The inflorescence is a key yield-determining organ, yet its complex morphology makes manual pixel-level annotation time-consuming, leading to a scarcity of high-quality segmentation datasets. To address this bottleneck, we propose ZCAT (Zero-shot Cross-crop Annotation Transfer), a novel paradigm for zero-annotation cross-crop pseudo-mask screening. ZCAT completely eliminates pixel-level manual annotation of the target crop, requiring only holistic quality assessment of model-generated pseudo-masks (5-10 s per image). Specifically, we train a SegFormer model on public rice panicle datasets (CVRP and RiceSEG) and transfer it across crops to the wheat spike segmentation task. The key innovation is the introduction of a human-defined quality function Q, which circumvents the fundamental challenge in self-learning algorithms: the inability of computers to autonomously distinguish good masks from bad ones. Through iterative human-in-the-loop pseudo-label screening with a curriculum learning strategy, each round adds only a few high-quality pseudo-masks to the training set, continuously improving model performance. After four iterations, ZCAT produces pseudo-masks with an average Spike IoU of 0.7003, evaluated against the GWFSS manual annotations as ground truth. Moreover, the pseudo-mask dataset exhibited higher benchmark performance than the GWFSS manual annotations (Spike IoU 0.7612 vs. 0.7027; mIoU 0.8627 vs. 0.8247), suggesting stronger self-consistency. A generalization test on a strictly held-out set of 100 manually annotated wheat spike images showed that the model trained on ZCAT-generated pseudo-masks achieved marginally better performance than that trained on the GWFSS manual annotations (Spike IoU: 0.5112 vs. 0.4927; mIoU: 0.5627 vs. 0.5247). The time budget of the ZCAT pipeline was substantially lower than that of manual annotation. ZCAT opens a new pathway for rapid annotation of plant reproductive structures or other organs and significantly reduces data preparation costs in plant phenomics. The generated wheat spike pseudo-mask dataset and the mask quality screening tool (Mask Quality Screener) are open-sourced.
Why it matches plant phenotyping methods植物器官セグメンテーションのためのゼロショット転移、擬似マスク品質評価、反復学習パイプラインを開発・検証しており、表現型取得基盤が中心である。
abstractThe key innovation is the introduction of a human-defined quality function Q
Reproduction assets foundThe paper explicitly open-sources two paper-specific assets: the ZCAT-generated wheat spike pseudo-mask dataset and the Mask Quality Screener tool, both with public GitHub URLs in the Data availability statement.Dataset · publicThe wheat spike pseudo-mask dataset and Mask Quality Screener are available at https://github.com/zyxyes1/MaskQualityScreener and https://github.com/zyxyes1/Wheat-Spike-Semantic-Segmentation , respectively.Open asset ↗Wheat-Spike-Semantic-Segmentationlines:415-440Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
This study compiles and releases the first standardized rapeseed phenology observation dataset spanning forty-four years (1981-2024) over the core winter rapeseed production region of the Middle and Lower Yangtze River Plain in China. The data originate from systematic observations at 50 national-level agrometeorological stations across six provinces: Jiangsu, Zhejiang, Anhui, Jiangxi, Hubei, and Hunan. The dataset provides complete records of the specific dates for each phenology stage from sowing to maturity, including eight key phenology periods: Sowing (SO), Emergence (EM), Five-leaf (FV), Bud Formation (BF), Stem Elongation (SE), Flowering (FL), Green Ripening (GR), and Maturity (MA), along with the calculated durations of six distinct growth lengths. We implemented a multi-level quality control protocol encompassing internal logical checks, statistical outlier detection, climatological validation, time series homogenization, and expert arbitration. This protocol effectively constrained data uncertainty and corrected non-climatic discontinuities. Univariate linear regression was further employed to quantify the decadal change trends of each phenology period and growth length, supplemented by Kernel Density Estimation (KDE) to characterize their probability distribution features. The final dataset is presented as structured tables (in xlsx format) and high-resolution diagnostic plots (including trend and density plots), with a total volume of approximately 470 MB, systematically organized by province and station. This dataset fills a critical gap in long-term, standardized rapeseed phenology data for the region. The integrated analysis of phenology dates, growth stage durations, and their trends across the entire network provides an indispensable, high-quality empirical foundation. It is designed to support in-depth investigations into the nonlinear response mechanisms of overwintering crops to climate warming, improve crop model parameterization and validation, and inform regional adaptive management strategies.
Why it matches plant phenotyping methods44年間のナタネの生育段階日を標準化・品質管理して公開するデータセット研究であり、植物状態（フェノロジー）の測定データ整備が中心です。
abstractThis study compiles and releases the first standardized rapeseed phenology observation dataset spanning forty-four years (1981-2024)
Reproduction assets foundThe paper's rapeseed phenology dataset (1981–2024, 50 stations) is openly deposited in Science Data Bank under DOI 10.57760/sciencedb.34086, containing structured xlsx tables and diagnostic plots. No custom code was created per the authors.Dataset · publicThe dataset described in this work has been deposited in the Science Data Bank (ScienceDB) under
accession code https://doi.org/10.57760/sciencedb.34086 [27].Open asset ↗Science Data Bank · 10.57760/sciencedb.34086pdf-page:12 lines:1-68Code / dataset availability confirmedOpenAlex · Europe PMC · Crossref · checked 5 Sept 2026
Sweet potato virus disease (SPVD) is one of the most destructive diseases affecting sweet potato production worldwide, causing severe yield losses and posing a significant threat to food security. Vision-based intelligent diagnosis has emerged as a promising solution for large-scale SPVD monitoring due to its low cost and scalability. However, existing publicly available datasets for SPVD are extremely limited and typically focus on a single task, such as disease classification or lesion segmentation, under constrained imaging conditions. This lack of comprehensive, task-oriented datasets significantly restricts the development, evaluation, and fair comparison of advanced computer vision methods for SPVD analysis. In this study, we present SPVD-Field, a task-oriented multi-task visual dataset suite composed of two independently collected sub-datasets optimized for different computer vision tasks. Rather than constructing a single homogeneous dataset, SPVD-Field is deliberately organized into two complementary task-oriented sub-datasets: SPVD-DET, designed for disease detection with bounding-box annotations, and SPVD-SEG, designed for fine-grained lesion segmentation with pixel-level masks. The two sub-datasets were independently collected using different acquisition protocols optimized for their respective tasks, while sharing a unified semantic definition of SPVD symptoms, crop growth stages, and field environments. SPVD-Field captures substantial real-world variability in imaging scale, viewpoint, illumination, background complexity, and symptom manifestation, reflecting the inherent challenges of fieldbased disease diagnosis. We provide detailed documentation of data acquisition, annotation strategies, and quality control procedures, along with baseline benchmark results for both detection and segmentation tasks to demonstrate the usability and difficulty of the dataset. By offering a structured dataset suite rather than a single-task collection, SPVD-Field aims to support diverse research directions, including detection, segmentation, multi-task learning, and disease severity analysis, and to facilitate reproducible and comparable research in SPVD-related plant phenotyping.
Why it matches plant phenotyping methodsサツマイモの病徴を対象とする画像データセットで、検出・病斑セグメンテーション、データ取得・アノテーション・品質管理、ベンチマークを中心的に提供しており、植物病害状態の画像フェノタイピング手法・データ基盤に該当する。
abstractIn this study, we present SPVD-Field, a task-oriented multi-task visual dataset suite composed of two independently collected sub-datasets optimized for different computer vision tasks.
Reproduction assets foundThe paper's core asset is the SPVD-Field dataset (SPVD-DET detection images with bounding-box annotations and SPVD-SEG segmentation images with pixel-level masks), explicitly deposited in a public repository via the data availability statement with a DOI link.Dataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://dx.doi.org/10.21227/hq1q-jp43 .Open asset ↗10.21227/hq1q-jp43lines:664-703Code / dataset availability confirmedOpenAlex · Europe PMC · bioRxiv · Crossref · checked 15 Sept 2026
BarleyCommon beanCowpeaGrowth chamberMesh / voxelLiDAR / point cloudMultispectral / hyperspectralLeafWhole plant / canopy / plot / fieldAnnotation / quality control
Abstract High-throughput 3D multispectral plant phenotyping platforms generate large volumes of point cloud files, but trait extraction is typically performed by sensor-bundled software whose internal algorithms are not publicly documented, which limits reproducibility and integration into custom research pipelines. Here we present PhytoScan3D, an open-source Python pipeline that extracts morphological and spectral phenotypic traits, spanning plant height, 3D leaf area, digital biomass, convex hull volume, leaf inclination, canopy geometry, NDVI, hue, and vegetation indices, from both PLY and PCD point cloud files generated by Phenospex PlantEye F500 and F600 sensors, and is portable to point clouds from any acquisition platform. PhytoScan3D was validated against HortControl (PhenoSpex) ground-truth measurements on 936 barley ( Hordeum vulgare ) pot-date observations from the growth chamber trial (20 Norwegian cultivars, 12 scan dates, Septemenr 2025 to January 2026), achieving Pearson r = 0.913 to 0.999 and ratio approximately 1.000 for Plant Height Max, 3D Leaf Area, and NDVI Average. A vectorised mesh face filtering implementation achieved a 120x speed improvement, increasing valid 3D Leaf Area coverage from 0.6% to 100% of files. Cross-format validation on 223 PlantEye F600 PCD files from the ICRISAT LeasyScan platform (four legume species: mungbean, cowpea, lima bean, and common bean; 1,523 plant observations) yielded r = 0.884 against independent cuboid annotation heights. The systematic positive bias (mean +27.2 mm, ratio = 1.44) is attributable to PhytoScan3D computing height from raw point cloud Z-range while cuboid annotations are fitted to segmented plant points only, with the offset consistent across all four species (per-species r = 0.880 to 0.888). Cross-dataset processing of 1,180 PLY files from the Crops3D benchmark (8 species, 3 acquisition methods) confirmed zero extraction errors. PhytoScan3D is available at “github.com/kovimallik/phytoscan3d” under the MIT licence and processes 1,651 files across three independent datasets in under 12 minutes on GPU hardware. Highlights PhytoScan3D is the first open-source Python pipeline for batch extraction of phenotypic traits, including plant height, 3D leaf area, digital biomass, convex hull volume, leaf inclination, NDVI, and excess green index, from both PLY and PCD point cloud files generated by Phenospex PlantEye sensors. Primary validation against HortControl ground-truth measurements on 936 barley pot-date observations achieved Pearson r = 0.913-0.999 for Plant Height Max, 3D Leaf Area, and NDVI Average. A 120x computational speedup in mesh face filtering (vectorised NumPy vs. set-based loop) increased the coverage of valid 3D Leaf Area extraction from 0.6% to 100% of files. Cross-format validation on 223 PlantEye F600 PCD files from ICRISAT LeasyScan (four legume species, 1,523 plants) achieved r = 0.884 against independent cuboid annotation heights. The systematic +27.2 mm bias reflects a methodological difference (raw Z-range vs. soil-segmented annotations), is consistent and predictable across all four species (per-species r = 0.880-0.888), and is correctable by a single linear factor. Cross-dataset processing of 1,180 PLY files from the Crops3D benchmark (8 species, 3 acquisition methods) confirmed zero extraction errors. Significant scan-unit variation was detected for Plant Height Max (F = 5.71, p < 0.001, η 2 = 0.138) and Canopy Width X (F = 6.32, p < 0.001, η 2 = 0.150), demonstrating the biological utility of extracted traits.
Why it matches plant phenotyping methods植物の3D点群・マルチスペクトルデータから形態・スペクトル形質を抽出するオープンソース手法を開発し、複数データセットで技術検証・ベンチマークしているため、植物フェノタイピング手法が中心である。
abstractHere we present PhytoScan3D, an open-source Python pipeline that extracts morphological and spectral phenotypic traits
Reproduction assets foundThe paper's own analysis code (PhytoScan3D pipeline) is publicly released on GitHub under the MIT licence, and the two external 3D point cloud datasets used for validation (Crops3D and ICRISAT LeasyScan) are publicly available on figshare. The primary barley PLY dataset is not yet public (to be deposited in NVA upon).Code · publicditing, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
Data Availability
PhytoScan3D source code, documentation, and example datasets are available at
https://github.com/kovimallik/phytoscan3d under the MIT licence. The barley PLY dataset
will be deposited in the Norwegian Research Information Repository (NVA) upon acceptance.
The Crops3D benchmark dataset is publicly available at
https://doi.org/10.6084/m9.figshare.27313272 (Zhu et al. 2024). The ICRISAT LeasyScan
dataset is publicly available at https://doi.org/10Open asset ↗github.com/kovimallik/phytoscan3dpdf-raw-page:15 lines:1-36Dataset · publicData Availability
PhytoScan3D source code, documentation, and example datasets are available at
https://github.com/kovimallik/phytoscan3d under the MIT licence. The barley PLY dataset
will be deposited in the Norwegian Research Information Repository (NVA) upon acceptance.
The Crops3D benchmark dataset is publicly available at
https://doi.org/10.6084/m9.figshare.27313272 (Zhu et al. 2024). The ICRISAT LeasyScan
dataset is publicly available at https://doi.org/10.6084/m9.figshare.28270742 (Galba et al.
2025).
Acknowledgements
This work was supported by the PheNo, DLT-Farming and Soil2Milk from Research Council
of Norway and TWIN-NUE from Norwegian University of Life Sciences (NMBU). The
authoOpen asset ↗figshare · 10.6084/m9.figshare.27313272pdf-raw-page:15 lines:1-36Dataset · publicimallik/phytoscan3d under the MIT licence. The barley PLY dataset
will be deposited in the Norwegian Research Information Repository (NVA) upon acceptance.
The Crops3D benchmark dataset is publicly available at
https://doi.org/10.6084/m9.figshare.27313272 (Zhu et al. 2024). The ICRISAT LeasyScan
dataset is publicly available at https://doi.org/10.6084/m9.figshare.28270742 (Galba et al.
2025).
Acknowledgements
This work was supported by the PheNo, DLT-Farming and Soil2Milk from Research Council
of Norway and TWIN-NUE from Norwegian University of Life Sciences (NMBU). The
authors thank Sara Catarina Costa Laranjeira, Min Lin and other NMBU growth facility staff
for plant care and scanning operOpen asset ↗figshare · 10.6084/m9.figshare.28270742pdf-raw-page:15 lines:1-36Code / dataset availability confirmedCrossref · OpenAlex · Europe PMC · checked 5 Sept 2026
Attempts to deploy computer vision in agricultural tasks often suffer from a shortage of annotated data. One strategy to alleviate the impact of limited data is Self-Supervised Learning (SSL), which involves pre-training a model on a pretext task that utilizes automatically generated annotations. The primary objective of this study is to leverage a multi-camera view dataset of cotton boll images for contrastive learning in order to enable phenotyping tasks with minimal data annotation. This dataset was collected in the field using six camera views. The efficacy of two contrastive learning frameworks (SimCLR and MoCo) in producing representations when positive examples originate from different cameras was investigated, and a comprehensive study of how the camera positions affect performance was conducted. After self-supervised pre-training, linear evaluation and semi-supervised learning experiments were performed on boll detection and plot status downstream tasks. In general, using multiple camera views with SimCLR and MoCo improves cotton boll detection mean average precision by 14% compared to vanilla SimCLR and MoCo. Through careful investigation using synthetic data, it was determined that relative camera poses with an intermediate amount of overlap seem more likely to perform well. Neither MoCo nor SimCLR was consistently superior to the other in this context. The representations embed meaningful features about the cotton plants, such as overall boll density, but also less meaningful ones, such as lighting variations. This technique could potentially accelerate the development of phenotyping algorithms based on data collected from field robots. • A contrastive learning method based on comparing multi-camera views was developed. • The method was tested with images of cotton bolls from a ground robot. • The method outperformed baseline contrastive learning approaches.
Why it matches plant phenotyping methodsマルチカメラ画像とコントラスト学習による植物表現学習・フェノタイピング手法の開発と評価が中心であり、綿花のボール検出性能を検証している。
abstractThe primary objective of this study is to leverage a multi-camera view dataset of cotton boll images for contrastive learning in order to enable phenotyping tasks with minimal data annotation.
Reproduction assets foundThe paper's data availability statement points to a public GitHub repository containing the authors' code to reproduce the multi-camera contrastive learning phenotyping experiments. A processed-data Zenodo deposit (10.5281/zenodo.18164649) is also mentioned, but its URL is not among the allowed URLs, so only the code资产Code · publicThe code required to reproduce the above findings are available to download from https://github.com/UGA-BSAIL/self-supervised-learning .Open asset ↗UGA-BSAIL/self-supervised-learninglines:200-224Code / dataset availability confirmedEurope PMC · bioRxiv · checked 15 Sept 2026
Accurate and reproducible assessment of foliar disease severity is essential for evaluating the performance of heterogeneous plant communities and understanding host-pathogen interactions. However, traditional visual scoring methods remain subjective, with limited precision, and difficult to scale in large phenotyping experiments. Here, we present a semi-automated image analysis workflow designed to quantify multiple foliar disease symptoms simultaneously on wheat flag leaves sampled from varietal mixtures. The workflow combines three methodological components: (i) a standardized protocol for leaf sampling and imaging, (ii) supervised machine learning segmentation using Random Forest implemented in Ilastik to classify multiple symptoms (powdery mildew and yellow rust), and (iii) a graphical user interface facilitating pipeline deployment by non-specialist operators. To evaluate the influence of image representation on classification performance, four color spaces (RGB, HSV, HLS, LAB) were systematically compared. The approach was validated using images of durum wheat flag leaves collected from a field experiment assessing eight-way varietal mixtures under natural fungal pressure. Cross-validation against manually annotated images demonstrated high segmentation accuracy across all symptom. Comparison among color spaces revealed only minor differences in performance. Overall, this workflow offers a cost-effective, annotation-efficient and reproducible alternative to deep learning approaches, leveraging open-source and actively maintained tools while requiring limited training data and enabling objective, reproducible and scalable disease phenotyping.
Why it matches plant phenotyping methods葉の病害症状を画像解析で定量化するワークフローを開発し、色空間比較と手動アノテーションによる検証を行っており、植物表現型取得法が中心である。
abstractwe present a semi-automated image analysis workflow designed to quantify multiple foliar disease symptoms simultaneously
Reproduction assets foundThe paper's authors explicitly state that all code implementing the leaf disease quantification workflow (SegLeaf, including the graphical interface and documentation) is hosted in a public GitHub repository. No separate public phenotype dataset or trained model checkpoint is described in the supplied blocks.Code · publicted by the Agence Nationale de
la Recherche (ANR) (project SCOOP, grant no. ANR-19-CE32-0011; and project MOBIDIV,
grant no. ANR-20-PCPA-0006).
Code and Data Availability
The method and associated scripts developed in this work are freely available to the re-
search community. All code is hosted in a public GitHub repository at https://github.com/titouanlegourrierec/SegLeaf, which includes the full implementation of the method includ-
ing the graphical interface and documentation to guide users through the analysis pipeline.
15
.
CC-BY 4.0 International license
made available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display tOpen asset ↗titouanlegourrierec/SegLeafpdf-raw-page:15 lines:1-39Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
This paper addresses the problem of automated segmentation of plant green biomass in field crop images aimed at improving the accuracy of crop and weed identification. To construct a training dataset for neural network models, an automatic annotation algorithm is proposed, enabling the generation of polygonal object masks without human intervention. The method is based on adaptive analysis of color characteristics of plant fragments with iterative narrowing of the hue range in the HSV color space, combined with an integral quality metric that accounts for the dynamics of contour area and shape. The proposed method achieved an IoU of 93.22% and a DSC of 96.30%, demonstrating a high level of agreement between automatic and manual annotations. The generated masks are used to train segmentation models of the YOLO11-seg family. Models of different scales (n, s, m, l, x) were trained and evaluated using standard metrics, including Intersection over Union (IoU), mAP@0.5, mAP@0.5–0.95, F1-score, and Precision–Recall (PR) curves. Experimental results demonstrate that models trained on automatically generated annotations achieve stable segmentation performance of plant green biomass. The best results were obtained with the YOLO11m-seg model, achieving an F1-score of 0. 772. The results confirm the effectiveness of the proposed approach and demonstrate acceptable segmentation quality, supported by both quantitative metrics and visual analysis. The developed automatic annotation algorithm can be used to expand training datasets in computer vision tasks for agricultural applications.
Why it matches plant phenotyping methods植物の緑色バイオマスを画像から自動抽出するポリゴン注釈法を開発し、手動注釈との一致度で検証しているため、植物表現型取得・抽出法が中心である。
abstractan automatic annotation algorithm is proposed, enabling the generation of polygonal object masks without human intervention
Reproduction assets foundThe authors publicly released the paper-specific generated dataset of polygonal segmentation annotations (masks and supporting materials) on Hugging Face. CVAT is only a generic annotation tool, and no author analysis code or trained model checkpoints are explicitly deposited.Dataset · publicThe generated dataset with polygonal segmentation annotations of crop and weed plants, produced using the proposed algorithm and based on the LincolnBeet Dataset, is publicly available on Hugging Face at: https://huggingface.co/datasets/ivliev123/polygonal_marking_plant_objectsOpen asset ↗Hugging Face · ivliev123/polygonal_marking_plant_objectshtml-lines:438-462Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
Accurate and efficient grain quality assessment is critical for making informed decisions throughout the grain value chain. Early detection of disease enables actions to mitigate spread and further damage, and optimal batch mixing to fulfill specified quality requirements allows for maximizing value and minimizing scrapping. Vision based machine learning and deep learning approaches are gaining attention in the agricultural sector and are useful for the development of automated grain quality assessment. These techniques can reduce the current manual inspection load and are key for objective and precise analysis. Yet, the majority of prior studies are constrained to small or controlled and curated datasets. Practical challenges associated with real-world deployment and reliability are rarely addressed. That is the focus of this work. We present and demonstrate a structured approach for investigating convolutional neural networks (CNNs) and key factors influencing performance for wheat kernel classification. The objective is to determine a CNN model that ensures high and robust classification accuracy, while elucidating and explaining how different image dataset characteristics and training parameters affect performance and reliability. We use a commercial mirror-based imaging system that captures over 90% of each kernel's surface and contrast and compare model architectures, robustness, the effect on pre-processing and image resolution. Our results show similar and high overall performance for ResNet50V2 and EfficientNetV2B0 ([Formula: see text]% accuracy), but per-class analysis indicate that the smaller classes suffer from lack of representative examples, and that most classes benefit from pre-processing including downsampling whereas others benefit from higher resolution. Interactive visualizations reveal that another contributing factor is dubious annotation and multi-class belongingness. Thus, our step-by-step analysis of CNN performance underscores the need for representative data, proper pre-processing, and class-aware evaluation to ensure trustworthy deployment in wheat grain quality assessment.
Why it matches plant phenotyping methods小麦粒画像から品質・病害クラスを推定するCNN画像解析手法の性能、頑健性、前処理、解像度、データ特性を体系的に評価しており、フェノタイピング手法が中心的である。
abstractWe present and demonstrate a structured approach for investigating convolutional neural networks (CNNs) and key factors influencing performance for wheat kernel classification.
Reproduction assets foundThe paper's wheat grain image dataset has a publicly available subset deposited on Zenodo (DOI 10.5281/zenodo.17397123), explicitly stated in the Data Availability statement. The full dataset is proprietary; code is only available upon request, so no qualifying code asset.Dataset · publicA publicly available subset of the segmented wheat grain images used in this study has been deposited in Zenodo to support transparency and reproducibility. The dataset includes representative samples per class collected from instrument and can be accessed at https://doi.org/10.5281/zenodo.17397123.Open asset ↗Zenodo · 10.5281/zenodo.17397123html-lines:337-368Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
3D crop phenotyping technology provides critical support for screening morphology-related plant genes and identification of germplasm resource. Organ segmentation or recognition is the first key step in 3D crop phenotyping, where inductive deep learning currently dominates as the mainstream methodology. However, the high requirement for data annotation in inductive learning paradigm has transformed the manual data labeling into a labor-intensive task, thereby in turn restricting the progress of inductive learning. This problem has inspired us to leverage Graph Neural Networks (GNNs) as the transductive learning tool to directly segment organs on sparsely annotated crop point clouds. We propose a Dual-branch Graph Convolutional Network (DBGCN) that only requires sparse labels to perform organ instance inference directly on plant point clouds that have featureless point features. Different from existing graph-based networks, DBGCN not only carries out the static-feature-space graph convolutions that are good at mining and aggregating on local information on the point cloud, but also incorporates dynamic graph convolutions that captures the potential changes of the graph manifold in deep feature space. Extensive experiments prove that the fusion of two types of graph feature convolutions brings a high node (point) classification accuracy, outperforming mainstream GNNs and even several popular inductive deep architectures. On the PlantNet sub-dataset, DBGCN achieves an mAcc (mean accuracy of node classification) of 93.00% under 1.95% manual annotation ratio. On the Soybean-MVS sub-dataset, DBGCN achieves an mAcc of 91.05% under 4.88% manual annotation ratio. Furthermore, our DBGCN not only works well on crop 3D data but can also serve other applications such as the segmentation of point cloud data for large-scale street view. Our dataset and code can be found at https://github.com/chinazhouzhaoyi/DBGCN/tree/master/.
Why it matches plant phenotyping methods3D植物点群から器官を分割・推論する深層学習手法を開発し、植物フェノタイピングデータ上で精度検証しているため、表現型取得・抽出法が中心である。
abstractWe propose a Dual-branch Graph Convolutional Network (DBGCN) that only requires sparse labels to perform organ instance inference directly on plant point clouds
Reproduction assets foundThe authors explicitly state that their dataset (plant point clouds) and DBGCN code are publicly available on GitHub.Code · publicOur data and code are available at: https://github.com/chinazhouzhaoyi/DBGCN/tree/master/.Open asset ↗chinazhouzhaoyi/DBGCNhtml-lines:414-455Dataset · publicOur dataset and code can be found at https://github.com/chinazhouzhaoyi/DBGCN/tree/master/Open asset ↗chinazhouzhaoyi/DBGCNhtml-lines:88-94Code / dataset availability confirmedCrossref · Europe PMC · OpenAlex · checked 5 Sept 2026
Quantitative characterization of complete canopy architecture is essential for accurate evaluation of crop photosynthesis and yield potential, thereby supporting crop ideotype design. Although various sensing technologies enable three-dimensional (3D) reconstruction of individual plants and canopies, they often fail to describe canopy architecture accurately because of severe occlusion in dense populations. To address this limitation, we developed an effective framework for the 3D reconstruction of complex and dynamic population-scale canopy architecture in rapeseed using unmanned aerial vehicle multi-view imagery combined with a novel point cloud completion model. A complete point cloud generation pipeline was first established to enable automated training data annotation, allowing discrimination between surface points and occluded points within the canopy. The proposed crop population point cloud completion network (CP-PCN) integrates a multi-resolution dynamic graph convolutional encoder, a point pyramid decoder, a dynamic graph convolutional feature extractor, and a generative adversarial network-based loss function to predict occluded canopy points. CP-PCN achieved chamfer distance values of 3.35 to 4.51 cm across four growth stages, outperforming the state-of-the-art transformer-based method PoinTr. Ablation analyses confirmed that each of the four modules contributes to overall model accuracy. In addition, validation experiments showed that the improved architectural completeness achieved by CP-PCN resulted in more accurate yield estimation compared with incomplete and PoinTr-completed point clouds. CP-PCN also demonstrated strong cross-crop generalizability by successfully reconstructing mature rice canopies. Overall, this framework provides a scalable approach for quantitative analysis of complex canopy architectures in field-grown crops.
Why it matches plant phenotyping methodsUAVマルチビュー画像から遮蔽点を補完し、作物群落の3Dキャノピー構造を再構成する手法を開発・検証しており、植物表現型取得が研究の中心です。
abstractwe developed an effective framework for the 3D reconstruction of complex and dynamic population-scale canopy architecture in rapeseed using unmanned aerial vehicle multi-view imagery combined with a novel point cloud completion model
Reproduction assets foundThe paper's Data and code availability statement explicitly deposits all source code and test data for the CP-PCN phenotyping pipeline on a public GitHub repository, matching an allowed URL.Code · publicAll source code and test data used in this study are publicly available on GitHub ( https://github.com/Ziyue-Guo/CP-PCN.git ).Open asset ↗https://github.com/Ziyue-Guo/CP-PCN.git · CP-PCNlines:133-158Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
Field / plotAnnotation / quality controlVisualization / data management
BACKGROUND: Accurate acquisition of phenotypic data is critical for cataloguing and utilising genetic variation in cultivated crops, landraces, and their wild relatives. The collection of phenotypic data using handwritten notes often introduces errors which can and should be avoided. Electronic data collection is crucial for ensuring error prevention and data standardisation and thus ensuring high-quality, reliable data. IMPLEMENTATION: This paper describes the development of GridScore NEXT, a new plant phenotyping application that significantly advances the state of the art for collecting field trial data in plant genetics, pre-breeding and crop improvement research. Building on its predecessor, GridScore, the development of GridScore NEXT was driven by real life, in the field interactions with expert user groups across a number of crops. This iterative design methodology allowed the development and testing of new features. Collaborators from the 'Biodiversity for Opportunities, Livelihoods and Development' (BOLD) project, focusing on crops including rice, grasspea, and alfalfa, along with barley, potato, vegetable and blueberry teams, provided invaluable insights through training sessions and interviews and in the field use of the application. RESULTS: Key improvements to GridScore NEXT include enhanced data collection tools, supporting individual plant phenotyping within plots and enabling new data types such as GPS coordinates and image traits. GridScore NEXT provides customisable user defined validation rules to help prevent errors and incorporates barcode scanning for accurate, efficient data capture. The application offers an increased toolbox of data visualizations over its predecessor including heatmaps and statistical box plots, which aid in identifying potential data issues and understanding trial performance in the field. GridScore NEXT is cross-platform and can operate without an internet connection, making it ideal for field use in remote areas. Its adoption has led to standardisation of methods, significant error reduction, and the timely sharing of data, enabling quicker decision-making in pre-breeding and characterisation experiments. GridScore NEXT is available under an open-source (Apache 2.0) licence and freely available to all with no restrictions. It offers self-hosting options for enhanced data security and privacy. GridScore NEXT shows broad applicability across a diverse range of not only plant phenotyping experiments, but any experiment that requires the collection of accurate data.
Why it matches plant phenotyping methods植物表現型データ収集アプリケーションの開発と検証が論文の中心であり、個体表現型や画像形質を含む圃場データ取得を支援するため、対象範囲に含める。
abstractThis paper describes the development of GridScore NEXT, a new plant phenotyping application that significantly advances the state of the art for collecting field trial data in plant genetics, pre-breeding and crop improvement research.
Reproduction assets foundThe paper describes GridScore NEXT and its use in BOLD/CPC phenotyping. Authors' public code (GitHub, Zenodo) and public phenotype datasets (BOLD alfalfa, grasspea, rice; CPC characterisation data) are available; blueberry and UKVGB data are request-only.Dataset · publicDatasets used in this study were part of the BOLD project (alfalfa, grasspea and rice) which are available from https://germinate.hutton.ac.uk/cwr/alfalfa/, https://germinate.hutton.ac.uk/cwr/grasspea and https://germinate.hutton.ac.uk/cwr/rice/.Open asset ↗html-lines:528-593Dataset · publicDatasets used in this study were part of the BOLD project (alfalfa, grasspea and rice) which are available from https://germinate.hutton.ac.uk/cwr/alfalfa/, https://germinate.hutton.ac.uk/cwr/grasspea and https://germinate.hutton.ac.uk/cwr/rice/.Open asset ↗html-lines:528-593Dataset · publicDatasets used in this study were part of the BOLD project (alfalfa, grasspea and rice) which are available from https://germinate.hutton.ac.uk/cwr/alfalfa/, https://germinate.hutton.ac.uk/cwr/grasspea and https://germinate.hutton.ac.uk/cwr/rice/.Open asset ↗html-lines:528-593Dataset · publicThe CPC datasets used are characterisation datasets which are available from https://germinate.hutton.ac.uk/cpc.Open asset ↗html-lines:528-593Code / dataset availability confirmedEurope PMC · OpenAlex · Crossref · checked 5 Sept 2026
Background Rice plant architecture underpins yield and grain quality, yet two obstacles impede accurate field characterization in dense paddies. First, single-plant reconstruction is constrained by severe inter-plant occlusion, cluttered backgrounds, and limited viewpoints. These factors obscure culms, leaves, basal tillers, and the true physical scale of the plant. Active ranging devices are cumbersome in outdoor plots and can lose accuracy, whereas conventional passive photogrammetry performs poorly under such conditions. Second, delineating panicles within a 3D rice model is intrinsically difficult. Panicles are slender, highly branched, and visually similar to surrounding foliage, often interwoven and partially hidden. These factors result in fragmented boundaries and missing details. Direct point-cloud segmentation struggles with such discontinuous geometry and requires costly 3D annotation, whereas generic image segmentation models trained on natural scenes transfer poorly to paddy imagery. These challenges motivate a field-ready workflow that both reconstructs whole plants at high resolution in dense plantings and reliably segments panicles to enable trait extraction. Results A low-cost, in-field, multi-view pipeline for whole-plant three-dimensional reconstruction, termed One Stop 3D Target Reconstruction And segmentation (OSTRA), operates on color images with a reference-board setup. The pipeline builds detailed three-dimensional models of individual rice plants and automatically segments key organs (in this case, panicles), despite dense surrounding vegetation. When applied to 231 diverse rice landraces grown in a crowded field setting, the method produced high-fidelity plant models with clearly delineated panicle structures. From these reconstructions, three architectural traits were derived: plant height, leaf area, and panicle length. Genome-wide association analysis of the measured traits identified strong genotype-phenotype associations tagging known candidate genes. Natural variants at D2 and RFL/APO2 were associated with plant height variation, variants at FLW7 were linked to differences in leaf area, and allelic variation at AAI1 corresponded to panicle length variation. These loci are established regulators of plant growth and morphology, indicating that this three-dimensional phenotyping pipeline attains accuracy sufficient to rediscover meaningful genetic signals. Conclusions This study provides a practical tool for precise rice phenotyping even under dense field planting conditions, overcoming occlusion and structural complexity. By enabling non-destructive, field-based measurement of complete plant architecture and linking these phenotypes to specific genes, the pipeline bridges field phenomics and genomics. The integrated reconstruction and analysis framework advances the study of rice architecture and offers a general route to connect complex traits with their genetic determinants.
Why it matches plant phenotyping methods密植圃場でのイネ全体3D再構築、器官分割、形質抽出を中核とする画像ベース表現型解析手法の開発・実証であり、明確に収載対象。
abstractA low-cost, in-field, multi-view pipeline for whole-plant three-dimensional reconstruction, termed One Stop 3D Target Reconstruction And segmentation (OSTRA), operates on color images with a reference-board setup.
Reproduction assets foundThe paper explicitly states that the 3D rice plant models (231 landraces) are deposited on Zenodo and the OSTRA source code is publicly available on GitHub. Both are paper-specific, public, and actionable.Code · publicThe source code of OSTRA is available on GitHub at [http://github.com/ganlab/ostra] (http:/github.com/ganlab/ostra).Open asset ↗github · ganlab/ostralines:217-246Code / dataset availability confirmedarXiv · OpenAlex · checked 13 Sept 2026
Automating tasks in orchards is challenging because of the large amount of variation in the environment and occlusions. One of the challenges is apple pose estimation, where key points, such as the calyx, are often occluded. Recently developed pose estimation methods no longer rely on these key points, but still require them for annotations, making annotating challenging and time-consuming. Due to the abovementioned occlusions, there can be conflicting and missing annotations of the same fruit between different images. Novel 3D reconstruction methods can be used to simplify annotating and enlarge datasets. We propose a novel pipeline consisting of 3D Gaussian Splatting to reconstruct an orchard scene, simplified annotations, automated projection of the annotations to images, and the training and evaluation of a pose estimation method. Using our pipeline, 105 manual annotations were required to obtain 28,191 training labels, a reduction of 99.6%. Experimental results indicated that training with labels of fruits that are $\leq95\%$ occluded resulted in the best performance, with a neutral F1 score of 0.927 on the original images and 0.970 on the rendered images. Adjusting the size of the training dataset had small effects on the model performance in terms of F1 score and pose estimation accuracy. It was found that the least occluded fruits had the best position estimation, which worsened as the fruits became more occluded. It was also found that the tested pose estimation method was unable to correctly learn the orientation estimation of apples.
Why it matches plant phenotyping methodsリンゴの姿勢推定アノテーションを大幅に効率化する3D再構成・自動ラベル投影パイプラインを開発し、姿勢推定性能も評価しているため、植物フェノタイピング手法が中心である。
abstractWe propose a novel pipeline consisting of 3D Gaussian Splatting to reconstruct an orchard scene, simplified annotations, automated projection of the annotations to images, and the training and evaluation of a pose estimation method.
Reproduction assets foundThe paper explicitly provides two paper-specific public assets: the authors' phenotyping/pose-estimation pipeline code on GitHub and the collected apple orchard image dataset on a 4TU DOI. Both are directly used for the paper's measurements and analysis.Dataset · publicIn total, 367 images were collected. The dataset is available at https://doi.org/10.4121/976c94f2-028f-4291-adfd-20eb82b0f647Open asset ↗10.4121/976c94f2-028f-4291-adfd-20eb82b0f647lines:92-108Code / dataset availability confirmedEurope PMC · bioRxiv · checked 15 Sept 2026
FlowerAnnotation / quality controlObject detectionGrowth / development / phenology
ABSTRACT Herbarium specimens represent critical historical records of plant phenology, yet automating annotation of reproductive structures remains challenging given the diversity of floral morphologies, specimen age and quality, and image quality. Here, we present a machine learning pipeline that uses an ensemble modeling approach to detect flowers on herbarium specimens and deliver these data to the phenology research community. After testing multiple strategies for generating training data, we found in-house expert-curated annotations were essential for producing reliable results. Expert validation found relatively strong accuracy for detecting present floral structures, but still had moderately high false negative rates. Applying the ensemble to our filtered final image dataset of 22 million records resulted in 11.1 million records labeled with flowers present. However, only 2.9 million of these contained complete metadata necessary for downstream phenology research, highlighting the need for full label digitization efforts. Still, this dataset represents a large compilation of historical herbarium-derived phenology records available as a resource for the phenology community. We end by demonstrating how integrating these machine-labeled records into Phenobase, a publicly-available phenology database, expands taxonomic and temporal coverage for large-scale phenological analyses, and discuss remaining challenges and next steps.
Why it matches plant phenotyping methods植物標本画像から花の存在を自動検出し、精度検証と大規模な phenology データセット化を行う機械学習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適格です。
abstractwe present a machine learning pipeline that uses an ensemble modeling approach to detect flowers on herbarium specimens and deliver these data to the phenology research community.
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits the ensemble models, training/validation/test images, training data and final ensemble output on Zenodo, and the analysis code on GitHub; machine-labeled records are also served via the public Phenobase portal. All are paper-specific, public, and actionable.Dataset · publicors contributed to drafts and gave final
454
approval for publication.
455
456
Data Availability Statement
457
The ensemble data models and a corresponding JSON file with model metadata data are
458
housed on Zenodo (https://doi.org/10.5281/zenodo.17079402). Images used in training,
459
validation, and testing are located here: https://zenodo.org/records/17675089. Code used for
460
this project can be found on github (https://github.com/rafelafrance/phenobase/tree/v1.0.0).461
Training data and final ensemble output can be found on Zenodo
462
(https://doi.org/10.5281/zenodo.17675089).463
464
Supporting Information
465
Additional Supporting Information may be found online in the SupportinOpen asset ↗Zenodo · 17675089pdf-raw-page:19 lines:1-55Dataset · publictps://doi.org/10.5281/zenodo.17079402). Images used in training,
459
validation, and testing are located here: https://zenodo.org/records/17675089. Code used for
460
this project can be found on github (https://github.com/rafelafrance/phenobase/tree/v1.0.0).461
Training data and final ensemble output can be found on Zenodo
462
(https://doi.org/10.5281/zenodo.17675089).463
464
Supporting Information
465
Additional Supporting Information may be found online in the Supporting Information section at
466
the end of the article.
467
Appendix S1. List of difficult-to-annotate genera and families removed from training and
468
downstream data.
469
Appendix S2. Table S1. Validation results for held-oOpen asset ↗Zenodo · 10.5281/zenodo.17675089pdf-raw-page:19 lines:1-55Code · publiclity Statement
457
The ensemble data models and a corresponding JSON file with model metadata data are
458
housed on Zenodo (https://doi.org/10.5281/zenodo.17079402). Images used in training,
459
validation, and testing are located here: https://zenodo.org/records/17675089. Code used for
460
this project can be found on github (https://github.com/rafelafrance/phenobase/tree/v1.0.0).461
Training data and final ensemble output can be found on Zenodo
462
(https://doi.org/10.5281/zenodo.17675089).463
464
Supporting Information
465
Additional Supporting Information may be found online in the Supporting Information section at
466
the end of the article.
467
Appendix S1. List of difficult-to-annotaOpen asset ↗GitHub · rafelafrance/phenobasepdf-raw-page:19 lines:1-55Code / dataset availability confirmedCrossref · OpenAlex · Europe PMC · checked 14 Sept 2026
MultimodalLiDAR / point cloudAnnotation / quality controlClassificationObject detectionCalibration / preprocessingSegmentationTracking
Plant phenomics, the comprehensive study of plant phenotypes, has gained prominence as a vital tool for understanding the intricate relationships between genotypes and the environment. Image-based plant phenomics has progressed rapidly, and three-dimensional (3D) phenotyping is a valuable extension of traditional 2D phenomics. However, the increased data dimensionality poses challenges to feature extraction and phenotyping. In recent decades, deep learning has led to remarkable progress in revolutionizing 3D phenotyping. Therefore, this review highlights the importance of using deep learning in 3D plant phenomics. It systematically overviews the capabilities of deep learning for 3D computer vision, covering 3D representation, classification, detection and tracking, semantic segmentation, instance segmentation, and generation. Additionally, deep learning techniques for 3D point preprocessing (e.g., annotation, downsampling, and dataset organization) and various plant phenotyping tasks are discussed. Finally, the challenges and perspectives associated with deep learning in 3D plant phenomics are summarized, including (1) benchmark dataset construction by using synthetic datasets and methods such as generative artificial intelligence and unsupervised or weakly supervised learning; (2) accurate and efficient 3D point cloud analysis by leveraging multitask learning, lightweight models, and self-supervised learning; and (3) deep learning for 3D plant phenomics by exploring interpretability, extensibility, and multimodal data utilization. The exploration of deep learning in 3D plant phenomics is poised to spur breakthroughs in a new dimension of plant science.
Why it matches plant phenotyping methods3D植物フェノミクスにおける深層学習手法を体系的にレビューしており、植物形質の抽出・推定手法が中心である。
abstractTherefore, this review highlights the importance of using deep learning in 3D plant phenomics.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe dataset can be downloaded from https://github.com/Jinlab-AiPhenomics/Mazie3D.Open asset ↗Jinlab-AiPhenomics/Mazie3Dhtml-lines:332-336Code / dataset availability confirmedCrossref · Europe PMC · checked 6 Sept 2026
This data descriptor presents novel, annotated 3D point cloud plant scans generated by a high-throughput phenotyping platform (LeasyScan, ICRISAT, India). It focuses on broad-leaf legume species (mungbean, common bean, cowpea, and lima bean). The dataset, generated by PlantEye(R) F600 technology, captures multispectral 3D scans of plant canopies. It includes 223 scans, providing detailed organ-level segmentation annotations for embryonic leaves, leaves, petioles, stems, and whole plants. The dataset fills a critical gap in plant phenomics research by offering a base of annotated data to support AI model development efforts in 3D computer vision. Data preprocessing, annotation procedures, and potential applications in crop research disciplines are further discussed. The dataset, preprocessing code, annotations, and a MIAPPE-compliant data sheet are also presented via the GitHub repository for further updates and expansion.
Why it matches plant phenotyping methods植物フェノタイピングプラットフォームで取得した3D点群と器官レベル注釈を提供するデータセットで、再利用可能な画像解析・AI開発基盤が中心です。
abstractThis data descriptor presents novel, annotated 3D point cloud plant scans generated by a high-throughput phenotyping platform (LeasyScan, ICRISAT, India).
Reproduction assets foundThe paper's own annotated 3D point cloud dataset (223 scans of legumes with organ-level segmentation annotations), raw scanner data, MIAPPE metadata, and preprocessing/cuboid-generation/baseline-evaluation code are publicly deposited on Figshare and mirrored on GitHub.Code · publicinto this software. All the code and data are also available as the GitHub (https://github.com/kit-pef-czu-czOpen asset ↗GitHubpdf-page:2 lines:1-58Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
Accurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies. While RGB-D cameras enable high-throughput scanning in operational settings like harvesting conveyors, they produce incomplete, low-quality 3D models. Conversely, close-range Structure-from-Motion (SfM) produces high-quality reconstructions but is not suitable for high-throughput field application. This study bridges this gap through 3DPotatoTwin , a paired dataset containing 339 tuber samples across three cultivars collected in Hokkaido, Japan. Our dataset uniquely combines: (1) conveyor-acquired RGB-D point clouds, (2) ground measurement, (3) SfM reconstructions under indoor controlled environment, and (4) aligned model pairs with transformation matrices. The multi-sensory alignment employs an semi-supervised pin-guided pipeline incorporating single-pin extraction and referencing, cross-strip matching, and binary-color-enhanced ICP, achieving 0.59 ± 0.11 mm registration accuracy. Beyond serving as a benchmark for 3D phenotyping algorithms, the dataset enables training of 3D completion networks to reconstruct high-quality 3D models from partial RGB-D point clouds. Meanwhile, the proposed semi-automated annotation pipeline has the potential to accelerate 3D dataset generation for similar studies. The presented methodology demonstrates broader applicability for multi-sensor data fusion across crop phenotyping applications. The dataset and pipeline source code are publicly available at HuggingFace and GitHub, respectively.
Why it matches plant phenotyping methodsジャガイモ塊茎の3D表現型計測を対象に、RGB-D・SfM・地上計測を統合したデータセット、位置合わせパイプライン、ベンチマークを開発しており、表現型取得手法が中心である。
abstractAccurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicAll the batch processing scripts mentioned in this section were provided in the 3dscan folder at Github (https://github.com/UTokyo-FieldPhenomics-Lab/PotatoScan/).Open asset ↗UTokyo-FieldPhenomics-Lab/PotatoScanhtml-lines:119-131Code / dataset availability confirmedEurope PMC · bioRxiv · checked 6 Sept 2026
RGB / grayscaleLeafAnnotation / quality controlClassificationGrowth / development / phenology
ABSTRACT Plant phenology dictates many aspects of community function and ecosystem dynamics. Yet, global phenology data are still limited, especially in areas lacking monitoring programs. Here we present a new data resource, PhenoVision–Leaf, which extends a computer-vision pipeline utilizing iNaturalist digital image vouchers to produce global-scale leaf phenophase data for deciduous, woody genera. We first discuss our implementation of a new human annotation framework for leaf phenology on iNaturalist, aligning with phenophase definitions used by the larger phenology community. We then showcase the use of 165,988 crowdsourced annotated records to train a Vision Transformer model with a two-stage regime to maximize accuracy across single- and multi-image records. This approach extends Phenovision from scoring individual images to aggregating at the iNaturalist record level, better aligning with human annotation processes. Post-hoc validation showed high performance for detecting present green and colored leaves (>98% accuracy), and reasonable accuracy for breaking leaf buds (>87% accuracy). Applying PhenoVision–Leaf to over 26 million iNaturalist records yielded 5.6 million record-level phenology observations across 6,500 species and 57 families, filling geographic and taxonomic gaps. These data, now accessible through the Phenobase portal, establish a foundation for near real-time monitoring of leaf phenology, supporting global-scale synthesis analyses.
Why it matches plant phenotyping methods葉のフェノロジー状態を画像から推定するコンピュータビジョン手法と、注釈・学習・検証・大規模データ生成基盤が研究の中心であるため。
abstractwe present a new data resource, PhenoVision–Leaf, which extends a computer-vision pipeline utilizing iNaturalist digital image vouchers to produce global-scale leaf phenophase data
Reproduction assets foundThe paper's PhenoVision–Leaf record-level leaf phenology dataset (5.6M machine-labeled observations) is publicly available via the Phenobase portal, and the underlying iNaturalist images used for training and machine labeling are available through the iNaturalist open data repository on AWS. No author analysis code or Dataset · publicAll images associated with these records were
downloaded using iNaturalist’s open data repository on AWS
(https://registry.opendata.aws/inaturalist-open-data/).Open asset ↗pdf-page:4 lines:1-44Code / dataset availability confirmedCrossref · checked 6 Sept 2026
Object detection is essential for precision agriculture applications like automated plant counting, but the minimum dataset requirements for effective model deployment remain poorly understood for arable crop seedling detection on orthomosaics. This study investigated how much annotated data is required to achieve standard counting accuracy (R2 = 0.85) for maize seedlings across different object detection approaches. We systematically evaluated traditional deep learning models requiring many training examples (YOLOv5, YOLOv8, YOLO11, RT-DETR), newer approaches requiring few examples (CD-ViTO), and methods requiring zero labeled examples (OWLv2) using drone-captured orthomosaic RGB imagery. We also implemented a handcrafted computer graphics algorithm as baseline. Models were tested with varying training sources (in-domain vs. out-of-distribution data), training dataset sizes (10–150 images), and annotation quality levels (10–100%). Our results demonstrate that no model trained on out-of-distribution data achieved acceptable performance, regardless of dataset size. In contrast, models trained on in-domain data reached the benchmark with as few as 60–130 annotated images, depending on architecture. Transformer-based models (RT-DETR) required significantly fewer samples (60) than CNN-based models (110–130), though they showed different tolerances to annotation quality reduction. Models maintained acceptable performance with only 65–90% of original annotation quality. Despite recent advances, neither few-shot nor zero-shot approaches met minimum performance requirements for precision agriculture deployment. These findings provide practical guidance for developing maize seedling detection systems, demonstrating that successful deployment requires in-domain training data, with minimum dataset requirements varying by model architecture.
Why it matches plant phenotyping methodsトウモロコシ幼苗の個体数という植物形質を画像から推定する物体検出手法について、複数モデル、データ量、アノテーション品質を系統的に比較・評価しており、手法の性能検証が中心である。
abstractThis study investigated how much annotated data is required to achieve standard counting accuracy (R2 = 0.85) for maize seedlings across different object detection approaches.
Reproduction assets foundThe paper's Data Availability Statement provides two paper-specific public assets: the authors' handcrafted-method analysis code on a GitHub gist and the ID (in-distribution) annotation datasets created for this study on Zenodo. Both have explicit availability language and public URLs.Code · publicThe code for the handcrafted methods used in this study is available at
https://gist.github.com/SamueleBumbaca/4a227bbe7b78d6be3424899c16c60bb4 (accessed on 20
June 2025).Open asset ↗gist.github.com/SamueleBumbacapdf-page:23 lines:1-52Dataset · publicThe datasets created during this study (ID datasets) are available at the Zenodo repository
https://doi.org/10.5281/zenodo.15235602 (accessed on 20 June 2025)Open asset ↗Zenodo · 10.5281/zenodo.15235602pdf-page:23 lines:1-52Code / dataset availability confirmedOpenAlex · arXiv · checked 15 Sept 2026
Quantitative descriptions of the complete canopy architecture are essential for accurately evaluating crop photosynthesis and yield performance to guide ideotype design. Although various sensing technologies have been developed for three-dimensional (3D) reconstruction of individual plants and canopies, they failed to obtain an accurate description of canopy architectures due to severe occlusion among complex canopy architectures. We proposed an effective method for 3D reconstruction of complex, dynamic population canopy architecture for rapeseed crops with a novel point cloud completion model. A complete point cloud generation framework was developed for automated annotation of the training dataset by distinguishing surface points from occluded points within canopies. The crop population point cloud completion network (CP-PCN) was then designed with a multi-resolution dynamic graph convolutional encoder (MRDG) and a point pyramid decoder (PPD) to predict occluded points. To further enhance feature extraction, a dynamic graph convolutional feature extractor (DGCFE) module was proposed to capture structural variations over the whole rapeseed growth period. The results demonstrated that CP-PCN achieved chamfer distance (CD) values of 3.35 cm -4.51 cm over four growth stages, outperforming the state-of-the-art transformer-based method (PoinTr). Ablation studies confirmed the effectiveness of the MRDG and DGCFE modules. Moreover, the validation experiment demonstrated that the silique efficiency index developed from CP-PCN improved the overall accuracy of rapeseed yield prediction by 11.2% compared to that of using incomplete point clouds. The CP-PCN pipeline has the potential to be extended to other crops, significantly advancing the quantitatively analysis of in-field population canopy architectures.
Why it matches plant phenotyping methods作物群落キャノピーの3D形態を復元する点群補完法を開発し、既存法との比較、アブレーション、収量予測への有効性検証まで行っており、植物フェノタイピング手法が研究の中心である。
abstractWe proposed an effective method for 3D reconstruction of complex, dynamic population canopy architecture for rapeseed crops with a novel point cloud completion model.
Reproduction assets foundThe paper's availability statement explicitly deposits all source code and test data (rapeseed canopy point cloud completion, CP-PCN) on GitHub at the allowed URL.Code · publicn Wang, Yi Feng,
Mengjie Gong and Guangyu Wu, for their participation in the experiments, and
to the Jiaxing Academy of Agricultural Sciences for their assistance with the
experimental data acquisition.
Availability of supporting data and source code
All source codes and test data involved in this study are available on
GitHub (https://github.com/Ziyue-Guo/RP-PCN.git).
Declaration of Competing Interest
The authors declare that they have no known competing financial interests
or personal relationships that could have appeared to influence the work
reported in this paper.
Contributions
Z. G. designed the study, conducted the experiments, and wrote the
manuscript. Y. S. contributed to the expeOpen asset ↗Ziyue-Guo/RP-PCNpdf-layout-page:42 lines:1-42Code / dataset availability confirmedCrossref · checked 6 Sept 2026
Extracting the quantity and geolocation data of small objects at the organ level via large-scale aerial drone monitoring is both essential and challenging for precision agriculture. The quality of reconstructed digital orthophoto maps (DOMs) often suffers from seamline distortion and ghost effects, making it difficult to meet the requirements for organ-level detection. While raw images do not exhibit these issues, they pose challenges in accurately obtaining the geolocation data of detected small objects. The detection of small objects was improved in this study through the fusion of orthophoto maps with raw images using the EasyIDP tool, thereby establishing a mapping relationship from the raw images to geolocation data. Small object detection was conducted by using the Slicing-Aided Hyper Inference (SAHI) framework and YOLOv10n on raw images to accelerate the inferencing speed for large-scale farmland. As a result, comparing detection directly using a DOM, the speed of detection was accelerated and the accuracy was improved. The proposed SAHI-YOLOv10n achieved precision and mean average precision (mAP) scores of 0.825 and 0.864, respectively. It also achieved a processing latency of 1.84 milliseconds on 640×640 resolution frames for large-scale application. Subsequently, a novel crop canopy organ-level object detection dataset (CCOD-Dataset) was created via interactive annotation with SAHI-YOLOv10n, featuring 3986 images and 410,910 annotated boxes. The proposed fusion method demonstrated feasibility for detecting small objects at the organ level in three large-scale in-field farmlands, potentially benefiting future wide-range applications.
Why it matches plant phenotyping methodsUAV画像と生画像の融合、SAHI-YOLOv10nによる作物器官の検出・位置推定を中心に開発・評価し、器官レベルの大規模データセットも構築しているため、植物表現型取得法が研究の中心である。
abstractThe detection of small objects was improved in this study through the fusion of orthophoto maps with raw images using the EasyIDP tool, thereby establishing a mapping relationship from the raw images to geolocation data.
Reproduction assets foundThe paper's CCOD-Dataset (3986 UAV images, 410,910 annotated bounding boxes of crop canopy organs) is publicly released on Hugging Face, and the authors' SAHI-YOLOv10 detection framework code is hosted in a public GitHub repository. Other code (fusion/EasyIDP pipeline) is only available upon request.Dataset · publicThe CCOD-Dataset link
is publicly available at Hugging Face at https://huggingface.co/datasets/Nir-Open asset ↗CCOD-Datasetpdf-page:8 lines:1-51Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Ontologies are highly prevalent in biology and medicine and are always evolving. Annotating biological text, such as observed phenotype descriptions, with ontology terms is a challenging and tedious task. The process of annotation requires a contextual understanding of the input text and of the ontological terms available. While text-mining tools are available to assist, they are largely based on directly matching words and phrases and so lack understanding of the meaning of the query item and of the ontology term labels. Large Language Models (LLMs), however, excel at tasks that require semantic understanding of input text and therefore may provide an improvement for the auto-annotation of text with ontological terms. Here we describe a series of workflows incorporating OpenAI GPT's capabilities to annotate Arabidopsis thaliana and forest tree phenotypic observations with ontology terms, aiming for results that resemble manually curated annotations. These workflows make use of an LLM to intelligently parse phenotypes into short concepts, followed by finding appropriate ontology terms via embedding vector similarity or via Retrieval-Augmented Generation (RAG). The RAG model is a state-of-the-art approach that augments conversational prompts to the LLM with context-specific data to empower it beyond its pre-trained parameter space. We show that the RAG produces the most accurate automated annotations that are often highly similar or identical to expert-curated annotations.
Why it matches plant phenotyping methods植物の表現型観察記述をオントロジー語に自動アノテーションするLLM/RAGワークフローの開発・精度評価が中心であり、再利用可能な計算ツールとして植物表現型データを処理する。
abstractHere we describe a series of workflows incorporating OpenAI GPT's capabilities to annotate Arabidopsis thaliana and forest tree phenotypic observations with ontology terms
Reproduction assets foundThe paper's phenotype descriptors, gold-standard annotations, LLM-parsed concepts, auto-annotations, and evaluation scores are publicly available as supplementary files, and the authors' analysis code (DE, DCE, DCRAG workflows) is publicly deposited on GitHub. A specific AraPheno trait (#278) used as an input example/`Code · publicCode to execute the DE, DCE and DCRAG workflows is available at https://github.com/dkainer/LLMannotator .Open asset ↗github.com/dkainer/LLMannotatorlines:229-257Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
In the era of machine learning-driven plant imaging, the production of annotated datasets is a very important contribution. In this data paper, a unique annotated dataset of seedling emergence kinetics is proposed. It is composed of almost 70,000 RGB-depth frames and more than 700,000 plant annotations. The dataset is shown valuable for training deep learning models and performing high-throughput phenotyping by imaging. The ability of such models to generalize to several species and outperform the state-of-the-art owing to the delivered dataset is demonstrated. We also discuss how this dataset raises new questions in plant phenotyping.
Why it matches plant phenotyping methods植物の出芽速度を対象とする大規模RGB深度画像・アノテーションデータセットを提供し、深層学習および高スループット表現型解析への利用性を実証しており、表現型取得基盤が中心である。
abstracta unique annotated dataset of seedling emergence kinetics is proposed
Reproduction assets foundThis is a data paper whose core contribution is a public annotated RGB-depth seedling dataset (~70,000 frames, >700,000 annotations) deposited in DATA INRAE with DOI 10.57745/AMFJTK, explicitly stated as publicly accessible. Other allowed URLs (license, Intel datasheet, Jülich record) are not paper-specific assets.Dataset · publicSynthesis of the full time-lapse and RGB-Depth full frame quantity per species
Species
Pots time-lapse
Labelled pots time-lapse
RGB-depth full frame
Rapeseed
1 760
336
15 218
Tomatoes
1 960
480
33 283
Beans
2 320
400
21 445
Total
6 040
1 216
69 946
The dataset is publicly accessible in the DATA INRAE repository, DOI: https://doi.org/10.57745/AMFJTK . The file tree structure is illustrated in Fig. 4 . The dataset is organized into 11 compressed .zip files, each corresponding to a distinct trial. Within these files, images are sorted chronologically by acquisition start date, then by camera, and stored in .png format within dedicated color and depth folders. Labels are alsoOpen asset ↗DATA INRAE · 10.57745/AMFJTKlines:105-195Code / dataset availability confirmedOpenAlex · Crossref · checked 8 Sept 2026
Computer vision techniques offer promising tools for disease detection in orchards and can enable effective phenotyping for the selection of resistant cultivars in breeding programmes and research. In this study, a digital phenotyping system for disease detection and monitoring was developed using drones, object detection and photogrammetry, focusing on European pear rust (Gymnosporangium sabinae) as a model pathogen. High-resolution RGB images from ten low-altitude drone flights were collected in 2021, 2022 and 2023. A total of 16,251 annotations of leaves with pear rust symptoms were created on 584 images using the Computer Vision Annotation Tool (CVAT). The YOLO algorithm was used for the automatic detection of symptoms. A novel photogrammetric approach using Agisoft’s Metashape Professional software ensured the accurate localisation of symptoms. The geographic information system software QGIS calculated the infestation intensity per tree based on the canopy areas. This drone-based phenotyping system shows promising results and could considerably simplify the tasks involved in fruit breeding research.
Why it matches plant phenotyping methodsドローン画像、物体検出、写真測量を統合し、ナシ樹のさび病症状を検出・局在化して樹体ごとの感染強度を推定するデジタル表現型解析システムの開発が中心である。
abstracta digital phenotyping system for disease detection and monitoring was developed using drones, object detection and photogrammetry
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits the annotated UAV image dataset on Mendeley Data and the trained model, detection workflow, and Metashape loading script on figshare, both with public URLs matching allowed entries.Dataset · publicsource repository Mendeley Data (https://data.mendeley.com/datasets/44kjgc4gkc/1, accessed on 8Open asset ↗Mendeley Data · 44kjgc4gkc/1pdf-page:15 lines:1-67Code · publicThe model, the detection workflow with instructions and the script for
loading the detections into Agisoft’s Metashape are available in the open-source figshare repository
(https://doi.org/10.6084/m9.figshare.27225312.v2, accessed on 28 October 2024).Open asset ↗figshare · 10.6084/m9.figshare.27225312.v2pdf-page:15 lines:1-67Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Growth monitoring of crops is a crucial aspect of precision agriculture, essential for optimal yield prediction and resource allocation. Traditional crop growth monitoring methods are labor-intensive and prone to errors. This study introduces an automated segmentation pipeline utilizing multi-date aerial images and ortho-mosaics to monitor the growth of cauliflower crops (Brassica Oleracea var. Botrytis) using an object-based image analysis approach. The methodology employs YOLOv8, a Grounding Detection Transformer with Improved Denoising Anchor Boxes (DINO), and the Segment Anything Model (SAM) for automatic annotation and segmentation. The YOLOv8 model was trained using aerial image datasets, which then facilitated the training of the Grounded Segment Anything Model framework. This approach generated automatic annotations and segmentation masks, classifying crop rows for temporal monitoring and growth estimation. The study’s findings utilized a multi-modal monitoring approach to highlight the efficiency of this automated system in providing accurate crop growth analysis, promoting informed decision-making in crop management and sustainable agricultural practices. The results indicate consistent and comparable growth patterns between aerial images and ortho-mosaics, with significant periods of rapid expansion and minor fluctuations over time. The results also indicated a correlation between the time and method of observation which paves a future possibility of integration of such techniques aimed at increasing the accuracy in crop growth monitoring based on automatically derived temporal crop row segmentation masks.
Why it matches plant phenotyping methods航空画像・オルソモザイクから作物列を自動セグメンテーションし、時系列の生育・成長を推定する画像解析パイプラインが研究の中心であり、植物形質の取得手法として適格。
abstractThis study introduces an automated segmentation pipeline utilizing multi-date aerial images and ortho-mosaics to monitor the growth of cauliflower crops
Reproduction assets foundThe paper's Data Availability Statement points to the authors' public Mendeley Data repository (GobhiSet, DOI 10.17632/dcjjcwc5dh.4), which contains the raw, manually, and automatically annotated RGB aerial images and ortho-mosaics of cauliflower used for the YOLOv8x-seg and Grounded SAM training and growth analysis inDataset · publicon of the manuscript.
Funding: This research received no external funding.
Data Availability Statement: No new data was created. However, the data that were used to perform
this research can be found in the article published at https://doi.org/10.1016/j.dib.2024.110506 and
available in the repository DOI: 10.17632/dcjjcwc5dh.4 (https://data.mendeley.com/drafts/dcjjcwc5dh).Conflicts of Interest: The authors declare no conflicts of interest.
References
1. Di, L.; Ustundag, B. Crop Growth Modeling and Yield Forecasting. In Agro-Geoinformatics; Springer: Cham, Switzerland, 2021.
[CrossRef]
2. Mithen, S.; Jenkins, E.; Jamjoum, K.; Nuimat, S.; Nortcliff, S.; Finlayson, B. Experimental crop growingOpen asset ↗data.mendeley.com · 10.17632/dcjjcwc5dh.4pdf-raw-page:17 lines:1-52Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Annotation / quality controlVisualization / data managementDisease symptoms / severity
Phytoplasmas are small, intracellular bacteria that infect a vast range of plant species, causing significant economic losses and impacting agriculture and farmers' livelihoods. Early and rapid diagnosis of phytoplasma infections is crucial for preventing the spread of these diseases, particularly through early symptom recognition in the field by farmers and growers. A symptom database for phytoplasma infections can assist in recognizing the symptoms and enhance early detection and management. In this study, nearly 35,000 phytoplasma sequence entries were retrieved from the NCBI nucleotide database using the keyword "phytoplasma" and information on phytoplasma disease-associated plant hosts and symptoms was gathered. A total of 945 plant species were identified to be associated with phytoplasma infections. Subsequently, links to symptomatic images of these known susceptible plant species were manually curated, and the Phytoplasma Disease Symptom Database ( i PhyDSDB) was established and implemented on a web-based interface using the MySQL Server and PHP programming language. One of the key features of i PhyDSDB is the curated collection of links to symptomatic images representing various phytoplasma-infected plant species, allowing users to easily access the original source of the collected images and detailed disease information. Furthermore, images and descriptive definitions of typical symptoms induced by phytoplasmas were included in i PhyDSDB. The newly developed database and web interface, equipped with advanced search functionality, will help farmers, growers, researchers, and educators to efficiently query the database based on specific categories such as plant host and symptom type. This resource will aid the users in comparing, identifying, and diagnosing phytoplasma-related diseases, enhancing the understanding and management of these infections.
Why it matches plant phenotyping methods植物の病徴画像と症状定義を体系的に収録し、植物病害状態の認識・診断に利用するデータベースとウェブインターフェースを開発した研究であり、病徴という植物状態の取得・参照基盤が中心です。
abstractSubsequently, links to symptomatic images of these known susceptible plant species were manually curated, and the Phytoplasma Disease Symptom Database ( i PhyDSDB) was established and implemented on a web-based interface using the MySQL Server and PHP programming language.
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 · publicwe established a database that consists of various phytoplasma diseases and their associated symptoms, and we implemented it on a web-based interface ( https://plantpathology.ba.ars.usda.gov/iphydsdb/iphydsdb.html , accessed on 23 May 2024). The database is called the Phytoplasma Disease and Symptom Database ( i PhyDSDB), which includes 1264 links to symptomatic images collected from 372 out of 945 plant speciesOpen asset ↗lines:30-40Code / dataset availability confirmedOpenAlex · checked 7 Sept 2026
Wheat varieties show a large diversity of traits and phenotypes. Linking them to genetic variability is essential for shorter and more efficient wheat breeding programs. A growing number of plant molecular information networks provide interlinked interoperable data to support the discovery of gene-phenotype interactions. A large body of scientific literature and observational data obtained in-field and under controlled conditions document wheat breeding experiments. The cross-referencing of this complementary information is essential. Text from databases and scientific publications has been identified early on as a relevant source of information. However, the wide variety of terms used to refer to traits and phenotype values makes it difficult to find and cross-reference the textual information, e.g. simple dictionary lookup methods miss relevant terms. Corpora with manually annotated examples are thus needed to evaluate and train textual information extraction methods. While several corpora contain annotations of human and animal phenotypes, no corpus is available for plant traits. This hinders the evaluation of text mining-based crop knowledge graphs (e.g. AgroLD, KnetMiner, WheatIS-FAIDARE) and limits the ability to train machine learning methods and improve the quality of information. The Triticum aestivum trait Corpus is a new gold standard for traits and phenotypes of wheat. It consists of 528 PubMed references that are fully annotated by trait, phenotype, and species. We address the interoperability challenge of crossing sparse assay data and publications by using the Wheat Trait and Phenotype Ontology to normalize trait mentions and the species taxonomy of the National Center for Biotechnology Information to normalize species. The paper describes the construction of the corpus. A study of the performance of state-of-the-art language models for both named entity recognition and linking tasks trained on the corpus shows that it is suitable for training and evaluation. This corpus is currently the most comprehensive manually annotated corpus for natural language processing studies on crop phenotype information from the literature.
Why it matches plant phenotyping methods小麦の形質・表現型抽出とエンティティ linking のための手動アノテーションコーパスを構築し、言語モデルの訓練・評価に用いており、表現型情報の取得手法とデータセットが研究の中心である。
titleA manually annotated text dataset for trait and phenotype extraction and entity linking in wheat breeding literature
Reproduction assets foundThe paper's core assets are publicly available: the TaeC annotated corpus (trait/phenotype/species annotations of 528 PubMed wheat references) on Recherche Data Gouv, the Wheat Trait and Phenotype Ontology on AgroPortal, the AlvisNLP bread wheat workflow on Forgemia, and the ToMap method code on GitHub, all with authorDataset · publicThe corpus dataset TaeC is available under CC-BY-ND License at: https://entrepot.recherche.data.gouv.fr/dataset.xhtml?persistentId=doi:10.57745/GCYG3QOpen asset ↗entrepot.recherche.data.gouv.fr · doi:10.57745/GCYG3Qlines:142-152Code · publicThe code of the ToMap method is available under Apache License at https://github.com/Bibliome/alvisnlp/tree/master/alvisnlp-bibliome/src/main/java/fr/inra/maiage/bibliome/alvisnlp/bibliomefactory/modules/tomapOpen asset ↗github.comlines:142-152Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Deep learning and multimodal remote and proximal sensing are widely used for analyzing plant and crop traits, but many of these deep learning models are supervised and necessitate reference datasets with image annotations. Acquiring these datasets often demands experiments that are both labor-intensive and time-consuming. Furthermore, extracting traits from remote sensing data beyond simple geometric features remains a challenge. To address these challenges, we proposed a radiative transfer modeling framework based on the Helios 3-dimensional (3D) plant modeling software designed for plant remote and proximal sensing image simulation. The framework has the capability to simulate RGB, multi-/hyperspectral, thermal, and depth cameras, and produce associated plant images with fully resolved reference labels such as plant physical traits, leaf chemical concentrations, and leaf physiological traits. Helios offers a simulated environment that enables generation of 3D geometric models of plants and soil with random variation, and specification or simulation of their properties and function. This approach differs from traditional computer graphics rendering by explicitly modeling radiation transfer physics, which provides a critical link to underlying plant biophysical processes. Results indicate that the framework is capable of generating high-quality, labeled synthetic plant images under given lighting scenarios, which can lessen or remove the need for manually collected and annotated data. Two example applications are presented that demonstrate the feasibility of using the model to enable unsupervised learning by training deep learning models exclusively with simulated images and performing prediction tasks using real images.
Why it matches plant phenotyping methods植物のRGB・マルチ/ハイパースペクトル・熱・深度画像と植物形質ラベルを生成するシミュレーション基盤を開発しており、表現型取得・学習用データ生成が中心的な方法論的貢献である。
abstractwe proposed a radiative transfer modeling framework based on the Helios 3-dimensional (3D) plant modeling software designed for plant remote and proximal sensing image simulation.
Reproduction assets foundThe paper's phenotyping analysis relies on three public, paper-specific assets: the Helios framework code (used to generate the synthetic annotated images), the MSU-PID bean image dataset, and the strawberry.00 annotated dataset, all with explicit open-availability statements and URLs.Dataset · publicThe Bean data that support the findings of this study are openly available in MSU-PID at https://www.cse.msu.edu/computervision/MVA15-MSU-PID.zipOpen asset ↗MSU-PIDlines:255-283Dataset · publicThe strawberry data that support the findings of this study are openly available in strawberry.00 at https://universe.roboflow.com/skripsie/strawberry.00Open asset ↗strawberry.00lines:255-283Code / dataset availability confirmedOpenAlex · Europe PMC · Crossref · checked 15 Sept 2026
BACKGROUND: The study of plant photosynthesis is essential for productivity and yield. Thanks to the development of high-throughput phenotyping (HTP) facilities, based on chlorophyll fluorescence imaging, photosynthetic traits can be measured in a reliable, reproducible and efficient manner. In most state-of-the-art HTP platforms, these traits are automatedly analyzed at individual plant level, but information at leaf level is often restricted by the use of manual annotation. Automated leaf tracking over time is therefore highly desired. Methods for tracking individual leaves are still uncommon, convoluted, or require large datasets. Hence, applications and libraries with different techniques are required. New phenotyping platforms are initiated now more frequently than ever; however, the application of advanced computer vision techniques, such as convolutional neural networks, is still growing at a slow pace. Here, we provide a method for leaf segmentation and tracking through the fine-tuning of Mask R-CNN and intersection over union as a solution for leaf tracking on top-down images of plants. We also provide datasets and code for training and testing on both detection and tracking of individual leaves, aiming to stimulate the community to expand the current methodologies on this topic. RESULTS: We tested the results for detection and segmentation on 523 Arabidopsis thaliana leaves at three different stages of development from which we obtained a mean F-score of 0.956 on detection and 0.844 on segmentation overlap through the intersection over union (IoU). On the tracking side, we tested nine different plants with 191 leaves. A total of 161 leaves were tracked without issues, accounting to a total of 84.29% correct tracking, and a Higher Order Tracking Accuracy (HOTA) of 0.846. In our case study, leaf age and leaf order influenced photosynthetic capacity and photosynthetic response to light treatments. Leaf-dependent photosynthesis varies according to the genetic background. CONCLUSION: The method provided is robust for leaf tracking on top-down images. Although one of the strong components of the method is the low requirement in training data to achieve a good base result (based on fine-tuning), most of the tracking issues found could be solved by expanding the training dataset for the Mask R-CNN model.
Why it matches plant phenotyping methodsCNNによる葉のセグメンテーション・追跡手法を開発し、検出・追跡精度を検証した植物フェノタイピング研究である。
abstractHere, we provide a method for leaf segmentation and tracking through the fine-tuning of Mask R-CNN and intersection over union as a solution for leaf tracking on top-down images of plants.
Reproduction assets foundThe paper's authors explicitly state that the full project library (leaf detection/tracking code and dataset) is available as a public GitHub repository, which directly reproduces this paper's phenotyping analysis.Code · publicof the PyTorch-Vision GitHub repository was used for the model training, specifically the reference scripts found in the folder detection. These scripts are included in the project GitHub under the modelTraining folder without relevant modifications. The full library of this project is available as a public repository at GitHub https://github.com/Fedjurrui/Leaf-Tracking .
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
No competing interests declared.
References
1.Open asset ↗Fedjurrui/Leaf-Trackinglines:174-269Code / dataset availability confirmedOpenAlex · arXiv · checked 7 Sept 2026
Wheat varieties show a large diversity of traits and phenotypes. Linking them to genetic variability is essential for shorter and more efficient wheat breeding programs. Newly desirable wheat variety traits include disease resistance to reduce pesticide use, adaptation to climate change, resistance to heat and drought stresses, or low gluten content of grains. Wheat breeding experiments are documented by a large body of scientific literature and observational data obtained in-field and under controlled conditions. The cross-referencing of complementary information from the literature and observational data is essential to the study of the genotype-phenotype relationship and to the improvement of wheat selection. The scientific literature on genetic marker-assisted selection describes much information about the genotype-phenotype relationship. However, the variety of expressions used to refer to traits and phenotype values in scientific articles is a hinder to finding information and cross-referencing it. When trained adequately by annotated examples, recent text mining methods perform highly in named entity recognition and linking in the scientific domain. While several corpora contain annotations of human and animal phenotypes, currently, no corpus is available for training and evaluating named entity recognition and entity-linking methods in plant phenotype literature. The Triticum aestivum trait Corpus is a new gold standard for traits and phenotypes of wheat. It consists of 540 PubMed references fully annotated for trait, phenotype, and species named entities using the Wheat Trait and Phenotype Ontology and the species taxonomy of the National Center for Biotechnology Information. A study of the performance of tools trained on the Triticum aestivum trait Corpus shows that the corpus is suitable for the training and evaluation of named entity recognition and linking.
Why it matches plant phenotyping methods小麦の形質・表現型を文献から抽出・リンクするための注釈付きデータセットを開発し、ツール性能も評価しており、植物表現型情報の計算的抽出が中心である。
abstractThe Triticum aestivum trait Corpus is a new gold standard for traits and phenotypes of wheat.
Reproduction assets foundThe paper's core asset, the TaeC annotated wheat trait/phenotype corpus, is publicly deposited on Recherche Data Gouv under CC-BY-ND. The authors' AlvisNLP wheat text-mining workflow and the ToMap method code are also publicly available. The WTO ontology used for annotation is public on AgroPortal.Dataset · publicTaeC is
available under CC-BY-ND License at:
https://entrepot.recherche.data.gouv.fr/dataset.xhtml?persistentId=doi:10.57745/GCYG3Q.Open asset ↗entrepot.recherche.data.gouv.fr · doi:10.57745/GCYG3Qpdf-page:9 lines:1-56Code · publicThe AlvisNLP bread wheat workflow is available at : https://forgemia.inra.fr/migale/wheat-tm.
It includes the wheat-specific lexica of ToMap.Open asset ↗forgemia.inra.frpdf-page:13 lines:1-53Code · publicThe code of the ToMap method is available at
https://github.com/Bibliome/alvisnlp/tree/master/alvisnlp-Open asset ↗github.com/Bibliome/alvisnlppdf-page:13 lines:1-53Code / dataset availability confirmedCrossref · Europe PMC · checked 15 Sept 2026
Plant disease diagnosis with estimation of disease severity at early stages still remains a significant research challenge in agriculture. It is helpful in diagnosing plant diseases at the earliest so that timely action can be taken for curing the disease. Existing studies often rely on labor-intensive manually annotated large datasets for disease severity estimation. In order to conquer this problem, a lightweight framework named “PDSE-Lite” based on Convolutional Autoencoder (CAE) and Few-Shot Learning (FSL) is proposed in this manuscript for plant disease severity estimation with few training instances. The PDSE-Lite framework is designed and developed in two stages. In first stage, a lightweight CAE model is built and trained to reconstruct leaf images from original leaf images with minimal reconstruction loss. In subsequent stage, pretrained layers of the CAE model built in the first stage are utilized to develop the image classification and segmentation models, which are then trained using FSL. By leveraging FSL, the proposed framework requires only a few annotated instances for training, which significantly reduces the human efforts required for data annotation. Disease severity is then calculated by determining the percentage of diseased leaf pixels obtained through segmentation out of the total leaf pixels. The PDSE-Lite framework’s performance is evaluated on Apple-Tree-Leaf-Disease-Segmentation (ATLDS) dataset. However, the proposed framework can identify any plant disease and quantify the severity of identified diseases. Experimental results reveal that the PDSE-Lite framework can accurately detect healthy and four types of apple tree diseases as well as precisely segment the diseased area from leaf images by using only two training samples from each class of the ATLDS dataset. Furthermore, the PDSE-Lite framework’s performance is compared with existing state-of-the-art techniques, and it is found that this framework outperformed these approaches. The proposed framework’s applicability is further verified by statistical hypothesis testing using Student t-test. The results obtained from this test confirm that the proposed framework can precisely estimate the plant disease severity with a confidence interval of 99%. Hence, by reducing the reliance on large-scale manual data annotation, the proposed framework offers a promising solution for early-stage plant disease diagnosis and severity estimation.
Why it matches plant phenotyping methods植物葉画像から病変画素率を算出して病害重症度を推定する画像解析手法を開発し、データセット上で比較・統計検証しており、植物表現型取得が中心である。
abstracta lightweight framework named “PDSE-Lite” based on Convolutional Autoencoder (CAE) and Few-Shot Learning (FSL) is proposed in this manuscript for plant disease severity estimation with few training instances.
Reproduction assets foundThe paper's plant-phenotyping measurements (apple leaf disease detection and severity estimation) were performed on the publicly available Apple-Tree-Leaf-Disease-Segmentation (ATLDS) dataset, which the authors link via a Science Data Bank deposit. No authors' analysis code or trained model checkpoints are explicitlydeDataset · publicrk of this research also includes the deployment of the PDSE-Lite framework on different IoT devices, such as Unmanned Aerial Vehicles (UAVs), to enable real-time monitoring of plant diseases in agricultural fields.
Data availability statement
Publicly available datasets were analyzed in this study. This data can be found here: https://www.scidb.cn/en/detail?dataSetId=0e1f57004db842f99668d82183afd578 .
Author contributions
PB: Conceptualization, Formal analysis, Methodology, Project administration, Supervision, Writing – review & editing. PG: Conceptualization, Methodology, Software, Visualization, Writing – original draft. SM: Formal analysis, Resources, Writing – review & editing.
Funding Open asset ↗lines:501-513Code / dataset availability confirmedOpenAlex · Europe PMC · bioRxiv · checked 7 Sept 2026
Abstract Image segmentation is commonly used to estimate the location and shape of plants and their external structures. Segmentation masks are then used to localize landmarks of interest and compute other geometric features that correspond to the plant’s phenotype. Despite its prevalence, segmentation-based approaches are laborious (requiring extensive annotation to train), and error-prone (derived geometric features are sensitive to instance mask integrity). Here we present a segmentation-free approach which leverages deep learning-based landmark detection and grouping, also known as pose estimation. We use a tool originally developed for animal motion capture called SLEAP (Social LEAP Estimates Animal Poses) to automate the detection of distinct morphological landmarks on plant roots. Using a gel cylinder imaging system across multiple species, we show that our approach can reliably and efficiently recover root system topology at high accuracy, few annotated samples, and faster speed than segmentation-based approaches. In order to make use of this landmark-based representation for root phenotyping, we developed a Python library ( sleap-roots ) for trait extraction directly comparable to existing segmentation-based analysis software. We show that landmark-derived root traits are highly accurate and can be used for common downstream tasks including genotype classification and unsupervised trait mapping. Altogether, this work establishes the validity and advantages of pose estimation-based plant phenotyping. To facilitate adoption of this easy-to-use tool and to encourage further development, we make sleap-roots , all training data, models, and trait extraction code available at: https://github.com/talmolab/sleap-roots and https://osf.io/k7j9g/ .
Why it matches plant phenotyping methods植物根の形態ランドマークをポーズ推定で検出し、根系形質を抽出する手法とソフトウェアを開発・検証した研究であり、植物フェノタイピング手法が中心である。
abstractHere we present a segmentation-free approach which leverages deep learning-based landmark detection and grouping, also known as pose estimation.
Reproduction assets foundThe paper makes its root phenotyping assets public: labeled training data, trained SLEAP models, and analysis files on OSF, the sleap-roots trait-extraction codebase on GitHub, and a separate figure-replication code repository.Dataset · publicThe datasets generated and/or analyzed during the current study are available in the Open Science Framework (OSF) repository. This includes the labeled data, predictive models, and analysis files which can be accessed via the following link: https://osf.io/k7j9g/ .Open asset ↗osf.io/k7j9glines:752-811Code · publicAdditionally, the specific code utilized for replicating the figures presented in this study can be found in a separate GitHub repository here: https://github.com/talmolab/Berrigan_et_al_sleap-roots .Open asset ↗talmolab/Berrigan_et_al_sleap-rootslines:752-811Code / dataset availability confirmedEurope PMC · Crossref · checked 14 Sept 2026
Applying deep learning to images of cropping systems provides new knowledge and insights in research and commercial applications. Semantic segmentation or pixel-wise classification, of RGB images acquired at the ground level, into vegetation and background is a critical step in the estimation of several canopy traits. Current state of the art methodologies based on convolutional neural networks (CNNs) are trained on datasets acquired under controlled or indoor environments. These models are unable to generalize to real-world images and hence need to be fine-tuned using new labelled datasets. This motivated the creation of the VegAnn - Vegetation Annotation - dataset, a collection of 3775 multi-crop RGB images acquired for different phenological stages using different systems and platforms in diverse illumination conditions. We anticipate that VegAnn will help improving segmentation algorithm performances, facilitate benchmarking and promote large-scale crop vegetation segmentation research.
Why it matches plant phenotyping methods作物RGB画像から植生を分割し、キャノピー形質推定に用いる注釈付きデータセットを作成・ベンチマークする研究であり、フェノタイピング用データ基盤が中心である。
abstractThis motivated the creation of the VegAnn - Vegetation Annotation - dataset, a collection of 3775 multi-crop RGB images acquired for different phenological stages using different systems and platforms in diverse illumination conditions.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe dataset can be downloaded from Zenodo: https://doi.org/10.5281/zenodo.763640828 and is under the
CC-BY license, allowing for reuse without restrictions.Open asset ↗Zenodo · 10.5281/zenodo.7636408pdf-page:8 lines:1-50Code / dataset availability confirmedEurope PMC · Crossref · checked 14 Sept 2026
Background The current development of sensor technologies towards ever more cost-effective and powerful systems is steadily increasing the application of low-cost sensors in different horticultural sectors. In plant in vitro culture, as a fundamental technique for plant breeding and plant propagation, the majority of evaluation methods to describe the performance of these cultures are based on destructive approaches, limiting data to unique endpoint measurements. Therefore, a non-destructive phenotyping system capable of automated, continuous and objective quantification of in vitro plant traits is desirable. Results An automated low-cost multi-sensor system acquiring phenotypic data of plant in vitro cultures was developed and evaluated. Unique hardware and software components were selected to construct a xyz-scanning system with an adequate accuracy for consistent data acquisition. Relevant plant growth predictors, such as projected area of explants and average canopy height were determined employing multi-sensory imaging and various developmental processes could be monitored and documented. The validation of the RGB image segmentation pipeline using a random forest classifier revealed very strong correlation with manual pixel annotation. Depth imaging by a laser distance sensor of plant in vitro cultures enabled the description of the dynamic behavior of the average canopy height, the maximum plant height, but also the culture media height and volume. Projected plant area in depth data by RANSAC (random sample consensus) segmentation approach well matched the projected plant area by RGB image processing pipeline. In addition, a successful proof of concept for in situ spectral fluorescence monitoring was achieved and challenges of thermal imaging were documented. Potential use cases for the digital quantification of key performance parameters in research and commercial application are discussed. Conclusion The technical realization of "Phenomenon" allows phenotyping of plant in vitro cultures under highly challenging conditions and enables multi-sensory monitoring through closed vessels, ensuring the aseptic status of the cultures. Automated sensor application in plant tissue culture promises great potential for a non-destructive growth analysis enhancing commercial propagation as well as enabling research with novel digital parameters recorded over time.
Why it matches plant phenotyping methods植物組織培養の形質を自動・非破壊・連続測定するマルチセンサーフェノタイピングシステムを開発し、画像分割や深度計測を検証しており、方法が研究の中心である。
abstractAn automated low-cost multi-sensor system acquiring phenotypic data of plant in vitro cultures was developed and evaluated.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicThe dataset supporting the conclusions of this article (Hard- and Software of “Phenomenon” phenotyping system) are available in an open-access Github repository, https://github.com/halube/Phenomenon .Open asset ↗halube/Phenomenonlines:224-282Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Maize is one of the most important staple food and cash crops that are largely produced by majority of smallholder farmers throughout the humid and sub-humid tropic of Africa. Despite its significance in the household food security and income, diseases, especially Maize Lethal Necrosis and Maize Streak, have been significantly affecting production of this crop. This paper offers a dataset of well curated images of maize crop for both healthy and diseased leaves captured using smartphone camera in Tanzania. The dataset is the largest publicly accessible dataset for maize leaves with a total of 18,148 images, which can be used to develop machine learning models for the early detection of diseases affecting maize. Moreover, the dataset can be used to support computer vision applications such as image segmentation, object detection and classification. The goal of generating this dataset is to assist the development of comprehensive tools that will help farmers in the diagnosis of diseases and the enhancement of maize yields thus eradicating the problem of fod security in Tanzania and other parts in Africa.
Why it matches plant phenotyping methodsトウモロコシの健全・罹病葉画像を収録した公開データセットで、植物病徴の画像ベース判定モデル開発を直接支援するため、フェノタイピング用データセットが中心です。
abstractThis paper offers a dataset of well curated images of maize crop for both healthy and diseased leaves captured using smartphone camera in Tanzania.
Reproduction assets foundThe paper's core asset is its own maize leaf imagery dataset (18,148 images of healthy/MLN/MSV leaves), publicly deposited by the authors on Harvard Dataverse with an explicit DOI and direct URL. The annotation tools cited (VisiPics, LabelMe, Makerere web annotation tool) are generic third-party tools, not paper assetsDataset · publica source location
•
Institution: The Nelson Mandela African Institution of Science and Technology (NM-AIST), Tanzania Agricultural Research Institute (TARI)
•
City/Town/Region: Arusha
•
Country: Tanzania
Data accessibility
Repository name: Harvard Dataverse
Data identification number: doi: 10.7910/DVN/GDON8Q
Direct URL to data: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GDON8Q
Open in a new tab
Value of the Data
•Open asset ↗Harvard Dataverse · doi:10.7910/DVN/GDON8Qlines:1-98Code / dataset availability confirmedEurope PMC · Crossref · checked 7 Sept 2026
Background Plant architecture can influence crop yield and quality. Manual extraction of architectural traits is, however, time-consuming, tedious, and error prone. The trait estimation from 3D data addresses occlusion issues with the availability of depth information while deep learning approaches enable learning features without manual design. The goal of this study was to develop a data processing workflow by leveraging 3D deep learning models and a novel 3D data annotation tool to segment cotton plant parts and derive important architectural traits. Results The Point Voxel Convolutional Neural Network (PVCNN) combining both point- and voxel-based representations of 3D data shows less time consumption and better segmentation performance than point-based networks. Results indicate that the best mIoU (89.12%) and accuracy (96.19%) with average inference time of 0.88 s were achieved through PVCNN, compared to Pointnet and Pointnet++. On the seven derived architectural traits from segmented parts, an R 2 value of more than 0.8 and mean absolute percentage error of less than 10% were attained. Conclusion This plant part segmentation method based on 3D deep learning enables effective and efficient architectural trait measurement from point clouds, which could be useful to advance plant breeding programs and characterization of in-season developmental traits. The plant part segmentation code is available at https://github.com/UGA-BSAIL/plant_3d_deep_learning .
Why it matches plant phenotyping methods3D深層学習による綿花の器官分割と建築形質抽出ワークフローを開発・比較検証しており、植物フェノタイピング手法が中心である。
abstractThe goal of this study was to develop a data processing workflow by leveraging 3D deep learning models and a novel 3D data annotation tool to segment cotton plant parts and derive important architectural traits.
Reproduction assets foundThe paper's plant part segmentation code is explicitly stated as publicly available in the authors' GitHub repository. The underlying datasets are only available on request, so they are noted as request-only.Code · publicThe plant part segmentation code is available at https://github.com/UGA-BSAIL/plant_3d_deep_learning .Open asset ↗UGA-BSAIL/plant_3d_deep_learninglines:1-72Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
A large amount of training data is usually lacking at the beginning of system development and labeling such a large number of RGB (red, green, blue) images is laborious. Interactive recurrent annotation is beneficial to incrementally gain training images in the stream of the system development and provides an opportunity to reduce human workload. We developed a software package, ROOSTER, to integrate both labeling and prediction in a single user-friendly graphic user interface with interactive deep learning to reduce the laborious human labeling for fast development of machine vision systems. Predictions can be performed under both single-image mode and batch mode for multiple images. The prediction results can be used as the initial image labeling and manually adjusted under a single image mode. Human labeling and machine predictions are visualized on the same image. ROOSTER provides fully automatic labeling for abundantly available initial images of wheat stripe rust to gain essential predictability. The navigation of integrating prediction with labeling benefits human adjustment to iteratively improve predictability. The development of a detection system for wheat stripe rust was presented as a use case to demonstrate the efficiency of using interactive deep learning to develop machine vision systems.
Why it matches plant phenotyping methods植物病害（コムギ縞萎縮病）の画像検出を対象とした対話型画像ラベリング・分類ソフトウェアを開発しており、植物病害状態の取得手法が中心である。
abstractWe developed a software package, ROOSTER, to integrate both labeling and prediction in a single user-friendly graphic user interface with interactive deep learning to reduce the laborious human labeling for fast development of machine vision systems.
Reproduction assets foundThe paper's wheat stripe rust use case is supported by a public Zenodo underlying dataset (400 author-captured training images and use case output files) and public author source code (zzlab.net, GitHub, archived Zenodo). The independent test data from Schirrmann et al. is only available on request.Dataset · publicilability
Underlying data
The independent data used to test ROOSTER was sourced from Schirrmann et al.,10
see here: https://doi.org/10.3389/fpls.2021.469689). Please contact the corresponding author of this article (mschirrmann@atb-potsdam.de) to request
access to the test data if interested.
Zenodo: ROOSTER underlying dataset. https://doi.org/10.5281/zenodo.7530460.11
This project contains the following underlying data:
- RawImages.zip (400 input training images used to develop the model, and captured by the authors of this
article).
- UseCase.zip (use case output files).
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).
Software aOpen asset ↗Zenodo · 10.5281/zenodo.7530460pdf-raw-page:5 lines:1-44Code · publicmages used to develop the model, and captured by the authors of this
article).
- UseCase.zip (use case output files).
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).
Software availability
Software available from: https://zzlab.net/ROOSTER.
Source code available from: https://github.com/12HuYang/ROOSTER.
Archived source code at time of publication: https://doi.org/10.5281/zenodo.7320405.12
License: MIT
Page 5 of 9Open asset ↗GitHub · 12HuYang/ROOSTERpdf-layout-page:5 lines:1-63Code · publicseCase.zip (use case output files).
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).
Software availability
Software available from: https://zzlab.net/ROOSTER.
Source code available from: https://github.com/12HuYang/ROOSTER.
Archived source code at time of publication: https://doi.org/10.5281/zenodo.7320405.12
License: MIT
Page 5 of 9Open asset ↗Zenodo · 10.5281/zenodo.7320405pdf-layout-page:5 lines:1-63Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Object detection models have become the current tool of choice for plant disease detection in precision agriculture. Most existing research improved the performance by ameliorating networks and optimizing the loss function. However, because of the vast influence of data annotation quality and the cost of annotation, the data-centric part of a project also needs more investigation. We should further consider the relationship between data annotation strategies, annotation quality, and the model's performance. In this paper, a systematic strategy with four annotation strategies for plant disease detection is proposed: local, semi-global, global, and symptom-adaptive annotation. Labels with different annotation strategies will result in distinct models' performance, and their contrasts are remarkable. An interpretability study of the annotation strategy is conducted by using class activation maps. In addition, we define five types of inconsistencies in the annotation process and investigate the severity of the impact of inconsistent labels on model's performance. Finally, we discuss the problem of label inconsistency during data augmentation. Overall, this data-centric quantitative analysis helps us to understand the significance of annotation strategies, which provides practitioners a way to obtain higher performance and reduce annotation costs on plant disease detection. Our work encourages researchers to pay more attention to annotation consistency and the essential issues of annotation strategy. The code will be released at: https://github.com/JiuqingDong/PlantDiseaseDetection_Yolov5 .
Why it matches plant phenotyping methods植物病害の画像検出におけるアノテーション戦略・一貫性とモデル性能を体系的に評価する手法研究であり、植物の病徴・病害状態の抽出方法が中心的である。
abstracta systematic strategy with four annotation strategies for plant disease detection is proposed: local, semi-global, global, and symptom-adaptive annotation.
Reproduction assets foundThe paper's authors explicitly state a public GitHub repository for the plant disease detection analysis code (YOLOv5-based annotation strategy/consistency experiments). No separate phenotype dataset deposit by the authors is stated; cited datasets (PlantVillage, etc.) are prior work.Code · publicThe code will be released at: https://github.com/JiuqingDong/PlantDiseaseDetection_Yolov5 .Open asset ↗JiuqingDong/PlantDiseaseDetection_Yolov5lines:224-322Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Here we provide the 'Global Spectrum of Plant Form and Function Dataset', containing species mean values for six vascular plant traits. Together, these traits -plant height, stem specific density, leaf area, leaf mass per area, leaf nitrogen content per dry mass, and diaspore (seed or spore) mass - define the primary axes of variation in plant form and function. The dataset is based on ca. 1 million trait records received via the TRY database (representing ca. 2,500 original publications) and additional unpublished data. It provides 92,159 species mean values for the six traits, covering 46,047 species. The data are complemented by higher-level taxonomic classification and six categorical traits (woodiness, growth form, succulence, adaptation to terrestrial or aquatic habitats, nutrition type and leaf type). Data quality management is based on a probabilistic approach combined with comprehensive validation against expert knowledge and external information. Intense data acquisition and thorough quality control produced the largest and, to our knowledge, most accurate compilation of empirically observed vascular plant species mean traits to date.
Why it matches plant phenotyping methods植物形質の大規模再利用可能データセットを構築し、確率的品質管理と外部情報による検証を実施しており、形質データ基盤が研究の中心である。
abstractHere we provide the 'Global Spectrum of Plant Form and Function Dataset', containing species mean values for six vascular plant traits.
Reproduction assets foundThe paper's core asset is the 'Global Spectrum of Plant Form and Function Dataset' (species mean values for six plant traits plus categorical traits and references), explicitly deposited publicly under a CC-BY license in the TRY File Archive with DOI 10.17871/TRY.81. This is a paper-specific, publicly actionable trait/Dataset · publicThe dataset is available under a CC-BY license at the TRY File Archive (https://www.try-db.org/TryWeb/Data.php):
Díaz, S. et al. The global spectrum of plant form and function: enhanced species-level trait dataset. TRY File
Archive https://doi.org/10.17871/TRY.81 (2022)244Open asset ↗TRY File Archive · 10.17871/TRY.81pdf-page:5 lines:1-62Code / dataset availability confirmedEurope PMC · Crossref · checked 8 Sept 2026
Objectives A well-known drawback to the implementation of Convolutional Neural Networks (CNNs) for image-recognition is the intensive annotation effort for large enough training dataset, that can become prohibitive in several applications. In this study we focus on applications in the agricultural domain and we implement Deep Learning (DL) techniques for the automatic generation of meaningful synthetic images of plant leaves, which can be used as a virtually unlimited dataset to train or validate specialized CNN models or other image-recognition algorithms. Methods Following an approach based on DL generative models, we introduce a Leaf-to-Leaf Translation (L2L) algorithm, able to produce collections of novel synthetic images in two steps: first, a residual variational autoencoder architecture is used to generate novel synthetic leaf skeletons geometry, starting from binarized skeletons obtained from real leaf images. Second, a translation via Pix2pix framework based on conditional generator adversarial networks (cGANs) reproduces the color distribution of the leaf surface, by preserving the underneath venation pattern and leaf shape. Results The L2L algorithm generates synthetic images of leaves with meaningful and realistic appearance, indicating that it can significantly contribute to expand a small dataset of real images. The performance was assessed qualitatively and quantitatively, by employing a DL anomaly detection strategy which quantifies the anomaly degree of synthetic leaves with respect to real samples. Finally, as an illustrative example, the proposed L2L algorithm was used for generating a set of synthetic images of healthy end diseased cucumber leaves aimed at training a CNN model for automatic detection of disease symptoms. Conclusions Generative DL approaches have the potential to be a new paradigm to provide low-cost meaningful synthetic samples. Our focus was to dispose of synthetic leaves images for smart agriculture applications but, more in general, they can serve for all computer-aided applications which require the representation of vegetation. The present L2L approach represents a step towards this goal, being able to generate synthetic samples with a relevant qualitative and quantitative resemblance to real leaves.
Why it matches plant phenotyping methods植物葉画像を生成し、葉形状・葉脈・表面色を再現する画像生成手法を開発・評価しており、植物フェノタイピング関連の画像解析ワークフローが中心である。
abstractwe implement Deep Learning (DL) techniques for the automatic generation of meaningful synthetic images of plant leaves
Reproduction assets foundThe paper explicitly states that the authors' code and data for the Leaf2Leaf generative leaf-image synthesis pipeline are publicly available on GitHub, which is a paper-specific, actionable asset.Code · publicData Availability: The code and data for reproducibility are available on GitHub ( https://github.com/AleBenfe/Leaf2Leaf ).Open asset ↗AleBenfe/Leaf2Leaflines:135-147Code / dataset availability confirmedOpenAlex · checked 15 Sept 2026
Abstract. The SiDroForest (Siberian drone-mapped forest inventory) data collection is an attempt to remedy the scarcity of forest structure data in the circumboreal region by providing adjusted and labeled tree-level and vegetation plot-level data for machine learning and upscaling purposes. We present datasets of vegetation composition and tree and plot level forest structure for two important vegetation transition zones in Siberia, Russia; the summergreen–evergreen transition zone in Central Yakutia and the tundra–taiga transition zone in Chukotka (NE Siberia). The SiDroForest data collection consists of four datasets that contain different complementary data types that together support in-depth analyses from different perspectives of Siberian Forest plot data for multi-purpose applications. i. Dataset 1 provides unmanned aerial vehicle (UAV)-borne data products covering the vegetation plots surveyed during fieldwork (Kruse et al., 2021, https://doi.org/10.1594/PANGAEA.933263). The dataset includes structure-from-motion (SfM) point clouds and red–green–blue (RGB) and red–green–near-infrared (RGN) orthomosaics. From the orthomosaics, point-cloud products were created such as the digital elevation model (DEM), canopy height model (CHM), digital surface model (DSM) and the digital terrain model (DTM). The point-cloud products provide information on the three-dimensional (3D) structure of the forest at each plot.ii. Dataset 2 contains spatial data in the form of point and polygon shapefiles of 872 individually labeled trees and shrubs that were recorded during fieldwork at the same vegetation plots (van Geffen et al., 2021c, https://doi.org/10.1594/PANGAEA.932821). The dataset contains information on tree height, crown diameter, and species type. These tree and shrub individually labeled point and polygon shapefiles were generated on top of the RGB UVA orthoimages. The individual tree information collected during the expedition such as tree height, crown diameter, and vitality are provided in table format. This dataset can be used to link individual information on trees to the location of the specific tree in the SfM point clouds, providing for example, opportunity to validate the extracted tree height from the first dataset. The dataset provides unique insights into the current state of individual trees and shrubs and allows for monitoring the effects of climate change on these individuals in the future.iii. Dataset 3 contains a synthesis of 10 000 generated images and masks that have the tree crowns of two species of larch (Larix gmelinii and Larix cajanderi) automatically extracted from the RGB UAV images in the common objects in context (COCO) format (van Geffen et al., 2021a, https://doi.org/10.1594/PANGAEA.932795). As machine-learning algorithms need a large dataset to train on, the synthetic dataset was specifically created to be used for machine-learning algorithms to detect Siberian larch species.iv. Dataset 4 contains Sentinel-2 (S-2) Level-2 bottom-of-atmosphere processed labeled image patches with seasonal information and annotated vegetation categories covering the vegetation plots (van Geffen et al., 2021b, https://doi.org/10.1594/PANGAEA.933268). The dataset is created with the aim of providing a small ready-to-use validation and training dataset to be used in various vegetation-related machine-learning tasks. It enhances the data collection as it allows classification of a larger area with the provided vegetation classes. The SiDroForest data collection serves a variety of user communities. The detailed vegetation cover and structure information in the first two datasets are of use for ecological applications, on one hand for summergreen and evergreen needle-leaf forests and also for tundra–taiga ecotones. Datasets 1 and 2 further support the generation and validation of land cover remote-sensing products in radar and optical remote sensing. In addition to providing information on forest structure and vegetation composition of the vegetation plots, the third and fourth datasets are prepared as training and validation data for machine-learning purposes. For example, the synthetic tree-crown dataset is generated from the raw UAV images and optimized to be used in neural networks. Furthermore, the fourth SiDroForest dataset contains S-2 labeled image patches processed to a high standard that provide training data on vegetation class categories for machine-learning classification with JavaScript Object Notation (JSON) labels provided. The SiDroForest data collection adds unique insights into remote hard-to-reach circumboreal forest regions.
Why it matches plant phenotyping methodsUAV画像・点群から森林の3D構造や個体樹木の高さ・樹冠径を扱う再利用可能なデータセットを提供し、抽出結果の検証や機械学習に用いるため、植物表現型データ基盤が中心です。
abstractThe SiDroForest (Siberian drone-mapped forest inventory) data collection is an attempt to remedy the scarcity of forest structure data in the circumboreal region by providing adjusted and labeled tree-level and vegetation plot-level data for machine learning and upscaling purposes.
Reproduction assets foundThe paper is a data description paper for the SiDroForest collection; all four datasets (UAV-SfM point clouds/orthomosaics, individually labeled trees, synthetic tree-crown images, Sentinel-2 labeled patches) are published on PANGAEA with explicit public download availability.Dataset · publice future users time when attempting to classify
vegetation of central Siberian and eastern Siberian boreal forests.
5 Data availability
All four datasets of the SiDroForest data collection are published in the PANGAEA data repository and are available for download:
i.
UAV-SfM point clouds, point-cloud products, and orthoimages: https://doi.org/10.1594/PANGAEA.933263 (Kruse
et al., 2021b),
ii.
Individually labeled trees: https://doi.org/10.1594/PANGAEA.932821 (van Geffen et al., 2021c),
iii.
Synthetically created tree-crown dataset: https://doi.org/10.1594/PANGAEA.932795 (van Geffen et al., 2021a),
iv.
Sentinel-2 labeled image patches: https://doi.org/10.1594/PANGAEA.933268 (van Geffen et aOpen asset ↗PANGAEA · 10.1594/PANGAEA.933263lines:557-585Dataset · publicerian boreal forests.
5 Data availability
All four datasets of the SiDroForest data collection are published in the PANGAEA data repository and are available for download:
i.
UAV-SfM point clouds, point-cloud products, and orthoimages: https://doi.org/10.1594/PANGAEA.933263 (Kruse
et al., 2021b),
ii.
Individually labeled trees: https://doi.org/10.1594/PANGAEA.932821 (van Geffen et al., 2021c),
iii.
Synthetically created tree-crown dataset: https://doi.org/10.1594/PANGAEA.932795 (van Geffen et al., 2021a),
iv.
Sentinel-2 labeled image patches: https://doi.org/10.1594/PANGAEA.933268 (van Geffen et al., 2021b).
6 Conclusions
The circumboreal forests are covering large areas on the globe. EverOpen asset ↗PANGAEA · 10.1594/PANGAEA.932821lines:557-585Dataset · publice PANGAEA data repository and are available for download:
i.
UAV-SfM point clouds, point-cloud products, and orthoimages: https://doi.org/10.1594/PANGAEA.933263 (Kruse
et al., 2021b),
ii.
Individually labeled trees: https://doi.org/10.1594/PANGAEA.932821 (van Geffen et al., 2021c),
iii.
Synthetically created tree-crown dataset: https://doi.org/10.1594/PANGAEA.932795 (van Geffen et al., 2021a),
iv.
Sentinel-2 labeled image patches: https://doi.org/10.1594/PANGAEA.933268 (van Geffen et al., 2021b).
6 Conclusions
The circumboreal forests are covering large areas on the globe. Every new forest dataset collected, processed further, and published in a ready-to-use format for a wide range of biolOpen asset ↗PANGAEA · 10.1594/PANGAEA.932795lines:557-585Dataset · publicand orthoimages: https://doi.org/10.1594/PANGAEA.933263 (Kruse
et al., 2021b),
ii.
Individually labeled trees: https://doi.org/10.1594/PANGAEA.932821 (van Geffen et al., 2021c),
iii.
Synthetically created tree-crown dataset: https://doi.org/10.1594/PANGAEA.932795 (van Geffen et al., 2021a),
iv.
Sentinel-2 labeled image patches: https://doi.org/10.1594/PANGAEA.933268 (van Geffen et al., 2021b).
6 Conclusions
The circumboreal forests are covering large areas on the globe. Every new forest dataset collected, processed further, and published in a ready-to-use format for a wide range of biological and ecological applications is therefore quite rare and an important addition for scientific studiOpen asset ↗PANGAEA · 10.1594/PANGAEA.933268lines:557-585Code / dataset availability confirmedOpenAlex · Europe PMC · Crossref · checked 15 Sept 2026
RootAnnotation / quality controlSegmentationRoot system architecture
Accurate segmentation of root system architecture (RSA) from 2D images is an important step in studying phenotypic traits of root systems. Various approaches to image segmentation exist but many of them are not well suited to the thin and reticulated structures characteristic of root systems. The findings presented here describe an approach to RSA segmentation that takes advantage of the inherent structural properties of the root system, a segmentation network architecture we call ITErRoot. We have also generated a novel 2D root image dataset which utilizes an annotation tool developed for producing high quality ground truth segmentation of root systems. Our approach makes use of an iterative neural network architecture to leverage the thin and highly branched properties of root systems for accurate segmentation. Rigorous analysis of model properties was carried out to obtain a high-quality model for 2D root segmentation. Results show a significant improvement over other recent approaches to root segmentation. Validation results show that the model generalizes to plant species with fine and highly branched RSA's, and performs particularly well in the presence of non-root objects.
Why it matches plant phenotyping methods植物根系画像からRSAを抽出するセグメンテーション手法を開発し、データセット作成と他手法との検証・比較を行っており、植物フェノタイピング手法が中心です。
abstractAccurate segmentation of root system architecture (RSA) from 2D images is an important step in studying phenotypic traits of root systems.
Reproduction assets foundThe paper's Data availability statement provides public GitHub repositories for the authors' ITErRoot training code and the Friendly Ground Truth annotation tool used to create the paper's root segmentation ground truth. Both are paper-specific, public, and actionable. No separate phenotype image dataset deposit URL isCode · publicada First Research Excellence Fund. https://www.cfref-apogee.gc.ca/program-programme/communication_guidelines-lignes_directrices-eng.aspx . This work was also supported by the Google Cloud Platform (GCP) Research Credits Program.
Data availability
The code used to train the neural networks in this study is available on Github ( https://github.com/p2irc/ITErRoot ). The annotation tool used to create ground truth segmentations for training is available on Github ( https://github.com/p2irc/friendly_ground_truth ).
Competing interests
The authors declare no competing interests.
References
1.
Clark RT
Three-dimensional root phenotyping with a novel imaging and software platform
Plant PhysiOpen asset ↗p2irc/ITErRootlines:1379-1497Code · publicby volunteer Computer Science students with experience with other annotation tools. Friendly Ground Truth was successfully employed to generate a dataset of root images that were used to train and evaluate the segmentation network structure proposed in this work. The annotation tool has been made publicly available on GitHub ( https://github.com/p2irc/friendly_ground_truth ) for use by the community to generate root segmentation datasets.
Iterative neural network architectureOpen asset ↗p2irc/friendly_ground_truthlines:70-78Code / dataset availability confirmedOpenAlex · Europe PMC · checked 15 Sept 2026
Field / plotAnnotation / quality controlVisualization / data management
BACKGROUND: Plant breeding and crop research rely on experimental phenotyping trials. These trials generate data for large numbers of traits and plant varieties that needs to be captured efficiently and accurately to support further research and downstream analysis. Traditionally scored by hand, phenotypic data is nowadays collected using spreadsheets or specialized apps. While many solutions exist, which increase efficiency and reduce errors, none offer the same familiarity as printed field plans which have been used for decades and offer an intuitive overview over the trial setup, previously recorded data and plots still requiring scoring. RESULTS: We introduce GridScore which utilizes cutting-edge web technologies to reproduce the familiarity of printed field plans while enhancing the phenotypic data collection process by adding advanced features like georeferencing, image tagging and speech recognition. GridScore is a cross-platform open-source plant phenotyping app that combines barcode-based systems with a guided data collection approach while offering a top-down view onto the data collected in a field layout. GridScore is compared to existing tools across a wide spectrum of criteria including support for barcodes, multiple platforms, and visualizations. CONCLUSION: Compared to its competition, GridScore shows strong performance across the board offering a complete manual phenotyping experience.
Why it matches plant phenotyping methodsGridScoreは、植物表現型データの収集・可視化を目的とするオープンソースの横断的アプリであり、手動表現型計測ワークフロー自体が中心的な技術貢献です。
abstractWe introduce GridScore which utilizes cutting-edge web technologies to reproduce the familiarity of printed field plans while enhancing the phenotypic data collection process by adding advanced features like georeferencing, image tagging and speech recognition.
Reproduction assets foundThis is a software paper describing GridScore, a phenotyping data-collection app. The authors explicitly state the source code is publicly available on GitHub and a Docker container on Docker Hub, with the project home page at ics.hutton.ac.uk. No phenotype datasets or images from the paper's exemplar trials are sharedCode · publicThe source code is available on GitHub [ 19 ] and a Docker container is available on Docker Hub [ 20 ].Open asset ↗lines:182-244Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
A fundamental question in biology concerns how molecular and cellular processes become integrated during morphogenesis. In plants, characterization of 3D digital representations of organs at single-cell resolution represents a promising approach to addressing this problem. A major challenge is to provide organ-centric spatial context to cells of an organ. We developed several general rules for the annotation of cell position and embodied them in 3DCoordX, a user-interactive computer toolbox implemented in the open-source software MorphoGraphX. 3DCoordX enables rapid spatial annotation of cells even in highly curved biological shapes. Using 3DCoordX, we analyzed cellular growth patterns in organs of several species. For example, the data indicated the presence of a basal cell proliferation zone in the ovule primordium of Arabidopsis (Arabidopsis thaliana). Proof-of-concept analyses suggested a preferential increase in cell length associated with neck elongation in the archegonium of Marchantia (Marchantia polymorpha) and variations in cell volume linked to central morphogenetic features of a trap of the carnivorous plant Utricularia (Utricularia gibba). Our work demonstrates the broad applicability of the developed strategies as they provide organ-centric spatial context to cellular features in plant organs of diverse shape complexity.
Why it matches plant phenotyping methods植物器官の3Dデジタル表現から細胞位置を注釈し、細胞成長や形態特徴を解析する専用ツールを開発しており、表現型取得・抽出手法が研究の中心です。
abstractWe developed several general rules for the annotation of cell position and embodied them in 3DCoordX, a user-interactive computer toolbox implemented in the open-source software MorphoGraphX.
Reproduction assets foundThe paper deposits its phenotyping datasets (raw cell boundaries, PlantSeg predictions, segmented cells, annotated 3D cell meshes, and csv attribute files) in the BioStudies repository under accession S-BSST734, making the paper-specific 3D plant organ data publicly available.Dataset · publicThe datasets of this study have been deposited with the BioStudies data repository ( https://www.ebi.ac.uk/biostudies ) under the accession S-BSST734. Example dataset contains raw cell boundaries, cell boundaries, predictions from PlantSeg, nuclei images, segmented cells as well as the annotated 3D cell meshes, and the associated attribute files in csv format. The 3D meshes used in different manuscript figures are also available for download from the repository.Open asset ↗BioStudies · S-BSST734lines:161-178Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
RiceAnnotation / quality controlCalibration / preprocessingVisualization / data management
Background Developing a systematic phenotypic data analysis pipeline, creating enhanced visualizations, and interpreting the results is crucial to extract meaningful insights from data in making better breeding decisions. Here, we provide an overview of how the Rainfed Rice Breeding (RRB) program at IRRI has leveraged R computational power with open-source resource tools like R Markdown, plotly, LaTeX, and HTML to develop an open-source and end-to-end data analysis workflow and pipeline, and re-designed it to a reproducible document for better interpretations, visualizations and easy sharing with collaborators. Results We reported the state-of-the-art implementation of the phenotypic data analysis pipeline and workflow embedded into a well-descriptive document. The developed analytical pipeline is open-source, demonstrating how to analyze the phenotypic data in crop breeding programs with step-by-step instructions. The analysis pipeline shows how to pre-process and check the quality of phenotypic data, perform robust data analysis using modern statistical tools and approaches, and convert it into a reproducible document. Explanatory text with R codes, outputs either in text, tables, or graphics, and interpretation of results are integrated into the unified document. The analysis is highly reproducible and can be regenerated at any time. The analytical pipeline source codes and demo data are available at https://github.com/whussain2/Analysis-pipeline . Conclusion The analysis workflow and document presented are not limited to IRRI's RRB program but are applicable to any organization or institute with full-fledged breeding programs. We believe this is a great initiative to modernize the data analysis of IRRI's RRB program. Further, this pipeline can be easily implemented by plant breeders or researchers, helping and guiding them in analyzing the breeding trials data in the best possible way.
Why it matches plant phenotyping methods作物育種における表現型データの前処理・品質管理・統計解析・可視化を一貫して行う、再現可能なオープンソース解析パイプラインが中心である。
abstractHere, we provide an overview of how the Rainfed Rice Breeding (RRB) program at IRRI has leveraged R computational power with open-source resource tools like R Markdown, plotly, LaTeX, and HTML to develop an open-source and end-to-end data analysis workflow and pipeline
Reproduction assets foundThe paper's authors publicly release their phenotypic data analysis pipeline source codes, sample HTML workflow documents, and demo phenotypic dataset on GitHub, directly reproducing this paper's computational analysis.Code · publicThe analytical pipeline source codes and demo data are available at https://github.com/whussain2/Analysis-pipelineOpen asset ↗whussain2/Analysis-pipelinelines:1-75Dataset · publicAll the instructions, R source codes, examples, and the data sets are freely available in the GitHub repository at https://github.com/whussain2/Analysis-pipelineOpen asset ↗whussain2/Analysis-pipelinelines:80-91Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
With the ongoing cost decrease of genotyping and sequencing technologies, accurate and fast phenotyping remains the bottleneck in the utilizing of plant genetic resources for breeding and breeding research. Although cost-efficient high-throughput phenotyping platforms are emerging for specific traits and/or species, manual phenotyping is still widely used and is a time- and money-consuming step. Approaches that improve data recording, processing or handling are pivotal steps towards the efficient use of genetic resources and are demanded by the research community. Therefore, we developed PhenoApp, an open-source Android app for tablets and smartphones to facilitate the digital recording of phenotypical data in the field and in greenhouses. It is a versatile tool that offers the possibility to fully customize the descriptors/scales for any possible scenario, also in accordance with international information standards such as MIAPPE (Minimum Information About a Plant Phenotyping Experiment) and FAIR (Findable, Accessible, Interoperable, and Reusable) data principles. Furthermore, PhenoApp enables the use of pre-integrated ready-to-use BBCH (Biologische Bundesanstalt für Land- und Forstwirtschaft, Bundessortenamt und CHemische Industrie) scales for apple, cereals, grapevine, maize, potato, rapeseed and rice. Additional BBCH scales can easily be added. The simple and adaptable structure of input and output files enables an easy data handling by either spreadsheet software or even the integration in the workflow of laboratory information management systems (LIMS). PhenoApp is therefore a decisive contribution to increase efficiency of digital data acquisition in genebank management but also contributes to breeding and breeding research by accelerating the labour intensive and time-consuming acquisition of phenotyping data.
Why it matches plant phenotyping methods植物表現型データのデジタル記録・取得を目的とするオープンソースアプリの開発であり、表現型測定ワークフローが中心的です。
abstractTherefore, we developed PhenoApp, an open-source Android app for tablets and smartphones to facilitate the digital recording of phenotypical data in the field and in greenhouses.
Reproduction assets foundThe paper describes PhenoApp, an open-source Android phenotyping app. Authors provide the app's source code (Gitea, archived on Zenodo) and underlying example input/output phenotype data files on Zenodo under CC0. The SHAPE II project website is a project page, not a paper-specific data deposit, and is excluded.Code · publice ‘in’ folder of the app main directory and no additional source data is required).
-
Output_example.xls (sample output file created by PhenoApp).
Data are available under the terms of the
Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).
Software availability
Source code available from:
https://gitea.julius-kuehn.de/JKI/pheno-app
Archived source code at time of publication:
https://doi.org/10.5281/zenodo.5525779
36
License:
Apache-2.0
Acknowledgements
We are grateful to Moritz Cappel, Teresa Claus and Claudia Vogel for ongoing testing, recommendations and bug report of PhenoApp during development.
Funding Statement
This work was supported by grants fOpen asset ↗gitea.julius-kuehn.de · JKI/pheno-applines:333-433Code · publicput_example.xls (sample output file created by PhenoApp).
Data are available under the terms of the
Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).
Software availability
Source code available from:
https://gitea.julius-kuehn.de/JKI/pheno-app
Archived source code at time of publication:
https://doi.org/10.5281/zenodo.5525779
36
License:
Apache-2.0
Acknowledgements
We are grateful to Moritz Cappel, Teresa Claus and Claudia Vogel for ongoing testing, recommendations and bug report of PhenoApp during development.
Funding Statement
This work was supported by grants from the German Federal Ministry of Education and Research to FS (SelWineQ, FKZ 031B0889Open asset ↗Zenodo · 10.5281/zenodo.5525779lines:333-433Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Background The 3D point cloud is the most direct and effective data form for studying plant structure and morphology. In point cloud studies, the point cloud segmentation of individual plants to organs directly determines the accuracy of organ-level phenotype estimation and the reliability of the 3D plant reconstruction. However, highly accurate, automatic, and robust point cloud segmentation approaches for plants are unavailable. Thus, the high-throughput segmentation of many shoots is challenging. Although deep learning can feasibly solve this issue, software tools for 3D point cloud annotation to construct the training dataset are lacking. Results We propose a top-to-down point cloud segmentation algorithm using optimal transportation distance for maize shoots. We apply our point cloud annotation toolkit for maize shoots, Label3DMaize, to achieve semi-automatic point cloud segmentation and annotation of maize shoots at different growth stages, through a series of operations, including stem segmentation, coarse segmentation, fine segmentation, and sample-based segmentation. The toolkit takes ∼4-10 minutes to segment a maize shoot and consumes 10-20% of the total time if only coarse segmentation is required. Fine segmentation is more detailed than coarse segmentation, especially at the organ connection regions. The accuracy of coarse segmentation can reach 97.2% that of fine segmentation. Conclusion Label3DMaize integrates point cloud segmentation algorithms and manual interactive operations, realizing semi-automatic point cloud segmentation of maize shoots at different growth stages. The toolkit provides a practical data annotation tool for further online segmentation research based on deep learning and is expected to promote automatic point cloud processing of various plants.
Why it matches plant phenotyping methodsトウモロコシの3D点群を器官レベルに分割・注釈するツールを開発し、植物形態の表現型推定と深層学習用データ構築を技術的に支援するため、フェノタイピング手法が中心です。
abstractWe propose a top-to-down point cloud segmentation algorithm using optimal transportation distance for maize shoots.
Reproduction assets foundThe paper's authors publicly released the Label3DMaize toolkit (MATLAB source code and executable) on GitHub, which implements the paper's point cloud segmentation/annotation analysis for maize shoots. A supporting data deposit (GigaScience Database, 10.5524/100884) is cited but its URL is not among the allowed URLs,soCode · publicgmented point clouds. The segmentation algorithm and this toolkit will be extended to other crops according to their morphological characteristics, which will promote the automatic 3D point cloud segmentation of plants.
Availability of Supporting Source Code and Requirements
Project name: Label3DMaize Toolkit
Project home page: https://github.com/syau-miao/Label3DMaize.git
Source code and executable program: [ 57 ]
Operating systems: Windows
Programming languages: MATLAB
License: GNU General Public License (GPL)
RRID:SCR_021029
biotools ID: label3dmaize
Data AvailabilityOpen asset ↗Label3DMaize · syau-miao/Label3DMaizelines:277-291Code / dataset availability confirmedOpenAlex · Europe PMC · Crossref · checked 15 Sept 2026
Abstract Background:The use of 3D based high-throughput phenotyping improves theefficiency of crop management and monitoring practices. Thestructure-from-motion and multi-view stereo photogrammetry (SfM-MVS)technique, applicable to common RGB digital cameras, has been widely used forthis and can be implemented by many commercial and open-source tools. Byusing such tools, several outputs such as digital orthophoto map (DOM), digitalsurface model (DSM), and point cloud data (PCD) can be generated. However,there is a gap between these outputs and the final 3D plant phenotyping. Forexample, calculating plant height and canopy ground cover requires thesegmentation of each plot from the whole DOM, DSM, or original image. Theseintermediate processes are time-consuming, and to the best of our knowledge,there are no easy-to-use alternatives currently available. Results: In this study, a software package called EasyIDP (easy intermediatedata processor) was developed to link the products of SfM-MVS techniques with3D based plant phenotyping. A lotus (Nelumbo nucifera) breeding field was usedto demonstrate the following points: 1) clipping (segmenting) SfM-MVS productsaccording to a given plot boundary or region of interest (ROI); 2) transformingthe ROI of the SfM-MVS products into high-quality raw images to assist inobject detection; and 3) evaluating the accuracy of the previous transformationusing manual annotation. Conclusions: The proposed intermediate data processing tool showed anacceptable accuracy and potential to process the products from SfM-MVStechniques. By using the EasyIDP, a bridge between SfM-MVS products andplant phenotyping was conveniently achieved.
Why it matches plant phenotyping methodsEasyIDPはSfM-MVS生成物を植物表現型抽出へ接続する中間処理ソフトウェアとして開発・評価されており、表現型取得ワークフローが中心である。
abstractThe proposed intermediate data processing tool showed anacceptable accuracy and potential to process the products from SfM-MVStechniques.
Reproduction assets foundThe paper is a software article for EasyIDP, whose source code is publicly released on GitHub, and the authors explicitly state that the example data (UAV/SfM-MVS phenotyping case-study data) and Jupyter notebook analysis codes are available in a companion public repository (EasyIDP.paper). Both are paper-specific,公开,和Code · publica U19A2061.
479
Ethics approval and consent to participate
480
Not applicable.
481
Consent for publication
482
Not applicable.
483
Availability of data and materials
484
The download link of the example data, and Jupyter notebook codes for drawing all results figures, and the LaTeX
485
codes of this manuscript, are available on https://github.com/HowcanoeWang/EasyIDP.paper.486
Competing interests
487
The authors declare that they have no competing interests.
488
Author details
489
1
International Field Phenomics Research Laboratory, Institute for Sustainable Agro-ecosystem Services, Graduate
490
School of Agricultural and Life Science, The University of Tokyo, 188-0002 Tokyo, Japan. 2
Key LaOpen asset ↗HowcanoeWang/EasyIDP.paperpdf-raw-page:19 lines:1-164Code / dataset availability confirmedOpenAlex · checked 15 Sept 2026
Crop phenotypic data underpin many pre-breeding efforts to characterize variation within germplasm collections. Although there has been an increase in the global capacity for accumulating and comparing such data, a lack of consistency in the systematic description of metadata often limits integration and sharing. We therefore aimed to understand some of the challenges facing findable, accesible, interoperable and reusable (FAIR) curation and annotation of phenotypic data from minor and underutilized crops. We used bambara groundnut (Vigna subterranea) as an exemplar underutilized crop to assess the ability of the Crop Ontology system to facilitate curation of trait datasets, so that they are accessible for comparative analysis. This involved generating a controlled vocabulary Trait Dictionary of 134 terms. Systematic quantification of syntactic and semantic cohesiveness of the full set of 28 crop-specific COs identified inconsistencies between trait descriptor names, a relative lack of cross-referencing to other ontologies and a flat ontological structure for classifying traits. We also evaluated the Minimal Information About a Phenotyping Experiment and FAIR compliance of bambara trait datasets curated within the CropStoreDB schema. We discuss specifications for a more systematic and generic approach to trait controlled vocabularies, which would benefit from representation of terms that adhere to Open Biological and Biomedical Ontologies principles. In particular, we focus on the benefits of reuse of existing definitions within pre- and post-composed axioms from other domains in order to facilitate the curation and comparison of datasets from a wider range of crops. Database URL: https://www.cropstoredb.org/cs_bambara.html.
Why it matches plant phenotyping methods作物表現型データの標準化、Trait Dictionary、Crop Ontology、データキュレーションおよびFAIR準拠評価が研究の中心であり、再利用可能な表現型データ記述手法を扱うため。
abstractWe therefore aimed to understand some of the challenges facing findable, accesible, interoperable and reusable (FAIR) curation and annotation of phenotypic data from minor and underutilized crops.
Reproduction assets foundThe paper's bambara groundnut phenotype datasets and Trait Dictionary are publicly available: the curated bambara trait datasets are hosted in CropStoreDB, and the authors' Trait Dictionary (134 terms, crop code CO_366) is published on the Crop Ontology portal. Supplementary tables (S1–S7) contain the multi-species TD,Dataset · publicrepresentation of terms that adhere to Open Biological and Biomedical Ontologies principles. In particular, we focus on the benefits of reuse of existing definitions within pre- and post-composed axioms from other domains in order to facilitate the curation and comparison of datasets from a wider range of crops.
Database URL : https://www.cropstoredb.org/cs_bambara.html
status released display-pdf yes is-olf no is-manuscript no is-preprint no is-journal-matter no is-scanned no is-retracted no
Received 2020 Jun 27; Revised 2021 Apr 14; Accepted 2021 Apr 30; Collection date 2021.
Introduction
Technological advances in data acquisition have driven massive increases in the accumulation of crop tOpen asset ↗CropStoreDBlines:1-31Code / dataset availability confirmedCrossref · Europe PMC · checked 9 Sept 2026
Automated identification of plant diseases is very important for crop protection. Most automated approaches aim to build classification models based on leaf or fruit images. These approaches usually require the collection and annotation of many images, which is difficult and costly process especially in the case of new or rare diseases. Therefore, in this study, we developed and evaluated several methods for identifying plant diseases with little data. Convolutional Neural Networks (CNNs) are used due to their superior ability to transfer learning. Three CNN architectures (ResNet18, ResNet34, and ResNet50) were used to build two baseline models, a Triplet network and a deep adversarial Metric Learning (DAML) approach. These approaches were trained from a large source domain dataset and then tuned to identify new diseases from few images, ranging from 5 to 50 images per disease. The proposed approaches were also evaluated in the case of identifying the disease and plant species together or only if the disease was identified, regardless of the affected plant. The evaluation results demonstrated that a baseline model trained with a large set of source field images can be adapted to classify new diseases from a small number of images. It can also take advantage of the availability of a larger number of images. In addition, by comparing it with metric learning methods, we found that baseline model has better transferability when the source domain images differ from the target domain images significantly or are captured in different conditions. It achieved an accuracy of 99% when the shift from source domain to target domain was small and 81% when that shift was large and outperformed all other competitive approaches.
Why it matches plant phenotyping methods植物画像から病害状態を推定するCNN手法の開発と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手法に該当する。
titleConvolutional Neural Network for Automatic Identification of Plant Diseases with Limited Data
Reproduction assets foundThe paper's few-shot plant disease classification uses two public image datasets, both explicitly linked in the Data Availability Statement: PlantVillage (source domain) and the coffee leaf dataset (target domain). No author code or models are shared.Dataset · publicThe PlantVillage dataset is available at https://github.com/spMohanty/PlantVillage-Dataset and the Coffee dataset at https://github.com/esgario/lara2018/ .Open asset ↗spMohanty/PlantVillage-Datasetlines:741-764Dataset · publicThe PlantVillage dataset is available at https://github.com/spMohanty/PlantVillage-Dataset and the Coffee dataset at https://github.com/esgario/lara2018/ .Open asset ↗esgario/lara2018lines:741-764Code / dataset availability confirmedbioRxiv · checked 14 Sept 2026
Broad scale remote sensing promises to build forest inventories at unprecedented scales. A crucial step in this process is designing individual tree segmentation algorithms to associate pixels into delineated tree crowns. While dozens of tree delineation algorithms have been proposed, their performance is typically not compared based on standard data or evaluation metrics, making it difficult to understand which algorithms perform best under what circumstances. There is a need for an open evaluation benchmark to minimize differences in reported results due to data quality, forest type and evaluation metrics, and to support evaluation of algorithms across a broad range of forest types. Combining RGB, LiDAR and hyperspectral sensor data from the National Ecological Observatory Networks Airborne Observation Platform with multiple types of evaluation data, we created a novel benchmark dataset to assess individual tree delineation methods. This benchmark dataset includes an R package to standardize evaluation metrics and simplify comparisons between methods. The benchmark dataset contains over 6,000 image-annotated crowns, 424 field-annotated crowns, and 3,777 overstory stem points from a wide range of forest types. In addition, we include over 10,000 training crowns for optional use. We discuss the different evaluation sources and assess the accuracy of the image-annotated crowns by comparing annotations among multiple annotators as well as to overlapping field-annotated crowns. We provide an example submission and score for an open-source baseline for future methods.
Why it matches plant phenotyping methods個体樹冠の画像ベース delineation を評価する標準ベンチマークデータセットと評価用Rパッケージを構築しており、植物形態の抽出・比較手法が中心である。
abstractwe created a novel benchmark dataset to assess individual tree delineation methods.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · public375 developed an R package ( https://github.com/weecology/NeonTreeEvaluation_package) forOpen asset ↗weecology/NeonTreeEvaluation_packagepdf-page:21 lines:1-71Code / dataset availability confirmedCrossref · Europe PMC · checked 9 Sept 2026
Abstract We present a new large-scale three-fold annotated microscopy image dataset, aiming to advance the plant cell biology research by exploring different cell microstructures including cell size and shape, cell wall thickness, intercellular space, etc. in deep learning (DL) framework. This dataset includes 9,811 unstained and 6,127 stained (safranin-o, toluidine blue-o, and lugol’s-iodine) images with three-fold annotation including physical, morphological, and tissue grading based on weight, different section area, and tissue zone respectively. In addition, we prepared ground truth segmentation labels for three different tuber weights. We have validated the pertinence of annotations by performing multi-label cell classification, employing convolutional neural network (CNN), VGG16, for unstained and stained images. The accuracy has been achieved up to 0.94, while, F2-score reaches to 0.92. Furthermore, the ground truth labels have been verified by semantic segmentation algorithm using UNet architecture which presents the mean intersection of union up to 0.70. Hence, the overall results show that the data are very much efficient and could enrich the domain of microscopy plant cell analysis for DL-framework.
Why it matches plant phenotyping methodsジャガイモ塊茎の細胞形態・組織特性を対象とする大規模画像データセットを構築し、分類・セグメンテーションで検証しており、植物フェノタイピング用データ資源が中心である。
titleA large-scale optical microscopy image dataset of potato tuber for deep learning based plant cell assessment
Reproduction assets foundThe paper's potato tuber microscopy image dataset (raw stained/unstained images plus ground truth segmentation labels) is publicly deposited on figshare by the authors.Dataset · publicThis dataset is publicly available on figshare47 (https://doi.org/10.6084/m9.figshare.c.4955669) which can be
downloaded as a zip file.Open asset ↗figshare · 10.6084/m9.figshare.c.4955669pdf-page:5 lines:1-35Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
NASA’s ICESat-2space-borne photon-counting lidar mission is providing global elevation measurements that will provide significant benefits to a variety of bio-geoscience research applications. Given the novelty of elevation and the derived data products from the ICESat-2 mission, the research community needs software tools that can facilitate photon-level analyses to support product validation and development new analysis methods. Here, we describe PhotonLabeler, a free graphic user interface (GUI) for manual labeling and visualization of ICESat-2 Geolocated Photon data (ATL03). Developed in MATLAB, the GUI facilitates the reading and display of ATL03 Hierarchical Data Format (HDF) files, the manual labeling of individual photons into target classes of choice using a number of point selections tools and enables eventual saving of labeled data in ASCII format. Other capabilities include saving and loading of labeling sessions to manage labeling tasks over time. We expect labeled data generated using the application to serve two main purposes. First, serve as ground truth for validating various products from ICESat-2 mission, especially for study sites around the world that do not have existing reference datasets such as airborne lidar. Second, serve as training and validation data in the development of new algorithms for generating various ICESat-2 data products. We demonstrate the first use case through a validation case study for the land and vegetation product (ATL08), which provides canopy and terrain height estimates, over two sites. For the first site, located in northwestern Zambia, we used ICESat-2 ATL03 data acquired at night and for our second site in Texas, US, we used ATL03 data acquired during the day. The PhotonLabeler application is freely available as a compiled MATLAB binary to enable free access and utilization by interested researchers.
Why it matches plant phenotyping methodsICESat-2光子データを手動ラベリングし、植生・樹冠高を含む植物関連プロダクトの検証用データを作成するソフトウェアであり、植物状態の取得・検証ワークフローが中心的です。
abstractwe describe PhotonLabeler, a free graphic user interface (GUI) for manual labeling and visualization of ICESat-2 Geolocated Photon data (ATL03).
Reproduction assets foundThe paper's own PhotonLabeler software (a MATLAB GUI for labeling ICESat-2 ATL03 photon data, used to generate the validation measurements in the case study) is explicitly stated to be publicly available on the authors' GitHub with a compiled binary and user manual. The ATL03/ATL08 data are generic mission products, soCode · publicling sessions. A saved session file contains the state of the application at the time of saving and
253 stores input files path and parameters to enable one to pick up labeling from where they left.
254 2.2.5 Software availability
255 PhotonLabeler is available to interested scientists through our project website on GitHub
256 (https://github.com/Oht0nger/PhoLabeler/releases/tag/v1.0). The application is available as a compiled
257 binary, which one can install without a MATLAB license. The option requires a download of free
258 MATLAB runtime environment. On our GitHub page, we also provide a detailed user manual on how
259 to use the software.
260 2.3 Case study: Using manually labeled datOpen asset ↗Oht0nger/PhoLabeler · v1.0pdf-layout-page:9 lines:1-61Code / dataset availability confirmedEurope PMC · checked 9 Sept 2026
RootAnnotation / quality controlRoot system architecture
Motivation Trait data are fundamental to quantitatively describe plant form and function. Although root traits capture key dimensions related to plant responses to changing environmental conditions and effects on ecosystem processes, they have rarely been included in large-scale comparative studies and global models. For instance, root traits remain absent from nearly all studies that define the global spectrum of plant form and function. Thus, to overcome conceptual and methodological roadblocks preventing a widespread integration of root trait data into large-scale analyses we created the Global Root Trait (GRooT) Database. GRooT provides ready-to-use data by combining the expertise of root ecologists with data mobilization and curation. Specifically, we (i) determined a set of core root traits relevant to the description of plant form and function based on an assessment by experts, (ii) maximized species coverage through data standardization within and among traits, and (iii) implemented data quality checks. Main types of variables contained GRooT contains 114,222 trait records on 38 continuous root traits. Spatial location and grain Global coverage with data from arid, continental, polar, temperate, and tropical biomes. Data on root traits derived from experimental studies and field studies. Time period and grain Data recorded between 1911 and 2019 Major taxa and level of measurement GRooT includes root trait data for which taxonomic information is available. Trait records vary in their taxonomic resolution, with sub-species or varieties being the highest and genera the lowest taxonomic resolution available. It contains information for 184 sub-species or varieties, 6,214 species, 1,967 genera and 254 families. Due to variation in data sources, trait records in the database include both individual observations and mean values. Software format GRooT includes two csv file. A GitHub repository contains the csv files and a script in R to query the database.
Why it matches plant phenotyping methods植物の根形質を大規模に標準化・品質管理して提供する再利用可能なデータベースであり、植物フェノタイピング用データセットとして中心的な貢献である。
abstractwe created the Global Root Trait (GRooT) Database
Reproduction assets foundThe paper's core asset is the GRooT root trait database (two csv files) plus the authors' R script (GRooTExtraction) for querying/error-risk calculation, explicitly deposited in a public GitHub repository with a project website.Dataset · publicGRooT is public and will be maintained in a GitHub repository
(https://github.com/GRooT-Database/GRooT-Data).Open asset ↗GRooT-Database/GRooT-Datapdf-page:8 lines:1-48Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 9 Sept 2026
In order to train the neural network for plant phenotyping, a sufficient amount of training data must be prepared, which requires time-consuming manual data annotation process that often becomes the limiting step. Here, we show that an instance segmentation neural network aimed to phenotype the barley seed morphology of various cultivars, can be sufficiently trained purely by a synthetically generated dataset. Our attempt is based on the concept of domain randomization, where a large amount of image is generated by randomly orienting the seed object to a virtual canvas. The trained model showed 96% recall and 95% average Precision against the real-world test dataset. We show that our approach is effective also for various crops including rice, lettuce, oat, and wheat. Constructing and utilizing such synthetic data can be a powerful method to alleviate human labor costs for deploying deep learning-based analysis in the agricultural domain.
Why it matches plant phenotyping methods合成データとインスタンスセグメンテーションによる種子形態フェノタイピング手法を開発し、実画像で性能検証しているため、方法が研究の中心である。
abstractan instance segmentation neural network aimed to phenotype the barley seed morphology of various cultivars
Reproduction assets foundThe authors publicly release both the synthetic and real-world seed image datasets and the analysis code (Mask R-CNN deployment and multivariate analysis notebooks) via their GitHub repository, explicitly stated in Data availability and Code availability sections.Dataset · publicSynthetically generated and real-world datasets can be obtained from the following GitHub repository ( https://github.com/totti0223/crop_seed_instance_segmentation ).Open asset ↗https://github.com/totti0223/crop_seed_instance_segmentationlines:149-171Code · publicCode to reproduce the deployment of the trained Mask R-CNN and multivariate analysis is formatted as IPython notebooks and can also be obtained from the GitHub repository ( https://github.com/totti0223/crop_seed_instance_segmentation ).Open asset ↗https://github.com/totti0223/crop_seed_instance_segmentationlines:149-171Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 9 Sept 2026
Image-based plant phenotyping has been steadily growing and this has steeply increased the need for more efficient image analysis techniques capable of evaluating multiple plant traits. Deep learning has shown its potential in a multitude of visual tasks in plant phenotyping, such as segmentation and counting. Here, we show how different phenotyping traits can be extracted simultaneously from plant images, using Multi-Task Learning (MTL). MTL leverages information contained in the training images of related tasks to improve overall generalization and learns models with fewer labels. We present a Multi-Task Deep Learning framework for plant phenotyping, able to infer three traits simultaneously: (i) leaf count; (ii) projected leaf area (PLA); and (iii) genotype classification. We adopted a modified ResNet50 as a feature extractor, trained end-to-end to predict multiple traits. We also leverage MTL to show that through learning from more easily obtainable annotations (such as PLA and genotype) we can predict a better leaf count (harder to obtain annotation). We evaluate our findings on several publicly available datasets of top-view images of Arabidopsis thaliana. Experimental results show that the proposed MTL method improves the leaf count Mean Squared Error (MSE) by more than 40 %, compared to a single task network on the same dataset. We also show that our MTL framework can be trained with up to 75 % fewer leaf count annotations without significantly impacting performance, whereas a single task model shows a steady decline when fewer annotations are available.
Why it matches plant phenotyping methods植物画像から複数形質を同時推定するマルチタスク深層学習手法の開発・評価が中心であり、明確な植物フェノタイピング方法論研究である。
abstractWe present a Multi-Task Deep Learning framework for plant phenotyping, able to infer three traits simultaneously: (i) leaf count; (ii) projected leaf area (PLA); and (iii) genotype classification.
Reproduction assets foundThe paper's authors provide public analysis code (MTL phenotyping framework) on GitHub, and the study analyzes publicly available CVPPP plant image datasets (Ara2013, A1, A4) hosted on plant-phenotyping.org. Both are paper-specific, public, and actionable.Code · publicCode available at https://github.com/andobrescu/Multi_task_plant_phenotyping .Open asset ↗andobrescu/Multi_task_plant_phenotypinglines:224-295Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://www.plant-phenotyping.org/CVPPP2017-challenge .Open asset ↗lines:607-694Code / dataset availability confirmedOpenAlex · Europe PMC · checked 9 Sept 2026
Natural language descriptions of plant phenotypes are a rich source of information for genetics and genomics research. We computationally translated descriptions of plant phenotypes into structured representations that can be analyzed to identify biologically meaningful associations. These representations include the entity-quality (EQ) formalism, which uses terms from biological ontologies to represent phenotypes in a standardized, semantically rich format, as well as numerical vector representations generated using natural language processing (NLP) methods (such as the bag-of-words approach and document embedding). We compared resulting phenotype similarity measures to those derived from manually curated data to determine the performance of each method. Computationally derived EQ and vector representations were comparably successful in recapitulating biological truth to representations created through manual EQ statement curation. Moreover, NLP methods for generating vector representations of phenotypes are scalable to large quantities of text because they require no human input. These results indicate that it is now possible to computationally and automatically produce and populate large-scale information resources that enable researchers to query phenotypic descriptions directly.
Why it matches plant phenotyping methods植物表現型記述をNLPで構造化・ベクトル化し、手動キュレーションとの性能比較で検証する計算手法が中心である。
abstractWe computationally translated descriptions of plant phenotypes into structured representations that can be analyzed to identify biologically meaningful associations.
Reproduction assets foundThe paper's data availability statement explicitly deposits the authors' analysis code on GitHub (irbraun/phenologs) and all files needed to reproduce the results (including the phenotype/EQ datasets used) on Zenodo (doi 10.5281/zenodo.3255020). These are paper-specific, public, and actionable assets for the phenotype-Code · publicThe code used to produce the results of this work is available at github.com/irbraun/phenologs . Files necessary to reproduce the discussed results, datasets used to generate figures presented in this work, and other supplemental files are available at doi.org/10.5281/zenodo.3255020 .Open asset ↗irbraun/phenologs · 10.5281/zenodo.3255020lines:792-814Dataset · publicFiles necessary to reproduce the discussed results, datasets used to generate figures presented in this work, and other supplemental files are available at doi.org/10.5281/zenodo.3255020 . This data repository also includes versions of the previously described datasets available as supplemental data of Oellrich, Walls et al. (2015) and Lloyd and Meinke (2012) , for the purpose of making this study reproducible without any additional external files.Open asset ↗10.5281/zenodo.3255020lines:792-814Code / dataset availability confirmedOpenAlex · checked 15 Sept 2026
Abstract High‐throughput phenotyping systems provide abundant data for statistical analysis through plant imaging. Before usable data can be obtained, image processing must take place. In this study, we used supervised learning methods to segment plants from the background in such images and compared them with commonly used thresholding methods. Because obtaining accurate training data is a major obstacle to using supervised learning methods for segmentation, a novel approach to producing accurate labels was developed. We demonstrated that, with careful selection of training data through such an approach, supervised learning methods, and neural networks in particular, can outperform thresholding methods at segmentation.
Why it matches plant phenotyping methods植物画像から背景を分離するセグメンテーション手法を開発・比較し、教師データ生成法も提案しているため、表現型取得の技術が中心である。
abstractIn this study, we used supervised learning methods to segment plants from the background in such images and compared them with commonly used thresholding methods.
Reproduction assets foundThe paper's DATA AVAILABILITY statement points to the authors' public GitHub repository containing all segmentation analysis code and related data, and to CyVerse Data Commons hosting the raw maize image data used in the study.Code · publicdata were obtained for the rest of the plant. In more challeng-
ing cases of plant segmentation, such as the field environment,
our method serves as an excellent starting point, and its perfor-
mance could be improved by incorporating more training data.
DATA AVAILABILITY
All code along with related data are posted on Github at
https://github.com/jasonradams47/PlantSegmentationCode.The raw image data used in this study are hosted at CyVerse
(Liang & Schnable, 2017).
CONFLICT OF INTEREST
The authors have no competing financial interests.
ORCID
Jason Adams https://orcid.org/0000-0003-2085-4911
Yumou Qiu https://orcid.org/0000-0003-4846-1263
Yuhang Xu https://orcid.org/0000-0003-4351-4602
JamesOpen asset ↗jasonradams47/PlantSegmentationCodepdf-raw-page:10 lines:1-83Dataset · publicThe raw image data used in this study are hosted at CyVerse
(Liang & Schnable, 2017).Open asset ↗pdf-raw-page:10 lines:1-83Code / dataset availability confirmedEurope PMC · checked 9 Sept 2026
Genebanks are valuable sources of genetic diversity, which can help to cope with future problems of global food security caused by a continuously growing population, stagnating yields and climate change. However, the scarcity of phenotypic and genotypic characterization of genebank accessions severely restricts their use in plant breeding. To warrant the seed integrity of individual accessions during periodical regeneration cycles in the field phenotypic characterizations are performed. This study provides non-orthogonal historical data of 12,754 spring and winter wheat accessions characterized for flowering time, plant height, and thousand grain weight during 70 years of seed regeneration at the German genebank. Supported by historical weather observations outliers were removed following a previously described quality assessment pipeline. In this way, ready-to-use processed phenotypic data across regeneration years were generated and further validated. We encourage international and national genebanks to increase their efforts to transform into bio-digital resource centers. A first important step could consist in unlocking their historical data treasures that allows an educated choice of accessions by scientists and breeders.
Why it matches plant phenotyping methods7 दशकにわたるコムギ表現型データを大規模に整理・品質評価・検証し、再利用可能な処理済みデータとして提供することが中心であり、植物フェノタイピングデータセットとして適格です。
abstractThis study provides non-orthogonal historical data of 12,754 spring and winter wheat accessions characterized for flowering time, plant height, and thousand grain weight during 70 years of seed regeneration at the German genebank.
Reproduction assets foundThe paper deposits its historical wheat phenotypic data (FT, PH, TGW for 12,754 accessions), outlier-corrected and BLUE-processed datasets, and example R analysis scripts in the e!DAL-PGP repository under DOI 10.5447/IPK/2019/11, which is an allowed URL and appears verbatim in the text.Dataset · publicPhilipp, N. et al. Historical phenotypic data from seven decades of seed regeneration in a wheat ex situ collection hosted at the
Leibniz Institute of Plant Genetics and Crop Plant Research (IPK). e!DAL - Plant Genomics and Phenomics Research Data Repository,
https://doi.org/10.5447/IPK/2019/11 (2019).Open asset ↗e!DAL - Plant Genomics and Phenomics Research Data Repository · 10.5447/IPK/2019/11pdf-page:9 lines:1-56Code / dataset availability confirmedEurope PMC · OpenAlex · checked 15 Sept 2026
Annotation / quality controlVisualization / data management
GnpIS is a data repository for plant phenomics that stores whole field and greenhouse experimental data including environment measures. It allows long-term access to datasets following the FAIR principles: Findable, Accessible, Interoperable, and Reusable, by using a flexible and original approach. It is based on a generic and ontology driven data model and an innovative software architecture that uncouples data integration, storage, and querying. It takes advantage of international standards including the Crop Ontology, MIAPPE, and the Breeding API. GnpIS allows handling data for a wide range of species and experiment types, including multiannual perennial plants experimental network or annual plant trials with either raw data, i.e., direct measures, or computed traits. It also ensures the integration and the interoperability among phenotyping datasets and with genotyping data. This is achieved through a careful curation and annotation of the key resources conducted in close collaboration with the communities providing data. Our repository follows the Open Science data publication principles by ensuring citability of each dataset. Finally, GnpIS compliance with international standards enables its interoperability with other data repositories hence allowing data links between phenotype and other data types. GnpIS can therefore contribute to emerging international federations of information systems.
Why it matches plant phenotyping methods植物フェノミクスデータリポジトリの設計・標準化・相互運用性を扱う方法論的研究であり、フェノタイピングデータ基盤が中心です。
abstractGnpIS is a data repository for plant phenomics that stores whole field and greenhouse experimental data including environment measures.
Reproduction assets foundThis is an infrastructure/data-management paper describing the GnpIS phenotyping repository rather than a single measurement study. The paper-specific public assets are the GnpIS repository itself (hosting the curated phenotyping trial datasets the paper describes), the authors' public ontology versioning repository onDataset · publicGnpIS provides phenotyping data discovery capabilities and data aggregation among several datasets. The dedicated query form, available in the phenotyping section of GnpIS ( https://urgi.versailles.inra.fr/gnpis/ ), is based on three tabs: (i) “Genotype” for filtering the plant material by species, genetic panel, and collections, (ii) “Observation variables” that allows variables selection using a Breeding API compliant open source widget ( https://github.com/gnpis/trait-ontology-widget ), and (iii) “Trial” that contains filters for genOpen asset ↗GnpISlines:103-109Code / dataset availability confirmedEurope PMC · OpenAlex · Crossref · checked 15 Sept 2026
Background High-quality plant phenotyping and climate data lay the foundation for phenotypic analysis and genotype-environment interaction, providing important evidence not only for plant scientists to understand the dynamics between crop performance, genotypes, and environmental factors but also for agronomists and farmers to closely monitor crops in fluctuating agricultural conditions. With the rise of Internet of Things technologies (IoT) in recent years, many IoT-based remote sensing devices have been applied to plant phenotyping and crop monitoring, which are generating terabytes of biological datasets every day. However, it is still technically challenging to calibrate, annotate, and aggregate the big data effectively, especially when they were produced in multiple locations and at different scales. Findings CropSight is a PHP Hypertext Pre-processor and structured query language-based server platform that provides automated data collation, storage, and information management through distributed IoT sensors and phenotyping workstations. It provides a two-component solution to monitor biological experiments through networked sensing devices, with interfaces specifically designed for distributed plant phenotyping and centralized data management. Data transfer and annotation are accomplished automatically through an hypertext transfer protocol-accessible RESTful API installed on both device side and server side of the CropSight system, which synchronize daily representative crop growth images for visual-based crop assessment and hourly microclimate readings for GxE studies. CropSight also supports the comparison of historical and ongoing crop performance while different experiments are being conducted. Conclusions As a scalable and open-source information management system, CropSight can be used to maintain and collate important crop performance and microclimate datasets captured by IoT sensors and distributed phenotyping installations. It provides near real-time environmental and crop growth monitoring in addition to historical and current experiment comparison through an integrated cloud-ready server system. Accessible both locally in the field through smart devices and remotely in an office using a personal computer, CropSight has been applied to field experiments of bread wheat prebreeding since 2016 and speed breeding since 2017. We believe that the CropSight system could have a significant impact on scalable plant phenotyping and IoT-style crop management to enable smart agricultural practices in the near future.
Why it matches plant phenotyping methods分散型植物フェノタイピングのデータ収集・管理プラットフォームを開発し、センサーと画像による作物成長評価を統合しているため、方法が中心的である。
abstractCropSight is a PHP Hypertext Pre-processor and structured query language-based server platform that provides automated data collation, storage, and information management through distributed IoT sensors and phenotyping workstations.
Reproduction assets foundThe paper's authors publicly released the CropSight system source code (the software used for the paper's distributed plant phenotyping and IoT crop management) on GitHub under a BSD-3-Clause license, and Additional File 2 contains Python code to replicate the paper's plotted figures with datasets available in the sameCode · publicsimilar subsampling idea can be expanded to a larger and multi-site level, which can then truly help inform decision in crop research and agricultural practices across a country's arable land.
Availability of source code and requirements
Project name: CropSight for wheat prebreeding in Designing Future Wheat
Project home page: https://github.com/Crop-Phenomics-Group/cropsight/releases [ 35 ]
Operating system(s): Platform independent
Programming language: Python, PHP, JavaScript, SQL
Requirements: Apache (or other PHP5+) server, MySQL (or other SQL) server, a recent version of Chrome, Firefox, or Safari
License: BSD-3-Clause available at https://opensource.org/licenses/BSD-3-Clause
RRID:SCR_0Open asset ↗Crop-Phenomics-Group/cropsightlines:79-115Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 10 Sept 2026
The accuracy of machine learning tasks critically depends on high quality ground truth data. Therefore, in many cases, producing good ground truth data typically involves trained professionals; however, this can be costly in time, effort, and money. Here we explore the use of crowdsourcing to generate a large number of training data of good quality. We explore an image analysis task involving the segmentation of corn tassels from images taken in a field setting. We investigate the accuracy, speed and other quality metrics when this task is performed by students for academic credit, Amazon MTurk workers, and Master Amazon MTurk workers. We conclude that the Amazon MTurk and Master Mturk workers perform significantly better than the for-credit students, but with no significant difference between the two MTurk worker types. Furthermore, the quality of the segmentation produced by Amazon MTurk workers rivals that of an expert worker. We provide best practices to assess the quality of ground truth data, and to compare data quality produced by different sources. We conclude that properly managed crowdsourcing can be used to establish large volumes of viable ground truth data at a low cost and high quality, especially in the context of high throughput plant phenotyping. We also provide several metrics for assessing the quality of the generated datasets.
Why it matches plant phenotyping methods植物フェノタイピング画像のタッセル分割について、クラウドソーシングによる教師データ生成の品質・速度・評価指標を検証しており、表現型取得ワークフローが中心です。
abstractHere we explore the use of crowdsourcing to generate a large number of training data of good quality.
Reproduction assets foundThe paper's crowdsourced corn tassel bounding-box annotations (phenotyping measurements) are publicly deposited on figshare, and the analysis software is on GitHub, both explicitly stated in the Data Availability statement and Methods.Dataset · publicData Availability: The software for this project is available from: https://github.com/ashleyzhou972/Crowdsource-Corn-Tassels . The data for this project are available from: https://doi.org/10.6084/m9.figshare.6360236.v2 .Open asset ↗figshare · 10.6084/m9.figshare.6360236.v2lines:201-240Code · publicThe software for this study is available from: https://github.com/ashleyzhou972/Crowdsource-Corn-TasselsOpen asset ↗github · ashleyzhou972/Crowdsource-Corn-Tasselslines:257-277Code / dataset availability confirmedCrossref · OpenAlex · Europe PMC · checked 10 Sept 2026
BACKGROUND: Image-based plant phenotyping has become a powerful tool in unravelling genotype-environment interactions. The utilization of image analysis and machine learning have become paramount in extracting data stemming from phenotyping experiments. Yet we rely on observer (a human expert) input to perform the phenotyping process. We assume such input to be a 'gold-standard' and use it to evaluate software and algorithms and to train learning-based algorithms. However, we should consider whether any variability among experienced and non-experienced (including plain citizens) observers exists. Here we design a study that measures such variability in an annotation task of an integer-quantifiable phenotype: the leaf count. RESULTS: to measure intra- and inter-observer variability in a controlled study using specially designed annotation tools but also citizens using a distributed citizen-powered web-based platform. In the controlled study observers counted leaves by looking at top-view images, which were taken with low and high resolution optics. We assessed whether the utilization of tools specifically designed for this task can help to reduce such variability. We found that the presence of tools helps to reduce intra-observer variability, and that although intra- and inter-observer variability is present it does not have any effect on longitudinal leaf count trend statistical assessments. We compared the variability of citizen provided annotations (from the web-based platform) and found that plain citizens can provide statistically accurate leaf counts. We also compared a recent machine-learning based leaf counting algorithm and found that while close in performance it is still not within inter-observer variability. CONCLUSIONS: While expertise of the observer plays a role, if sufficient statistical power is present, a collection of non-experienced users and even citizens can be included in image-based phenotyping annotation tasks as long they are suitably designed. We hope with these findings that we can re-evaluate the expectations that we have from automated algorithms: as long as they perform within observer variability they can be considered a suitable alternative. In addition, we hope to invigorate an interest in introducing suitably designed tasks on citizen powered platforms not only to obtain useful information (for research) but to help engage the public in this societal important problem.
Why it matches plant phenotyping methods画像ベース植物フェノタイピングにおける葉数アノテーションの観察者間・内変動を、専用ツール、市民参加型基盤、機械学習アルゴリズムと比較検証しており、測定手法の技術的妥当性評価が中心である。
abstractHere we design a study that measures such variability in an annotation task of an integer-quantifiable phenotype: the leaf count.
Reproduction assets foundThe article states that the Arabidopsis image dataset used for the leaf-counting observer-variability study is publicly available at the plant-phenotyping.org datasets page, and the citizen-science annotations were collected via the authors' public Zooniverse 'Leaf Targeting' project. No author analysis code or trainedDataset · publicAvailability of data and materials
The image dataset used in this article is available at http://www.plant-phenotyping.org/datasets .Open asset ↗plant-phenotyping.orglines:307-379Dataset · publicThe A data (RPi) were included as part of a larger citizen-powered study (“Leaf Targeting”, available at https://www.zooniverse.org/projects/venchen/leaf-targeting ) built on ZooniverseOpen asset ↗Zooniverse · venchen/leaf-targetinglines:132-149