grape external appearance and proposes a small-sample red grape external appearance grading model based on transfer learning with convolutional neural networks (CNNs). Initially, the CNN transfer learning method was used to transfer the pre-trained AlexNet, VGG16, GoogleNet, InceptionV3, and ResNet50 network models on the ImageNet image dataset to the red grape image grading task. By comparing the classification performance of the CNN models of these five different network depths with fine-tuning, ResNet50 with a learning rate of 0.001 and a loop number of 10 was determined to be the best feature extractor for red grape images. Moreover, given the small number of red grape image samples in this study, different convolutional layer features output by the ResNet50 feature extractor were analyzed layer by layer to determine the effect of deep features extracted by each convolutional layer on SVM classification performance. This analysis helped to obtain a ResNet50+SVM red grape external appearance grading model based on the optimal ResNet50 feature extraction strategy. Experimental data showed that the classification model constructed using the feature parameters extracted from the 10th node of the ResNet50 network achieved an accuracy rate of 95.08% for red grape grading. These research results provide a reference for the online grading of red grape clusters based on external appearance quality and have certain guiding significance for the quality and efficiency of grape industry circulation and production.
Why it matches plant phenotyping methodsブドウ果実房の外観品質を画像から分類・等級化するCNN転移学習手法が研究の中心であり、植物器官の形態的状態を抽出する実質的な画像ベース手法に該当する。
abstractproposes a small-sample red grape external appearance grading model based on transfer learning with convolutional neural networks (CNNs).
Knowledge of the water status in commercial vineyards is of great importance when defining the production objectives and the composition of the grape must. Determining the appropriate irrigation doses allows for adjusting the balance between vigour and productive capacity of the vineyard. However, to accurately know the hydration status of the vines, it is necessary to use equipment such as pressure chambers that are hardly replicable. Much effort has been invested in finding a more straightforward simpler methodology that allows knowing the hydration of plants. In this respect, remote sensing technology is presented as an appropriate tool to obtain information from large areas quickly and efficiently. This work aimed to evaluate the accuracy of water stress detection based on thermal sensors onboard UAVs.The study was carried out in the Merlot vineyard located in Toledo-Spain; arranged on a trellis with a 2.60 x 1.10 m planting frame and established in 2002. High-resolution thermal images were obtained on different dates during the 2021 and 2022 irrigation campaign and at two intervals of the day (9:00 and 12:00 solar hours). Stem water potential (Ψm) and chlorophyll were measured at the same time.The results indicate that there are statistically significant differences between the different irrigation treatments. These differences were mainly observed in the water-steam potential measurements made in the morning.ReferencesAcevedo-Opazo, C., Tisseyre, B., Guillaume, S., & Ojeda, H. (2008). The potential of high spatial resolution information to define within-vineyard zones related to vine water status. Precision Agriculture, 9(5), 285–302. https://doi.org/10.1007/s11119-008-9073-1.Jackson, R. D. (1982). Canopy Temperature and Crop Water Stress. 1, 43–85. https://doi.org/10.1016/b978-0-12-024301-3.50009-5.Poblete-Echeverría, C., Sepulveda-Reyes, D., Ortega-Farias, S., Zuñiga, M., & Fuentes, S. (2016). Plant water stress detection based on aerial and terrestrial infrared thermography: A study case from vineyard and olive orchard. Acta Horticulturae, 1112, 141–146. https://doi.org/10.17660/ActaHortic.2016.1112.20. Acknowledgements:The authors want to thank Bodegas y Viñas Casa del Valle for allowing us to work in their vineyards and the company UTW for supply the drone images. Financial support provided by Comunidad de Madrid through calls for grants for the completion of Industrial Doctorates IND2020/AMB-17341 is greatly appreciated.
Why it matches plant phenotyping methodsUAV搭載熱センサーによるブドウ樹の水ストレス検出精度を評価しており、植物の生理状態を推定するセンシング手法の検証が中心である。
abstractThis work aimed to evaluate the accuracy of water stress detection based on thermal sensors onboard UAVs.
Water status in vineyards is a determining factor, given its relationship with productive and physiological parameters such as vegetative growth, berry ripening, yield and overall wine quality. In-field measurements, through a pressure chamber, provide very accurate and reliable measurements of midday stem water potential (Ψstem), a direct method for determining a plant’s water status by quantifying the tension with which water is retained in the leaf. Despite the robustness of this method, it is not practically applied to extensive commercial vineyards as it is a labour-intensive practice which can narrowly evaluate the significant intra-field variability. Remote sensing offers large-scale information at a single point in time without the need to be physically present in the field. This study aims to assess the use of multispectral imagery from Worldview-3, a commercial satellite, as a tool to indirectly estimate water status in the vineyard through different Vegetation Indexes (VI).This research was carried out in a commercial Merlot vineyard in Yepes (Toledo), an arid area in central Spain where rainfall and irrigation water availability is scarce. The vines were established in 2002 and arranged on a trellis with a plantation spacing of 2.6 x 1.1 m. Five different irrigation doses were tested to obtain variability in vine water status. Drip irrigation emitters were identical in all treatments ( 2 l h-1), but distances between emitters were adjusted to modify irrigation levels. Treatments were designed as follows: T1 (100% dose) emitters every 0.25 m, T2 (50%) emitters every 0.5 m, T3 (25%) emitters every 1.0 m, T4 (0%) no emitters and T5 (25%) underground emitters every 1.0 m. The results will be discussed in the context of deficit irrigation.
Why it matches plant phenotyping methodsWorldView-3マルチスペクトル画像と植生指数を用いて、ブドウ樹の水分状態という植物生理形質を間接推定する手法の評価が研究目的であり、測定法が中心的です。
abstractThis study aims to assess the use of multispectral imagery from Worldview-3, a commercial satellite, as a tool to indirectly estimate water status in the vineyard through different Vegetation Indexes (VI).
In agricultural systems, rapid information from data collection and processing is an important factor for stakeholders and researchers to correctly account for the spatial and temporal variability of crop and soil factors. The aim of the present study was to investigate soil-plant-water systems and interactions using manual and remote sensing techniques in a small agricultural catchment. Four land use types of forest, grassland, vineyard, and cropland (sunflower) were investigated in different slope positions. At the same time, three different tillage practices were applied in the vineyard between the rows: grassed (NT), cover cropped (CC), and tilled (T) inter rows. We evaluated NDVI measurements from three different sources (PlantPen - PP, Meter Group - MG, Sentinel-2 - S2) representing different scales (leaves, 0.33m2, and 100m2). We also compared ground and satellite measurements of varying vegetation indices.Spectral reflectance sensors were used on the slopes of grassland, cropland, and three vineyard sites. The Normalized Difference Vegetation Index (NDVI) and Photochemical Reflectance Index (PRI) sensors were used to measure leaf reflectance. A hemispherical sensor set was used for each measurement. Hand-held instruments were used to measure the topsoil soil water content (SWC) and temperature, leaf NDVI and chlorophyll concentrations, and Leaf Area Index (LAI) every two weeks. Satellite data, such as NDVI, green (GCI) and red edge (RECI) chlorophyll indices, and soil-adjusted vegetation index (SAVI), were obtained from the Sentinel-2 database on days when both ground and satellite overpass occurred within 24 hours.Land use types and slope position have a strong influence on vegetation growth. The highest overall NDVI and leaf chlorophyll values were observed in vineyard and forest samples, and the lowest in grassland. SWC and temperature were the lowest in the forest and vineyards. SWCs were significantly different for T and CC samples (p 0.05). For the other three land use types, there were no significant differences in values between slope positions. Chlorophyll data showed a very strong correlation between Sentinel-2 retrieved data and hand-held measurements, with r=0.84 for grassland (GCI), r=0.83 for NT (GCI), and r=0.87 for T (RECI). Strong correlations were found between the different sources of NDVI for the grassland samples (e.g. r=0.97, p
Why it matches plant phenotyping methods葉から圃場まで異なるセンサー・リモートセンシング手法によるNDVI等の植生指標を比較し、地上測定と衛星推定値の相関を評価しているため、植物状態の取得手法の検証が中心です。
abstractThe aim of the present study was to investigate soil-plant-water systems and interactions using manual and remote sensing techniques in a small agricultural catchment.
The grapevine is vulnerable to diseases, deficiencies, and pests, leading to significant yield losses. Current disease controls involve monitoring and spraying phytosanitary products at the vineyard block scale. However, automatic detection of disease symptoms could reduce the use of these products and treat diseases before they spread. Flavescence dorée (FD), a highly infectious disease that causes significant yield losses, is only diagnosed by identifying symptoms on three grapevine organs: leaf, shoot, and bunch. Its diagnosis is carried out by scouting experts, as many other diseases and stresses, either biotic or abiotic, imply similar symptoms (but not all at the same time). These experts need a decision support tool to improve their scouting efficiency. To address this, a dataset of 1483 RGB images of grapevines affected by various diseases and stresses, including FD, was acquired by proximal sensing. The images were taken in the field at a distance of 1-2 meters to capture entire grapevines and an industrial flash was ensuring a constant luminance on the images regardless of the environmental circumstances. Images of 5 grape varieties (Cabernet sauvignon, Cabernet franc, Merlot, Ugni blanc and Sauvignon blanc) were acquired during 2 years (2020 and 2021). Two types of annotations were made: expert diagnosis at the grapevine scale in the field and symptom annotations at the leaf, shoot, and bunch levels on computer. On 744 images, the leaves were annotated and divided into three classes: 'FD symptomatic leaves', 'Esca symptomatic leaves', and 'Confounding leaves'. Symptomatic bunches and shoots were, in addition of leaves, annotated on 110 images using bounding boxes and broken lines, respectively. Additionally, 128 segmentation masks were created to allow the detection of the symptomatic shoots and bunches by segmentation algorithms and compare the results to those of the detection algorithms.
Why it matches plant phenotyping methodsブドウ病害の症状を画像から抽出するための専門家アノテーション付きデータセットであり、植物体・葉・枝・果房の病徴状態を対象とするフェノタイピング手法・ベンチマークとして中心的です。
abstractTo address this, a dataset of 1483 RGB images of grapevines affected by various diseases and stresses, including FD, was acquired by proximal sensing.
Reproduction assets foundThis Data Brief describes a paper-specific grapevine disease image dataset (1483 RGB images with expert annotations) publicly deposited on Mendeley Data, with a direct URL provided in the article.Dataset · publice:
○
Plot 1: 44.6992974, -0.3924154
•
City/Town/Region: Rions, Gironde
Latitude and longitude:
○
Plot 1: 44.6704526, -0.3561660
○
Plot 2: 44.6726088, -0.3610193
•
City/Town/Region: Saint-Martin, Gironde
Latitude and longitude:
○
Plot 1: 44.5712274, -0.1697558
Data accessibility
Repository name: Mendeley Data
Direct URL to data: https://data.mendeley.com/datasets/3dr9r3w3jn/2
Related research article
Tardif, M., Amri, A., Keresztes, B., Deshayes, A., Martin, D., Greven, M., & Da Costa, J.-P. (2022). Two-stage automatic diagnosis of Flavescence Dorée based on proximal imaging and artificial intelligence: a multi-year and multi-variety experimental study. OENO One, 56(3), 371–384. https://doi.oOpen asset ↗Mendeley Data · 3dr9r3w3jn/2lines:46-137Plant phenotyping relevance match · UnverifiedEurope PMC · checked 14 Sept 2026
Phenotypic traits of grapevines are known to be closely related to grapevine yield, wine flavour and sensitivity to disease. Traditional phenotyping methods based on manual measurements face the bottleneck of intensely repetitive and time consuming measurement. As viticulturists turning their focus to the 3D automatic phenotyping domain, existing work on grape bunch phenotyping indicates deficiencies in incomplete reconstruction information, poor element coincidence with the ground truth and poor performance under field conditions. To this end, the proposed work introduces a novel reconstruction pipeline by dividing it into sub-problems of visible-berry-related reconstruction and invisible element prediction. By taking a 2D image of the target bunch as the only sensor input, visible berries are detected using image processing algorithms. With the detected berry information, their morphological positions are predicted, from which internodes that are associated with the detected berries are derived. Parameters of derived internodes are then estimated employing constraint-based optimisation theory, from which the visible-berry-related reconstruction is able to be achieved. Invisible element prediction is then conducted by filling elements according to a Restricted Reconstruction Grammar (RRG). A fully reconstructed bunch model is finally presented. Compared with existing work, the proposed grape bunch reconstruction pipeline achieved an improvement in quantity estimation of rachis internodes, tertiary internodes and pedicels, whose percentage errors were indicated as 21.2, 42.2 and 31.2% respectively. A better performance was also revealed in length estimations of secondary internodes and pedicels with percentage errors of 3.5 and 0.5%. This may largely facilitate related studies on disease control of grape bunches since internode numbers and lengths are closely related to bunch compactness which is the indicator of disease sensitivity. Especially, the proposed reconstruction pipeline shows an promising improvement in element coincidence with F1 scores of 0.90, 0.77, 0.45, 0.43 for respective element types. Knowing that element coincidence may influence the inner space utilisation of a bunch, the proposed work provides a better option for 3D grape bunch phenotyping.
Why it matches plant phenotyping methods画像からブドウ房を3D再構成し、房の構成要素の数量・長さ・位置を推定するパイプラインを開発・評価しており、植物表現型取得手法が研究の中心である。
abstractthe proposed work introduces a novel reconstruction pipeline
Precise and reliable identification of specific plant diseases is a challenge within precision agriculture nowadays. This is the case of esca, a complex grapevine trunk disease, that represents a major threat to modern viticulture as it is responsible for large economic losses annually. The lack of effective control strategies and the complexity of esca disease expression make essential the identification of affected plants, before symptoms become evident, for a better management of the vineyard. This study evaluated the suitability of a near-infrared hyperspectral imaging (HSI) system to detect esca disease in asymptomatic grapevine leaves of Tempranillo red-berried cultivar. For this, 72 leaves from an experimental vineyard, naturally infected with esca, were collected and scanned with a lab-scale HSI system in the 900-1700 nm spectral range. Then, effective image processing and multivariate analysis techniques were merged to develop pixel-based classification models for the distinction of healthy, asymptomatic and symptomatic leaves. Automatic and interval partial least squares variable selection methods were tested to identify the most relevant wavelengths for the detection of esca-affected vines using partial least squares discriminant analysis and different pre-processing techniques. Three-class and two-class classifiers were carried out to differentiate healthy, asymptomatic and symptomatic leaf pixels, and healthy from asymptomatic pixels, respectively. Both variable selection methods performed similarly, achieving good classification rates in the range of 82.77-97.17% in validation datasets for either three-class or two-class classifiers. The latter results demonstrated the capability of hyperspectral imaging to distinguish two groups of seemingly identical leaves (healthy and asymptomatic). These findings would ease the annual monitoring of disease incidence in the vineyard and, therefore, better crop management and decision making.
Why it matches plant phenotyping methodsブドウ葉の病徴状態をハイパースペクトル画像と画像処理・多変量解析で推定し、分類モデルを検証しており、植物表現型取得法が研究の中心である。
abstractThis study evaluated the suitability of a near-infrared hyperspectral imaging (HSI) system to detect esca disease in asymptomatic grapevine leaves
One of the primary issues with inside the agricultural region is crop sicknesses and automated detection is crucial for crop monitoring. Plant leaves generally display the maximum ailment symptoms, however professional laboratory leaf analysis is high priced and time-consuming. According to the Food and Agriculture Organization (FAO), agricultural pests lessen crop yields international via way of means of 20 to 40% according to year. Smart farming is good solution for farmers is to apply artificial intelligence techniques along with modern statistics and communication technologies to get rid of those dangerous insect infestations. Farmers can substantially lessen financial losses via way of means of treating plants directly and detecting sicknesses and pests in apple, mango and graphs leaves appropriately and timely. Image type accuracy has progressed notably because of latest advances in deep studying-primarily based totally convolutional neural networks (CNN). This article evolved strategies primarily based totally on deep studying to hit upon sicknesses and pests in apples, mango, and grape leaves. These strategies have been influenced via way of means of the achievement of CNNs in photograph type.
Why it matches plant phenotyping methods葉画像から植物の病害状態をCNNで推定する手法が中心であり、植物病害の画像ベース表現型評価に該当する。
abstractThis article evolved strategies primarily based totally on deep studying to hit upon sicknesses and pests in apples, mango, and grape leaves.
Climate and water availability greatly affect each season's grape yield and quality. Using models to accurately predict environment impacts on fruit productivity and quality is a huge challenge. We calibrated and validated the functional-structural model, GrapevineXL, with a data set including grapevine seasonal midday stem water potential (Ψ xylem ), berry dry weight (DW), fresh weight (FW), and sugar concentration per volume ([Sugar]) for a wine grape cultivar ( Vitis vinifera cv. Cabernet Franc) in field conditions over 13 years in Bordeaux, France. Our results showed that the model could make a fair prediction of seasonal Ψ xylem and good-to-excellent predictions of berry DW, FW, [Sugar] and leaf gas exchange responses to predawn and midday leaf water potentials under diverse environmental conditions with 14 key parameters. By running virtual experiments to mimic climate change, an advanced veraison (i.e. the onset of ripening) of 14 and 28 days led to significant decreases of berry FW by 2.70% and 3.22%, clear increases of berry [Sugar] by 2.90% and 4.29%, and shortened ripening duration in 8 out of 13 simulated years, respectively. Moreover, the impact of the advanced veraison varied with seasonal patterns of climate and soil water availability. Overall, the results showed that the GrapevineXL model can predict plant water use and berry growth in field conditions and could serve as a valuable tool for designing sustainable vineyard management strategies to cope with climate change.
Why it matches plant phenotyping methodsGrapevineXLという機能・構造モデルを13年間の圃場データで較正・検証し、水分状態、果粒成長、糖濃度などの植物形質を予測する方法が研究の中心である。
abstractWe calibrated and validated the functional-structural model, GrapevineXL, with a data set including grapevine seasonal midday stem water potential (Ψ xylem ), berry dry weight (DW), fresh weight (FW), and sugar concentration per volume ([Sugar])
In most of the countries, grapes are considered as a cash crop. Currently huge research is going on in development of automated grape harvesting systems. Speedy and reliable grape bunch detection is prime need for various deep learning based automated systems which deals with object detection and object segmentation tasks. But currently very few datasets are available on grape bunches in vineyard, because of which there is restriction to the research in this area. In comparison to the vineyard in outside countries, Indian vineyard structure is more complex, so it becomes hard to work in real-time. To overcome these problems and to make vineyard dataset for suitable for Indian vineyard scenarios, this paper proposed four different datasets on grape bunches in vineyard. For creating all datasets in GrapesNet, natural environmental conditions have been considered. GrapesNet includes total 11000+ images of grape bunches. Necessary data for weight prediction of grape cluster is also provided with dataset like height, width and real weight of cluster present in image. Proposed datasets can be used for prime tasks like grape bunch detection, grape bunch segmentation, and grape bunch weight estimation etc. of future generation automated vineyard harvesting technologies.
Why it matches plant phenotyping methodsブドウ果房画像データセットを構築し、果房の検出・セグメンテーションに加えて重量推定用の寸法と実重量を提供することが中心で、再利用可能な植物表現型データセットに該当する。
abstractthis paper proposed four different datasets on grape bunches in vineyard.
Reproduction assets foundThe paper is a data descriptor for GrapesNet, a public Mendeley Data repository of Indian vineyard RGB/RGB-D grape bunch image datasets with ground-truth cluster height, width, and weight measurements used for phenotyping tasks (detection, segmentation, weight estimation). The dataset is the paper's core asset and is aDataset · publicRepository name: GrapesNet: Indian Grape Clusters RGB & RGB-D Image Datasets
Data identification number (DOI): 10.17632/mhzmzd5cwx.1
Direct URL to data: https://data.mendeley.com/datasets/mhzmzd5cwx/1Open asset ↗10.17632/mhzmzd5cwx.1lines:1-95Code / dataset availability confirmedEurope PMC · Crossref · checked 14 Sept 2026
Even though mechanization has dramatically decreased labor requirements, vineyard management costs are still affected by selective operations such as winter pruning. Robotic solutions are becoming more common in agriculture, however, few studies have focused on grapevines. This work aims at fine-tuning and testing two different deep neural networks for: (i) detecting pruning regions (PRs), and (ii) performing organ segmentation of spur-pruned dormant grapevines. The Faster R-CNN network was fine-tuned using 1215 RGB images collected in different vineyards and annotated through bounding boxes. The network was tested on 232 RGB images, PRs were categorized by wood type (W), orientation (Or) and visibility (V), and performance metrics were calculated. PR detection was dramatically affected by visibility. Highest detection was associated with visible intermediate complex spurs in Merlot (0.97), while most represented coplanar simple spurs allowed a 74% detection rate. The Mask R-CNN network was trained for grapevine organs (GOs) segmentation by using 119 RGB images annotated by distinguishing 5 classes (cordon, arm, spur, cane and node). The network was tested on 60 RGB images of light pruned (LP), shoot-thinned (ST) and unthinned control (C) grapevines. Nodes were the best segmented GOs (0.88) and general recall was higher for ST (0.85) compared to C (0.80) confirming the role of canopy management in improving performances of hi-tech solutions based on artificial intelligence. The two fine-tuned and tested networks are part of a larger control framework that is under development for autonomous winter pruning of grapevines. Supplementary information The online version contains supplementary material available at 10.1007/s11119-023-10006-y.
Why it matches plant phenotyping methods深層学習によるブドウ樹の剪定領域検出と器官セグメンテーションを開発・評価しており、植物器官状態の画像ベース取得が中心である。
abstractThis work aims at fine-tuning and testing two different deep neural networks for: (i) detecting pruning regions (PRs), and (ii) performing organ segmentation of spur-pruned dormant grapevines.
Reproduction assets foundThe paper's annotated grapevine organ segmentation dataset (images with polygon/bounding-box annotations for cordon, arm, spur, cane, node) is publicly deposited on Zenodo. The pruning region detection dataset is not public and must be requested from the corresponding author. No author analysis code is available (code:Dataset · publicd Research, PRIN 20172HHNK5 Project.
Data availability
The pruning region detection dataset generated and/or analyzed during the presented study is currently not publicly available, but can be requested from the corresponding author on reasonable request. The annotated segmentation dataset is published on the zenodo platform at https://zenodo.org/record/5501784 .
Code availability
Not applicable.
Declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Ethical approval
The authors comply with the Journal’s Ethics guidelines confirming to respect third parties rights such as copyright and/or moral rights.
Consent to participate
NoOpen asset ↗zenodo · 5501784lines:583-615Plant phenotyping relevance match · UnverifiedOpenAlex · checked 13 Sept 2026
GrapevineAerial / UAVField / plotPhotogrammetry / SfM / MVSLiDAR / point cloudWhole plant / canopy / plot / field
The potential of precision viticulture has been highlighted since the first studies performed in the context of viticulture, but especially in the last decade there have been excellent results have been achieved in terms of innovation and simple application. The deployment of new sensors for vineyard monitoring is set to increase in the coming years, enabling large amounts of information to be obtained. However, the large number of sensors developed and the great amount of data that can be collected are not always easy to manage, as it requires cross-sectoral expertise. The preliminary section of the review presents the scenario of precision viticulture, highlighting its potential and possible applications. This review illustrates the types of sensors and their operating principles. Remote platforms such as satellites, unmanned aerial vehicles (UAV) and proximal platforms are also presented. Some supervised and unsupervised algorithms used for object-based image segmentation and classification (OBIA) are then discussed, as well as a description of some vegetation indices (VI) used in viticulture. Photogrammetric algorithms for 3D canopy modelling using dense point clouds are illustrated. Finally, some machine learning and deep learning algorithms are illustrated for processing and interpreting big data to understand the vineyard agronomic and physiological status. This review shows that to perform accurate vineyard surveys and evaluations, it is important to select the appropriate sensor or platform, so the algorithms used in post-processing depend on the type of data collected. Several aspects discussed are fundamental to the understanding and implementation of vineyard variability monitoring techniques. However, it is evident that in the future, artificial intelligence and new equipment will become increasingly relevant for the detection and management of spatial variability through an autonomous approach.
Why it matches plant phenotyping methodsブドウ園のセンサー、リモートセンシング、画像解析、3Dキャノピーモデリング、機械学習を植物の生理・農学的状態の評価に用いる方法を包括的にレビューしており、フェノタイピング手法が中心である。
abstractThis review illustrates the types of sensors and their operating principles.
GrapevineFruitGrowth / time-series analysisGrowth / development / phenology
Fruit growth and development consist of a continuous succession of physical, biochemical, and physiological changes driven by a genetic program that dynamically responds to environmental cues. Establishing recognizable stages over the whole fruit lifetime represents a fundamental requirement for research and fruit crop cultivation. This is especially relevant in perennial crops like grapevine ( Vitis vinifera L.) to scale the development of its fruit across genotypes and growing conditions. In this work, molecular-based information from several grape berry transcriptomic datasets was exploited to build a molecular phenology scale (MPhS) and to map the ontogenic development of the fruit. The proposed statistical pipeline consisted of an unsupervised learning procedure yielding an innovative combination of semiparametric, smoothing, and dimensionality reduction tools. The transcriptomic distance between fruit samples was precisely quantified by means of the MPhS that also enabled to highlight the complex dynamics of the transcriptional program over berry development through the calculation of the rate of variation of MPhS stages by time. The MPhS allowed the alignment of time-series fruit samples proving to be a complementary method for mapping the progression of grape berry development with higher detail compared to classic time- or phenotype-based approaches.
Why it matches plant phenotyping methodsブドウ果実の発達状態を転写データから推定・段階化する統計的パイプライン自体を開発し、従来の表現型ベース手法と比較しているため、分子データの単なる生物学的利用ではなく植物フェノタイピング手法が中心です。
abstractThe proposed statistical pipeline consisted of an unsupervised learning procedure yielding an innovative combination of semiparametric, smoothing, and dimensionality reduction tools.
Introduction Grapevine leafroll-associated viruses (GLRaVs) and grapevine red blotch virus (GRBV) cause substantial economic losses and concern to North America’s grape and wine industries. Fast and accurate identification of these two groups of viruses is key to informing disease management strategies and limiting their spread by insect vectors in the vineyard. Hyperspectral imaging offers new opportunities for virus disease scouting. Methods Here we used two machine learning methods, i.e., Random Forest (RF) and 3D-Convolutional Neural Network (CNN), to identify and distinguish leaves from red blotch-infected vines, leafroll-infected vines, and vines co-infected with both viruses using spatiospectral information in the visible domain (510-710nm). We captured hyperspectral images of about 500 leaves from 250 vines at two sampling times during the growing season (a pre-symptomatic stage at veraison and a symptomatic stage at mid-ripening). Concurrently, viral infections were determined in leaf petioles by polymerase chain reaction (PCR) based assays using virus-specific primers and by visual assessment of disease symptoms. Results When binarily classifying infected vs. non-infected leaves, the CNN model reaches an overall maximum accuracy of 87% versus 82.8% for the RF model. Using the symptomatic dataset lowers the rate of false negatives. Based on a multiclass categorization of leaves, the CNN and RF models had a maximum accuracy of 77.7% and 76.9% (averaged across both healthy and infected leaf categories). Both CNN and RF outperformed visual assessment of symptoms by experts when using RGB segmented images. Interpretation of the RF data showed that the most important wavelengths were in the green, orange, and red subregions. Discussion While differentiation between plants co-infected with GLRaVs and GRBV proved to be relatively challenging, both models showed promising accuracies across infection categories.
Why it matches plant phenotyping methodsハイパースペクトル画像と機械学習により、植物葉のウイルス感染状態を症状発現前後に推定する手法を開発・評価しており、フェノタイピング手法が中心である。
abstractHyperspectral imaging offers new opportunities for virus disease scouting.
Grapevine virus-associated disease such as grapevine leafroll disease (GLD) affects grapevine health worldwide. Current diagnostic methods are either highly costly (laboratory-based diagnostics) or can be unreliable (visual assessments). Hyperspectral sensing technology is capable of measuring leaf reflectance spectra that can be used for the non-destructive and rapid detection of plant diseases. The present study used proximal hyperspectral sensing to detect virus infection in Pinot Noir (red-berried winegrape cultivar) and Chardonnay (white-berried winegrape cultivar) grapevines. Spectral data were collected throughout the grape growing season at six timepoints per cultivar. Partial least squares-discriminant analysis (PLS-DA) was used to build a predictive model of the presence or absence of GLD. The temporal change of canopy spectral reflectance showed that the harvest timepoint had the best prediction result. Prediction accuracies of 96% and 76% were achieved for Pinot Noir and Chardonnay, respectively. Our results provide valuable information on the optimal time for GLD detection. This hyperspectral method can also be deployed on mobile platforms including ground-based vehicles and unmanned aerial vehicles (UAV) for large-scale disease surveillance in vineyards.
Why it matches plant phenotyping methodsブドウ樹のウイルス病状態を近接ハイパースペクトルセンシングで推定し、PLS-DAモデルの予測精度と検出時期を評価しており、病害フェノタイピング手法が中心である。
abstractHyperspectral sensing technology is capable of measuring leaf reflectance spectra that can be used for the non-destructive and rapid detection of plant diseases.
Accurate fruit counting helps grape wine industry make better logistics and decisions before harvest, and therefore produce higher quality wine. In view of poor real-time performance of the existing fruit tracking and counting methods, and a lack of effective counting methods for cluster-like fruits due to their huge shape variabilities. In this study, an end-to-end lightweight counting pipeline is developed to automate the processing of video data for real-time tracking and counting of grape clusters in field conditions. First, based on channel pruning algorithm, a more lightweight YOLOv5s cluster detection model is obtained, where number of model parameters, model size and floating-point operations (FLOPs) are reduced by 79 %, 76 %, and 58 %, respectively, and the pruned model size is only 3.4 MB. Secondly, the soft non-maximum suppression is introduced in prediction stage to improve detection performance for clusters with overlapping grapes. Test results show that mAP reaches 82.3 % and average inference time is 6.1 ms per image, which effectively reduces model parameters and complexity while ensuring detection accuracy. Finally, online multiple object tracking of clusters is implemented by integrating the detection results and SORT algorithm, where two counting modes are set by introducing counting lines. Test results on 8 videos indicated that the average counting accuracy of the proposed method reached 84.9 %, correlation coefficient with manual counting reached 0.9905, and speed of video processing reached up to 50.4 frames per second (FPS), meeting field real-time requirements. This study provides a timely technical reference for the development of orchard robots to achieve real-time automated yield estimation and accurate crop management decisions.
Why it matches plant phenotyping methodsブドウ房という植物器官の検出・追跡・計数による収量推定手法を開発し、精度・速度を検証しており、表現型取得が研究の中心である。
abstractan end-to-end lightweight counting pipeline is developed to automate the processing of video data for real-time tracking and counting of grape clusters in field conditions
Associated with climate change, the frequency, duration, and intensity of heatwaves are increasing in most of the key wine regions worldwide. Depending on timing, intensity, and duration, heatwaves can impact grapevine yield and berry composition, with implications for wine quality. To overcome these negative effects, two types of mitigation practices have been proposed (i) to enhance transpiration and (ii) to reduce the radiation load on the canopy. Here we use a biophysical model to quantify the impact of these practices on canopy gas exchange, vine water status, and leaf temperature (Tₗ). Model validation was performed in a commercial vineyard. Modelled Tₗ from 14 to 43 °C, and transpiration, from 0.1 to 5.4 mm d⁻¹, aligned around the identity line with measurements in field-grown vines; the RMSD was 2.6 ºC for temperature and 0.96 mm day⁻¹ for transpiration. Trellis system and row orientation modulate Tₗ. A sprawling single wire trellis with an EW orientation maintained the canopy around 1ºC cooler than a Vertical Shoot Positioned canopy with NS for the same range of total fraction of soil available water (TFAW). Although irrigation before a heatwave is a recommended practice, maximum transpiration can be sustained even when TFAW is reduced, limiting the heat dampening effect of irrigation. Alternatively, canopy cooling can be achieved through Kaolin application, the installation of shade cloth placement, or canopy trimming. Shade cloth produced a greater cooling than Kaolin in all the simulated scenarios; however, Tₗ differences between them varied. Trimming reduced Tₗ from 2 ºC to almost 8 ºC compared to its non-trimmed counterpart. Our analysis presents new insights to design heat wave mitigation strategies and supports agronomically meaningful definitions of heat waves that include not only temperature, but also wind, VPD, and radiation load as these factors influence crop physiology under heat stress.
Why it matches plant phenotyping methodsブドウの葉温・蒸散・水状態を推定する生物物理モデルを開発的に適用し、圃場測定で検証しているため、植物表現型の取得・推定が中心的です。
abstractHere we use a biophysical model to quantify the impact of these practices on canopy gas exchange, vine water status, and leaf temperature (Tₗ).