Accurate litchi counting from whole-tree images is essential for yield estimation, orchard management, and plant phenotyping, but remains challenging in real orchards because fruits occur in dense, heavily occluded clusters and vary markedly in scale, illumination, and appearance across ripening stages, particularly when green fruits resemble surrounding foliage. Existing methods have shown promise, but their robustness in complex orchard environments remains limited. To address these challenges, we propose FG-LCNet, a two-stage foreground-guided litchi counting framework. In the first stage, an enhanced fruit-cluster detector improves the localization of small and ambiguous clusters under complex canopy backgrounds. In the second stage, the detected foreground regions are fed into a density-regression network with hybrid attention, while a consistency-based training strategy is introduced to improve robustness to appearance and illumination variations. To support this study, a large-scale litchi counting dataset was established, consisting of 1,126 whole-tree images collected from five orchards and spanning three ripening stages, with approximately 120,000 fruit-level dot annotations and more than 20,000 cluster-level bounding boxes. FG-LCNet achieved the best overall counting performance, with an MAE of 7.44 and an RMSE of 11.01. It showed clear advantages in high-density fruit-cluster scenarios and cross-orchard validation, while maintaining competitive results across orchard-region and maturity-stage subsets. The framework further retained inference efficiency suitable for practical deployment. These results indicate that FG-LCNet provides an effective solution for robust litchi counting and offers potential for other clustered fruit-counting tasks.
Why it matches plant phenotyping methods果実数という植物器官形質を whole-tree 画像から推定する二段階画像解析手法を開発し、データセット構築と交差果樹園検証まで行っており、表現型取得・抽出法が中心である。
abstractwe propose FG-LCNet, a two-stage foreground-guided litchi counting framework.
Reproduction assets foundThe paper's implementation code is explicitly stated as publicly available at the authors' GitHub repository (FG-LCNet). The litchi counting dataset (1,126 whole-tree images with ~120,000 dot annotations and 20,000+ bounding boxes) is not yet fully public: a ~100-image annotated subset is promised upon acceptance, and,Code · publicdustry Technology Research System (CARS-32-21), Hainan Modern Agricul-
655
tural Industry Technology System (HNARS-08-G02).
656
Conflicts of Interest
657
The authors declare that there is no conflict of interest regarding the publication of this article.
658
Data Availability
659
The implementation code is publicly available at https://github.com/johnhamtom/FG-LCNet .
660
Upon acceptance, a representative subset of approximately 100 annotated litchi images will be
661
released to support reproducibility and preliminary benchmarking. The full dataset is being further
662
organized for future release. Before full release, the complete dataset can be obtained from the
663
corresponding authorOpen asset ↗johnhamtom/FG-LCNetpdf-raw-page:28 lines:1-81Plant phenotyping relevance match · UnverifiedOpenAlex · checked 15 Sept 2026
Automated fruit counting and yield estimation systems are necessary for efficient orchard management. This study presents a computer vision system based on video multi-object-tracking for fruit load estimation in apple orchards and provides a comprehensive analysis of the system performance under diverse scanning conditions. The system integrates fruit detection, tracking, localization within orchard, and fruit load map generation. Experiments were carried out in an experimental apple orchard containing 420 apple trees. Data was collected with two different RGB-D sensors (Azure Kinect DK and ZED 2) at three different scanning distances (125 cm, 175 cm, and 225 cm) on two different dates prior to the harvest. Comparing the two evaluated sensors, Azure Kinect provided more consistent performance across different dates. Results also show that the longer scanning distance improves accuracy due to seeing the full tree view gives better fruit counts than close partial views. Between the two dates, best results were achieved near harvest due to fruit color at this stage, achieving a Mean Absolute Percentage Error (MAPE) of 6.91 % and a determination coefficient (R 2 ) of 0.733 (using ZED2 sensor at 225 cm distance). Finally, a test comparing scanning from one or both sides of the tree row showed that bilateral scanning improved fruit load estimation at the stretch level by incorporating information from both sides of the canopy. The results of this work demonstrate the effectiveness of the video fruit tracking systems as a useful tool for automating fruit load estimation.
Why it matches plant phenotyping methods動画ベースの果実検出・追跡手法を開発・評価し、リンゴ樹の果実負荷量を推定すること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含める。
abstractThis study presents a computer vision system based on video multi-object-tracking for fruit load estimation in apple orchards and provides a comprehensive analysis of the system performance under diverse scanning conditions.
Fruit quality is a critical determinant of economic returns in pear production, and maintaining an appropriate fruit load (FL) is essential for achieving high yield and quality. As a direct indicator of canopy photosynthetic capacity and assimilate supply, leaf number constitutes the key biological basis of reasonable FL determination under the leaf-to-fruit ratio concept. However, accurate and efficient estimation of leaf number in mature pear trees remains technically challenging, limiting its practical use in precision FL regulation. Here, we propose a data-driven framework for leaf number and reasonable FL estimation by integrating 3D point cloud-derived canopy structure with machine learning. A pipeline for extracting 3D architectural traits was developed and implemented in the software tool FTPCT, enabling rapid and standardized trait acquisition. Through correlation analysis, multicollinearity diagnosis, and variance inflation factor screening, five key traits strongly associated with leaf number were identified and incorporated into five machine learning models optimized using Bayesian optimization. Among them, the optimized random forest regression model achieved the highest and most stable performance, with R 2 of 0.85, RMSE of 239.74, and MAE of 149.26 for test dataset. SHAP analysis identified tree crown volume as the dominant contributor to leaf number estimation. Field validation demonstrated that FL regulation guided by the proposed framework significantly improved fruit weight and size without reducing yield compared with conventional practices. Notably, the proposed approach avoids explicit leaf-level reconstruction and relies on less canopy-scale traits, substantially reducing data requirements and computational cost, and thereby offering strong potential for rapid, field-deployable FL regulation in large-scale orchards.
Why it matches plant phenotyping methods3D点群から樹冠構造形質を抽出し、葉数と適正着果量を推定する手法およびソフトウェアを開発・検証しており、植物表現型取得が中心である。
abstractA pipeline for extracting 3D architectural traits was developed and implemented in the software tool FTPCT, enabling rapid and standardized trait acquisition.
Reproduction assets foundThe paper's phenotyping analysis assets are the authors' publicly released LeafNumPred source code and trained models, and the FTPCT software for 3D trait extraction from pear tree point clouds. Phenotype/point-cloud datasets are only available on request.Code · public. Supplementary data
The following is the Supplementary data to this article:
Multimedia component 1
mmc1.docx (1.6MB, docx)
Data availability
Data will be made available on request. Anyone who wants to obtain other public data can contact us at taost@njau.edu.cn. The source codes and models have been made publicly available at https://github.com/Zhang-Fanhang/LeafNumPred, and the FTPCT software has been released at https://github.com/Zhang-Fanhang/FTPCT/tree/Installation-package.
References
1.Tao S., Khanizadeh S., Zhang H., Zhang S. Anatomy, ultrastructure and lignin distribution of stone cells in two Pyrus species. Plant Sci. 2009;176:413–419. [Google Scholar]
2.Zhang F., Wang Q., Yuan K.Open asset ↗Zhang-Fanhang/LeafNumPredhtml-lines:284-315Code · publical variations [34,35]. The method for calculating these traits are shown in the Supplementary information 1.
2.5. Software implementation for 3D trait extraction (FTPCT)
To facilitate efficient and standardized extraction of canopy structural traits from point cloud data, we used a standalone software tool, FTPCT (available at: https://github.com/Zhang-Fanhang/FTPCT/tree/Installation-package), which integrates the trait extraction procedures applied in this study. The software provides a graphical user interface, enabling users to process tree-level point cloud data and extract key 3D structural traits without requiring advanced programming skills.
FTPCT implements a series of predefined proOpen asset ↗Zhang-Fanhang/FTPCThtml-lines:138-149Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
Plant counting functions as a central role in the quantification system of plant phenotyping. Over the years, the field has evolved through successive technological paradigms, and its modern form has been strongly influenced by advances in visual counting from computer vision. In this review, we synthesize the historical origins of plant counting, assess its current fragmented landscape, and outline a roadmap toward standardized, universal plant counting systems. We propose a coherent conceptual framework— the four-level hierarchy of plant counting , which characterizes the environment, platform, sensor, and counting entity in biological organization —grounded in the need of high-throughput plant phenotyping. This framework aims to guide the development of plant counting systems that are not only accurate on individual dataset, but also reusable, comparable, and trustworthy across modern plant phenotyping scenarios. We argue that existing plant counting approaches are constrained by species-specific designs, sensor-dependent assumptions, and local, region-bound assessments. We hope this review can serve as a reference for building cross-species, cross-modal, and cross-scale visual plant counting systems.
Why it matches plant phenotyping methods植物フェノタイピングにおける画像ベースの植物個体数計測を主題とし、手法の歴史、現状評価、標準化・汎用化の枠組みと開発指針を扱うレビューであるため。
abstractPlant counting functions as a central role in the quantification system of plant phenotyping.
Cauliflower emergence rate and seedling growth are key indicators of field conditions and varietal potential. Traditional manual surveys are unsuitable for continuous monitoring across multiple varieties. This study integrates UAV RGB imagery with the DualSlim-YOLO model to estimate cauliflower emergence rates and monitor seedling growth. Built on YOLOv11, the model incorporates a lightweight feature extraction structure and an optimized detection-scale configuration. It reduces computational complexity while maintaining detection accuracy, thereby improving the efficiency of cauliflower seedling detection. DualSlim-YOLO achieved P, R, F1-score, mAP@0.5, and mAP@0.5:0.95 of 95.35%, 96.75%, 96.05%, 98.55%, and 86.65%, respectively. The number of parameters was reduced by 38.61%, while the inference speed increased by 22.16%, demonstrating good lightweight performance. Based on this model, UAV images of 171 cauliflower varieties acquired at 7, 21, and 28 d after transplanting were used for seedling detection and emergence rate estimation. In addition, 18 time-series seedling phenotypic traits were extracted, enabling a comprehensive quantitative evaluation of emergence dynamics and early-growth performance across multiple cauliflower varieties. This method effectively screens cauliflower varieties for high emergence rates, rapid emergence, and excellent seedling growth performance. It provides technical support for high-throughput, nondestructive seedling phenotyping and early germplasm screening under field conditions.
Why it matches plant phenotyping methodsUAV画像と軽量YOLOモデルを用いて、カリフラワー苗の検出、出芽率推定、18種類の時系列表現型形質抽出を行う手法が中心であり、モデル性能も検証している。
abstractThis study integrates UAV RGB imagery with the DualSlim-YOLO model to estimate cauliflower emergence rates and monitor seedling growth.
Existing conversational plant-phenotyping platforms are difficult for plant scientists to use and lack the reliability scientific research demands: failed analyses are reported as valid measurements rather than flagged as missing, statistical tests run without checking assumptions, predictions carry no uncertainty estimate, and specialised hardware limits accessibility. We present PhenoIntel, a lifecycle-aligned multi-agent web platform that turns the full machine-learning workflow into a reliable, user-friendly phenotyping system. Nine specialised agents divide the analysis into stages, from image collection through model selection, inference, and reporting, rather than handing the whole task to one AI manager. Independent checks separate these stages, and every agent reads from and writes to one shared, fixed-structure record, so an inconsistent output from one stage is caught before it reaches the next. Uncertainty is matched to each model family, conformal prediction, detection-confidence spread, or Monte Carlo Dropout, rather than applied uniformly, and quality thresholds adapt to crop and task instead of one global cutoff. When no suitable model exists, PhenoIntel can propose, validate, and integrate a new one on its own. The model repository spans ten trained models across five crops and four imaging modalities. Classification models reach Macro F1 of 0.78-0.996; object-detection models reach 0.96 mAP@50 with a 54% reduction in counting error over an unoptimised baseline; and a temporal model reaches held-out Macro F1 of 0.7050. PhenoIntel runs in a browser on standard hardware, requiring no GPU, and a 1,200-test automated suite confirms complete pipeline execution. Every result carries calibrated uncertainty, validated statistics, and FAIR-compliant provenance, a combination existing conversational phenotyping tools do not offer.
Why it matches plant phenotyping methods植物フェノタイピングの画像収集から推論・報告までを扱うウェブプラットフォームを開発し、複数モデル、精度、不確実性、検証スイートを評価しており、方法が研究の中心である。
abstractWe present PhenoIntel, a lifecycle-aligned multi-agent web platform that turns the full machine-learning workflow into a reliable, user-friendly phenotyping system.
Reproduction assets found論文固有の解析コードとモデル資産を公開するGitHubリポジトリを本文中の根拠とともに確認しました。Code · publiccode, model checkpoints, and the 1,200-test automated suite referenced
throughout this paper are maintained in a version-controlled
repository, available at
https://github.com/Naren1704/PhenoIntel-InternshipOpen asset ↗Naren1704/PhenoIntel-Internshiplines:2047-2163Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · checked 5 Sept 2026
Agricultural operations such as pepper harvesting, fruit counting, and field phenotyping rely on accurate visual recognition and instance segmentation algorithms. However, pepper fruits in complex field environments often exhibit slender and curved shapes, partial occlusion, ambiguous boundaries, and adhesion between adjacent instances. Existing object detection and instance segmentation methods therefore struggle to obtain complete fruit masks, which adversely affects subsequent fruit counting, contour measurement, and picking-point localization. To improve the instance segmentation accuracy of occluded peppers in complex field scenes, this study proposes OccPepSeg-YOLO, an improved model based on YOLO11n-seg. First, a P2FreqFusion module is introduced to fuse shallow, high-resolution detail features with deep semantic features, thereby enhancing the representation of fruit edges and tip regions. Second, an ASC module is designed to model the directional and scale-related morphological characteristics of pepper fruits, while a BoundaryGate module strengthens responses at occlusion interfaces and boundaries between adjacent instances. Finally, an OccPepSegment multi-scale prototype segmentation head is constructed, and a BDoU loss function is introduced to improve the boundary consistency of instance masks. Experiments on a self-constructed field-pepper instance segmentation dataset showed that OccPepSeg-YOLO achieved M-P, M-R, M-mAP50, and M-mAP50–95 values of 93.87%, 92.09%, 97.17%, and 82.31%, respectively, representing improvements of 5.59, 3.18, 3.83, and 9.52 percentage points over YOLO11n-seg. Further comparisons with representative YOLO-based instance segmentation models, including YOLOv8n-seg, YOLOv9c-seg, YOLO12n-seg, and YOLOv26n-seg, demonstrated that OccPepSeg-YOLO achieved the best overall segmentation performance. In particular, its M-mAP50–95 exceeded the best competing result obtained by YOLOv9c-seg by 8.35 percentage points. Under a unified repeated-inference protocol on an RTX 3090 GPU using FP32 precision, a batch size of 1, and 640 × 640 inputs, OccPepSeg-YOLO achieved a mean inference latency of 15.801 ± 1.238 ms, a P95 latency of 17.323 ms, and a throughput of 63.29 FPS. These results demonstrate that the proposed model can produce more complete pepper instance masks under leaf occlusion, fruit overlap, and complex background conditions, providing technical support for field-pepper recognition, fruit counting, and visual perception by agricultural robots.
Why it matches plant phenotyping methods圃場画像からピーマン果実のインスタンスマスクを抽出する手法を開発・比較検証しており、果実カウントや輪郭計測に利用可能な植物形質取得が中心である。
abstractAgricultural operations such as pepper harvesting, fruit counting, and field phenotyping rely on accurate visual recognition and instance segmentation algorithms.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 14 Sept 2026
Structural failure of cereal stems during late-season climate extremes is a critical determinant of yield stability. In barley, breakage of the stem below the spike, known as head loss, leads to major yield losses, particularly in hot and dry regions where the crop is widely grown. Despite a predicted increase in head loss risk due to global warming, current understanding of the genetic, physiological, anatomical, and environmental factors that control head loss remains limited. Overcoming these knowledge gaps is essential to providing a systems-level strategy for barley breeders to develop climate-ready cultivars that are resilient to stem breakage and suitable for industry adoption. Here, we review present knowledge and highlight opportunities for innovation to mitigate head loss through interdisciplinary approaches that combine precise phenotyping through mechanical testing of stem strength and flexibility, high-throughput phenotyping through drone-based spike counting, and genetic modification strategies informed by studies on hormonal regulation and cell wall composition. Coupled with genotypic data, these efforts will enable the development of a genomic selection platform to facilitate future breeding programs. The framework and tools discussed here are broadly applicable to improving stem resilience in other cereal crops.
Why it matches plant phenotyping methods茎の強度・柔軟性や穂数を対象とする表現型計測手法をレビューし、機械試験とドローンによる高スループット計測を育種基盤として論じているため、表現型手法が中心的です。
abstractHere, we review present knowledge and highlight opportunities for innovation to mitigate head loss through interdisciplinary approaches that combine precise phenotyping through mechanical testing of stem strength and flexibility, high-throughput phenotyping through drone-based spike counting
Abstract A systematic review based on 174 Scopus-records of studies using YOLO-type one-stage detectors for detecting citrus fruits, their count, and yield estimation followed PRISMA guidelines 2020. The title/abstract-screening process, done in duplicate (κ=0.920) yielded 90 included study-records, followed by two further post-hoc exclusions. Each study in the 80 reporting on fruit-level detection showed an average precision of 89.6%, recall of 85.9%, and mAP@0.5 of 91.0%. However, coverage for any individual metric rarely exceeded half of the studies, and only 13% were able to report the more stringent mAP@0.5:0.95. Both YOLOv8 and YOLOv5 were each utilized as the backbone architecture by approximately 22.2% of the studies. From 2025, YOLOv11 has also been emerging. Half of all studies modified architectural components including attention modules, lightweight architectures, and variants of IoU loss functions. Original contributions are generally concentrated in downstream tracking, sensor fusion, and yield modeling rather than the detector itself. A custom-made seven-domain risk of bias tool was developed and utilized by two reviewers who arbitrated discrepancies (91.5%). Results showed that all but one of the reviewed studies had a high level of risk due to almost universal lack of statistical validation and limited dataset diversity; a sensitivity analysis excluding the most risky studies left the performance profiles nearly identical. We conclude that the field has converged around a common technical toolkit but continues to lack standardized benchmarks, multispectral data, and rigorous field-deployment validation.
Why it matches plant phenotyping methods柑橘果実の検出・計数・収量推定に用いる画像解析手法を体系的にレビューし、性能評価、リスク・オブ・バイアス、標準化やベンチマーク不足を検討しており、植物フェノタイピング手法が中心である。
abstractA systematic review based on 174 Scopus-records of studies using YOLO-type one-stage detectors for detecting citrus fruits, their count, and yield estimation followed PRISMA guidelines 2020.
Boll-opening concentration is critical for mechanical cotton harvesting, yet it is still assessed mainly by manual records and single time-point indicators that miss temporal dynamics. To bridge the lack of a unified workflow linking vision foundation models, multi-temporal boll-opening monitoring, and harvest decision-making, we developed a cross-scale UAV high-throughput phenotyping framework centered on DINO-BollGX. DINO-BollGX couples a DINO v3 backbone, a RetinaNet detection head, and an adaptive refinement-and-suppression module for robust open-boll detection under complex field conditions. Using multi-temporal UAV imagery collected over two years for 383 cultivars, we reconstructed plot-scale time series of open-boll counts, derived dynamic features describing progression and intensity changes, and proposed a Cotton Boll-Opening Temporal Stability Index (CTSI) to quantify boll-opening rhythm and concentration; CTSI was further integrated with a time-based risk function to generate harvest decision curves. Under unified data and training settings, DINO-BollGX achieved precision = 0.91, F1 = 0.88, and AP@0.50 = 0.80, and provided accurate boll-count estimation (R 2 =0.98; MAE=3.10), outperforming YOLOv11, YOLOv12, YOLOv13, and RT-DETR. On an independent cross-year dataset acquired at 5 m altitude, it obtained precision = 0.98 and F1 = 0.87. An internal consistency analysis showed that CTSI had the expected negative association with Window_days (r = −0.83) and positive associations with Max_count (r = 0.78) and the boll-opening efficiency index (r = 0.92), reflecting the co-occurrence of temporal compactness and main-phase opening intensity in the cultivar population. CTSI ranged from −2.72 to 4.69 across cultivars, enabling identification of highly synchronized boll-opening. Harvest decision curves indicated that the relative net income index peaked at day 67 after the first observation and a compact optimal harvest window near the end of monitoring; on a fixed harvest date, Kuche 130292 (CTSI=4.69) produced 486 open bolls versus 182 for Xinluzao 36 (CTSI=0.53) and 90 for Andizhan-60 (CTSI=-2.72). Overall, the framework integrates dynamic boll-opening phenotyping with harvest timing optimization, supporting scalable cultivar screening and mechanization-ready deployment, with potential extension to harvest decision scenarios in other crops.
Why it matches plant phenotyping methodsUAV画像と基盤モデルを用いて綿花の開絮を検出・定量し、時系列表現型指標を構築・検証する方法が研究の中心であるため。
abstractwe developed a cross-scale UAV high-throughput phenotyping framework centered on DINO-BollGX.
Reproduction assets foundThe paper publicly releases its cotton boll-opening UAV image dataset (3638 patches, 94,774 YOLO-format bounding-box annotations) on GitHub, directly supporting the paper's phenotyping analysis. No author analysis code or trained model checkpoints are explicitly deposited.Dataset · publicentary information for evaluating cross-scale detection performance and characterizing macroscopic spatial patterns. The 5 m imagery acquired on 18 Sept 2024 is used exclusively for cross-year generalization assessment. All cropped images and the corresponding YOLO-format annotation files have been publicly released on GitHub ( https://github.com/mianchen0529/cotton-boll-dataset/tree/main ) to facilitate further research on cotton phenotyping, agricultural remote sensing, and intelligent analytics.
2.3.
Model construction
2.3.1.
Overall architecture of the DINO-BollGX network
The proposed DINO-BollGX network consists of four stages: image preprocessing, feature extraction, object prediction,Open asset ↗mianchen0529/cotton-boll-datasetlines:63-74Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
The edible yield and commercial value of durian are strongly determined by the number of fully developed internal locules, yet their assessment still relies largely on subjective manual inspection or costly destructive analysis. Existing non-destructive approaches, including tapping-based evaluation and X-ray imaging, are either insufficiently standardized or impractical for high-throughput phenotyping. This study proposes the Rotation-Synchronized Structural Locule Evaluator (RS-SLE). The framework integrates convolutional spatial learning with deterministic signal processing to estimate internal locule number from rotational RGB imaging. The proposed system learns frame-wise spatial morphology using convolutional heatmap regression and performs temporal inference through deterministic kinematic signal processing over a mechanically bounded 360° rotation, thereby avoiding reliance on data-intensive recurrent video models. Specifically, sequential 2D locule-probability heatmaps are transformed into a synchronized 1D morphological energy signal, which is subsequently refined using gradient-based integration, Tikhonov regularization, and adaptive morphological thresholding to recover cycle-consistent structural peaks corresponding to fertile locules. This design provides a transparent and computationally efficient alternative to learned temporal memory while maintaining robustness to viewpoint variation, partial occlusion, and high-frequency spine noise. Evaluated on 260 fruits of the ‘Monthong’ cultivar, the proposed framework achieved a mean absolute error of 0.289 locules and a 71.1% exact-match rate on independent rotational videos. When combined with morphological correlation analysis, the framework attained a 100% tolerance accuracy (±1 locule) on the test set with destructive ground-truth measurements. These results demonstrate that dynamic external morphology can serve as a reliable optical proxy for internal locule development. Crucially, the final locule count is derived entirely from the CNN-generated one-dimensional morphological signal. This demonstrates the necessity of localized structural analysis over simple macroscopic shape indices. The findings highlight the value of explainable, kinematics-informed vision systems for practical precision agriculture.
Why it matches plant phenotyping methods回転RGB画像からドリアン内部のlocule数という植物器官状態を推定する画像ベース表現型計測法を開発し、独立動画と破壊的グラウンドトゥルースで検証しているため、方法が研究の中心である。
abstractThis study proposes the Rotation-Synchronized Structural Locule Evaluator (RS-SLE).
Stomatal traits are key microscopic phenotypes for evaluating plant physiology, stress responses, and crop breeding potential. However, in vivo high-magnification microscopy often suffers from a shallow depth of field, causing noticeable defocus blur across different spatial locations and making it difficult to capture clear and complete stomatal structures in a single image. Multi-focus image fusion offers a practical solution, yet existing methods typically rely on supervised training, paired data, or hand-crafted rules, limiting their use in real agricultural microscopy scenarios. In this study, we propose an unsupervised multi-focus fusion framework for reconstructing fully focused stomatal microscopic images. The method integrates two-dimensional feature extraction with three-dimensional cross-focal-plane modeling to capture both spatial details and complementary information across focal planes. A max-response-guided spatial gating module is introduced to enhance focused regions while suppressing defocused responses. Additionally, dual sharpness priors based on perceptual features and wavelet high-frequency information enable pixel-wise pseudo-supervised learning without requiring all-in-focus ground-truth images. The model also predicts a probabilistic focal-plane volume for interpretable all-in-focus reconstruction. Experiments on a maize multi-focus image dataset demonstrate that the proposed method achieves superior or competitive performance across multiple fusion metrics, with entropy (EN), edge information preservation ( Q AB∕F ), Chen-Blum contrast metric ( Q CB ), and visual information fidelity for fusion (VIFF) reaching 7.43, 0.21, 0.41, and 1.01, respectively. Ablation studies confirm the effectiveness of the 3D modeling, spatial gating, and dual-prior sharpness supervision. More importantly, when the fused images serve as input to a YOLO-based stomatal instance segmentation model, the proposed method yields the best segmentation accuracy, with mAP50 and mAP50-95 reaching 0.9937 and 0.9121, respectively. Phenotypic measurements derived from the segmentation masks show high consistency with manual annotations, with the highest coefficient of determination R 2 = 0.97 achieved for stomatal count. These results indicate that the framework can act as an effective front-end module for automated microscopic stomatal phenotyping in agriculture.
Why it matches plant phenotyping methods植物の気孔表現型を対象に、マルチフォーカス画像融合、セグメンテーション、形質測定までを中核的に開発・検証しているため。
abstractwe propose an unsupervised multi-focus fusion framework for reconstructing fully focused stomatal microscopic images.
Reproduction assets foundThe authors state their data and code are publicly available on GitHub, covering the multi-focus stomatal microscopy dataset and the UMF-stomata fusion/phenotyping code.Code · publicOur data and code are available at: https://github.com/Longer-S/UMF-Stomata.Open asset ↗Longer-S/UMF-Stomatahtml-lines:640-655Dataset · publicOur data and code are available at: https://github.com/Longer-S/UMF-Stomata.Open asset ↗Longer-S/UMF-Stomatahtml-lines:683-756Code / dataset availability confirmedCrossref · Europe PMC · checked 5 Sept 2026
Sustainable wheat farming is challenging. Real-time information on crop health, disease transmission, and anticipated yields is essential for farmers. However, they frequently use slow, expensive, or non-communicative tools. This project develops a workable solution. There is no need for massive server farms because the entire system operates on a single graphics card. It incorporates images of wheat fields, Indian farming notes, greenhouse records, harvest statistics, and NASA meteorological data. Consider them as various “eyes” for crop photo analysis, and we tried several lightweight computer vision models. ConvNeXt-Tiny was slower but could operate on older equipment with 75% accuracy; EfficientNetB0 recognised wheat heads with 92% accuracy; and AgroMark, a hybrid solution that merged photo analysis with agricultural metadata (soil type, rainfall, increased to 87%, etc. Combining picture analysis with attention mechanisms (CBAM) allowed us to anticipate the amount of wheat that a field will yield based on these photo insights, and the results showed that our predictions were accurate, with an R 2 score of 0.97. Additionally, we developed a versatile detector that simultaneously detects disease, stress, head count, and pests. It is adjusted to deal with training data that is unbalanced (some diseases are common, while others are rare). As we packed everything into a 16-GB graphics card, we spent real time determining which strategies smaller training sets, removing weak features, and adjusting loss functions, work. We encounter real-world obstacles along the road, such as photographs from different locations not always match, mislabeled photographs from different locations not always match, mislabeled diseases, and neglected rare pests. Our step-by-step instructions, charts, and code are available.
Why it matches plant phenotyping methods小麦画像から病害・ストレス・穂数・収量などの植物形質・状態を推定するマルチモーダル手法を開発し、複数モデルの精度比較と実装上の検証を行っており、表現型取得・推定が研究の中心である。
abstractThis project develops a workable solution.
Reproduction assets foundThe paper builds its multimodal wheat phenotyping analysis on several explicitly cited public data assets: the Kaggle Wheat Plant Diseases image dataset (used for disease classification, Tables 2 and 9), the Global Wheat Head Detection dataset (used for head detection, Tables 1 and 6), FAOSTAT and India Open GovernmentDataset · publicAvailable online at: https://www.fao.org/faostat/ . FAOSTAT statistical database.Open asset ↗lines:1110-1162Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Abstract In peanut ( Arachis hypogaea L.), plant stand establishment, seedling vigor, and canopy growth are key determinants of crop performance; yet traditional ground‐based assessment methods can be destructive, labor‐intensive, and limited in throughput. This study evaluated the potential of vegetation metrics derived from unmanned aerial vehicle (UAV)‐based red‐green‐blue (RGB) and multispectral (MS) imagery for high‐throughput, nondestructive assessment of plant stand establishment, seedling vigor, and light interception in peanut. Six runner‐type peanut cultivars were evaluated in 2024 and seven in 2025, with each cultivar represented by two seed size classes (small and large), to generate variation in these traits. Within‐row vegetation discontinuity‐based plant stand ratings for estimating plant stand count ( R 2 = 0.81–0.90), together with canopy coverage for assessing seedling biomass ( R 2 = 0.77–0.82) and light interception ( R 2 = 0.96–0.98), were the best‐performing vegetation metrics. These vegetation metrics provided similar or greater cultivar separation compared with ground‐based measurements. In contrast, several vegetation indices exhibited strong correlations with ground‐based measurements but provided inconsistent cultivar rankings and statistical groupings. MS imagery outperformed RGB imagery for plant stand and seedling biomass assessment. Overall, these results demonstrate that UAV‐derived canopy metrics provide reliable, high‐throughput tools for early‐ to mid‐season crop assessment and offer scalable alternatives to traditional ground‐based approaches for agronomic, crop physiological, and plant breeding research.
Why it matches plant phenotyping methodsUAV画像から植物体の出芽・苗勢・バイオマス・光 interception を推定する植 phenotyping 手法を開発・評価しており、取得指標の性能検証が研究の中心です。
abstractThis study evaluated the potential of vegetation metrics derived from unmanned aerial vehicle (UAV)‐based red‐green‐blue (RGB) and multispectral (MS) imagery for high‐throughput, nondestructive assessment of plant stand establishment, seedling vigor, and light interception in peanut.
Background The quantification of fungal spores constitutes a fundamental metric in phytopathology, serving as the primary variable for inoculum standardization and being used as a proxy for disease severity. Historically, spore quantification has relied on manual hemocytometry, which remains the most precise counting process to date, where chambers such as the Malassez slide are used to count a subsample of the inoculum. However, this method applied manually is highly labor-intensive, time-consuming, and can be prone to operator-dependent variability. To overcome these limitations, we introduce MIRA (Microscopy Image Recognition & Analysis), a novel open-source software integrating You Only Look Once (YOLO) deep learning algorithms. Featuring a user-friendly graphical interface, MIRA is adaptable to multiple camera systems and supports advanced object detection models, including YOLOv11 and YOLOv26. Results We demonstrate that MIRA can be used to accurately detect and count spores from several phytopathogenic fungi, automatically measure spore surface area, and to differentiate spores across different genera. In an exhaustive comparative analysis using Pyricularia oryzae spores as an example, MIRA was benchmarked against manual gold-standard counting slides (Malassez and Kova) and indirect spectrophotometric methods (SPARK). The P. oryzae model loaded via MIRA achieved a strong correlation (R = 0.96) with manual gold standards while reducing processing time by over 90% for high-concentration samples (10⁶ spores/mL). Beyond this benchmark, we also successfully tested specific YOLO models designed to recognize macro- and microconidia of Fusarium oxysporum f. sp. cubense , a model for Pseudocercospora fijiensis , and a single multiclass model capable of identifying six different rice pathogenic fungi. We provide comprehensive tutorials for operating the software and training custom detection models for free using Roboflow and Google Colab. MIRA is available both as open-source Python code and as standalone executables for Windows and Linux. Conclusions MIRA provides a rapid, accurate, and highly reproducible alternative to manual spore counting, effectively removing a major bottleneck in phytopathology workflows. By combining advanced YOLO-based deep learning with an accessible interface and comprehensive training resources, MIRA makes accessible automated image analysis for researchers without programming expertise. Moreover, MIRA drastically improves the efficiency of high-throughput disease phenotyping and can be adapted for a wide range of microscopic quantification tasks across various biological disciplines.
Why it matches plant phenotyping methods植物病害に関わる胞子の画像検出・計数・サイズ測定ソフトウェアを開発し、手動計数法とのベンチマーク検証も行っている。病害フェノタイピングのための画像解析手法が中心である。
abstractwe introduce MIRA (Microscopy Image Recognition & Analysis), a novel open-source software integrating You Only Look Once (YOLO) deep learning algorithms.
Accurate and efficient acquisition of seedling density and growth information is of great significance for guiding modern agricultural field management. Although drone imagery has been widely used in seedling monitoring, the inherent trade-off between operational efficiency and image resolution limits the effectiveness of remote sensing-based seedling detection. To address this challenge, this study proposes an integrated analytical method combining super-resolution reconstruction and object detection. The approach first employs the Real-ESRGAN model to enhance low-resolution image quality, then utilizes the YOLOv12 model to accurately localize cotton seedlings, and finally generates visualizations of seedling density and growth uniformity. Experimental results demonstrate that super-resolution reconstruction enhances the detection algorithm's capability for small targets, increasing the object detection precision by 3.3%. With the incorporation of super-resolution reconstruction, the seedling counting accuracy reaches 92.08%, representing a 46.15% improvement over the method without super-resolution, thereby effectively enhancing the algorithm's counting capability. Furthermore, this method achieves image detail equivalent to that obtained at 7.5 meters flight altitude while operating at 30 meters, reducing data acquisition time to 1/16 of the original requirement. In practical applications, the visualized results of seedling density and growth uniformity provide precise decision-making support for thinning, replanting, and differentiated field management. The proposed method is not only applicable to cotton but can also be extended to staple crops such as corn, wheat, and rice, with additional potential applications in forestry and ecological monitoring.
Why it matches plant phenotyping methodsUAV画像の超解像化と物体検出により、ワタ苗の密度・生育均一性を定量化する手法が研究の中心であり、性能評価も行っているため。
abstractThe approach first employs the Real-ESRGAN model to enhance low-resolution image quality, then utilizes the YOLOv12 model to accurately localize cotton seedlings, and finally generates visualizations of seedling density and growth uniformity.
Why it matches plant phenotyping methods感染根のカルロース沈着という植物の病態・生理状態を、染色・蛍光画像・Fiji/TWSで検出および定量する方法を最適化した手法論文であり、表現型取得が中心です。
abstractHere, we have optimized a robust and reliable method for detecting callose deposition in soybean lateral roots during Macrophomina phaseolina infections.
Accurate plant population estimation is critical for crop monitoring, yield prediction, and field management in precision agriculture. However, challenges such as complex backgrounds, overlapping plants, and the need for large-scale coverage make seedling counting from UAV imagery difficult. In this study, we propose a novel automated pipeline for maize seedling counting based on high-resolution UAV orthomosaic images. The pipeline begins with an automatic field-based plot extraction method that isolates individual planting regions without the need for manual intervention. A lightweight object detection model, YOLOv8n-CA, is then employed, incorporating Coordinate Attention to enhance feature localization while maintaining fast inference. To further accelerate the process, we introduce the ‘In-Range Sliding’ strategy, which limits inference to only planting regions, reducing computational overhead. Extensive experiments demonstrate the effectiveness of our approach, achieving a mean absolute error (MAE) of 0.587 and a coefficient of determination ( R 2 ) of 94.19 % at the plot level. Additionally, our system enables spatial feature extraction such as seedling spacing, supporting more refined agronomic analysis. This work provides an efficient and scalable solution for plant counting, offering valuable insights for large-scale, rapid post-processing agricultural monitoring.
Why it matches plant phenotyping methodsUAV画像と深層学習によるトウモロコシ幼苗の計数・圃場区画抽出パイプラインを開発し、計数精度を検証しているため、植物表現型取得法が中心である。
abstractwe propose a novel automated pipeline for maize seedling counting based on high-resolution UAV orthomosaic images.
Rice panicle number per unit area is a key determinant of yield, but manual counting remains time-consuming and labor-intensive. This study proposes CEG-YOLO, a lightweight deep learning model for in-field rice panicle detection and counting using consumer-grade RGB imaging devices. The model introduces three improvements to YOLOv11s to address specific challenges in field scenarios: C2f-Fast replaces standard convolutions with depthwise convolutions to reduce computational cost for edge deployment; SPPF-ECA integrates an attention mechanism to suppress complex background interference; and GhostConv reduces feature redundancy to improve detection of dense panicles. A dataset of 5,175 images was collected from four rice cultivars planted at three densities using an iPhone 12. The proposed model achieved 93.9% average precision (AP) on the test set, outperforming YOLOv11s which achieved 89.1%, while reducing parameters to 7.8 million and floating-point operations (FLOPs) to 16.5 billion. Robustness evaluation yielded coefficients of determination (R²) values of 0.91 to 0.94 for lighting, 0.89 to 0.94 for planting density, and 0.90 to 0.94 for cultivar. A cross-year field test using an NVIDIA Jetson Orin NX edge device on 120 quadrats in 2025 achieved R² of 0.91, root mean square error (RMSE) of 4.0, and mean absolute error (MAE) of 3.3 at 20 frames per second, confirming practical deployability. This study demonstrates that smartphone-based proximal phenotyping with an optimized deep learning model can provide accurate, low-cost rice panicle counting for breeding and production applications.
Why it matches plant phenotyping methodsイネ穂数という植物形質をRGB画像から推定する深層学習モデルを開発し、精度・頑健性・実地展開性能を検証しており、フェノタイピング手法が研究の中心である。
abstractThis study proposes CEG-YOLO, a lightweight deep learning model for in-field rice panicle detection and counting using consumer-grade RGB imaging devices.
Accurate and non-destructive counting of rice seedlings is crucial for yield estimation and precision agriculture, yet remains challenging in UAV videos due to dense distribution and strong temporal appearance similarity. This study proposes an efficient tracking-based rice seedling counting framework that integrates an improved Yolov11n detector with a robust multi-object tracking strategy to achieve reliable video level counting. The proposed detector, termed DMNP-YOLO, enhances feature representation, localization robustness, and computational efficiency through Dynamic Snake Convolution, a multi-scale feature attention module, Shape-IoU combined with Normalized Wasserstein Distance, and BatchNorm scaling factor based structured channel pruning, resulting in reductions of 40.5% in Params and 15.2% in GFLOPs while achieving a precision of 0.901 and an mAP@0.5 of 0.921. Building upon accurate frame-level detections, a trajectory based counting mechanism is realized by embedding an Anchor–Angle–Distance association strategy into ByteTrack, which explicitly enforces geometric and temporal consistency across frames, significantly improving tracking stability in dense seedling scenes. As a result, Multi-Object Tracking Accuracy is increased by 5.3 percentage points, identity switches are reduced by 33.3%, and counting accuracy is improved by 3.7 percentage points. Extensive experiments demonstrate that the proposed tracking-based counting framework achieves a mean absolute error of 16.47, a mean absolute percentage error of 6.48%, and an R² of 0.95969. Field scale validation further confirms its practical applicability, achieving an overall rice seedling counting accuracy of 93.4% and demonstrating strong robustness in real world agricultural environments.
Why it matches plant phenotyping methodsUAV画像と検出・追跡アルゴリズムにより圃場のイネ幼苗数を推定する手法を開発し、精度検証と実圃場検証を行っており、植物表現型の取得方法が研究の中心である。
titlePrecisely tracking and counting of field rice seedlings based on UAV platform with DMNP-YOLO and Improved-Bytetrack
Wild rice (Oryza spp.) harbors abundant genetic variation and represents an important germplasm resource for improving yield-related traits in cultivated rice. Panicle number is a key phenotypic trait for evaluating tillering capacity and yield potential in wild rice. However, existing approaches for acquiring panicle-number phenotypes remain limited by low efficiency, high dependence on manual operation, and cumbersome matching between plant targets and accession identifiers. In this study, we proposed an AI-driven UAV phenotyping framework for high-throughput panicle counting of wild rice accessions for germplasm evaluation. The framework integrates field plant localization, accession identifier binding, flight route planning, plant-by-plant video acquisition, spatiotemporal registration, video slicing, and panicle detection and tracking, enabling structured panicle-number outputs indexed by accession identifier. To address the small scale, loose structure, morphological variation, and wind-induced swaying of wild rice panicles in UAV imagery, a wild rice panicle detection model was constructed, and WRPD-Tracker was developed for cross-frame identity association and non-redundant counting. The wild rice panicle detection model achieved an AP@50 of 91.56%, representing a 6.16-percentage-point improvement over the DEIM baseline, with 3.70 M parameters and 6.55 G FLOPs, while WRPD-Tracker achieved a HOTA of 65.1% and a MOTA of 79.0%, representing a 5.8-percentage-point improvement in HOTA over the baseline tracker. At the final counting level, UAV-based counts were highly consistent with manual ground counts, with an R² of 0.992 and an MAE of 0.37 panicles. This framework enables batch acquisition of panicle-number phenotypes in wild rice and provides quantitative support for germplasm evaluation, panicle-number trait comparison, and subsequent yield-related phenotypic studies.
Why it matches plant phenotyping methodsUAV画像とAIによるイネ穂数の取得・追跡・計数手法を開発し、手動計数と技術検証しており、植物フェノタイピング手法が研究の中心です。
abstractwe proposed an AI-driven UAV phenotyping framework for high-throughput panicle counting of wild rice accessions for germplasm evaluation.
The ready-to-eat lettuce industry is rapidly expanding, increasing the need for reliable, scalable methods to assess seed germination and early growth under realistic soil conditions. This study presents an automated imaging-based approach for quantifying germination dynamics and seedling vigor using a low-cost multi-camera system under greenhouse conditions. Lettuce seeds were grown in soil either inoculated or non-inoculated with the soil-borne pathogen Rhizoctonia solani. Top-view images were acquired using commercial surveillance cameras and processed through a calibrated pipeline including geometric correction, color normalization, vegetation segmentation, clustering, and temporal tracking of emergence events. Seedling vigor was quantified through projected leaf area estimation. The proposed method enables accurate estimation of germination kinetics and growth dynamics under field-like conditions. Automated counts were validated against manual measurements at both intermediate and final time points, achieving high agreement in both cases. At the final assessment, the method reached R² = 0.98 and RMSE = 1.12, while at the midterm evaluation it achieved improved performance with R² = 0.998 and RMSE = 0.5, reflecting the lower complexity of plant structure at earlier growth stages. Results showed that pathogen inoculation significantly reduced both germination rate and seedling vigor, with up to 70% reduction in biomass accumulation. The proposed framework provides a robust, low-cost solution for high-throughput phenotyping of early plant development in soil-based systems, supporting scalable agricultural experimentation.
Why it matches plant phenotyping methods低コスト多カメラ画像システムと画像解析パイプラインを開発・検証し、発芽動態と幼植物活力を定量化しているため、植物フェノタイピング手法が中心です。
abstractThis study presents an automated imaging-based approach for quantifying germination dynamics and seedling vigor using a low-cost multi-camera system under greenhouse conditions.
Plant disease phenotyping underpins resistance breeding, epidemiology and crop-loss management, yet it remains a recognised bottleneck. This review asked whether the two metrics that dominate the discipline, the disease severity index (DSI) and the area under the disease progress curve (AUDPC), adequately represent disease as a temporally unfolding process, and what the evidence says about dynamic alternatives. Reporting followed PRISMA 2020 and the Synthesis Without Meta-analysis (SWiM) guideline. Web of Science Core Collection, Scopus, PubMed and a Google Scholar grey-literature sweep were searched for records published between January 2020 and December 2025, retrieving 1,192 records; 874 remained after de-duplication, 128 full texts were assessed and 31 studies met the eligibility criteria. Citation chasing added 24 foundational works, giving 55 included studies. Records were dual-screened (Cohen's kappa = 0.86), appraised with an adapted Mixed Methods Appraisal Tool, and synthesised using vote counting by direction of effect, an evidence map and structured cross-study comparison; meta-analysis was inappropriate because outcomes were not commensurable. Thirty studies (54.5%) represented disease at a single assessment and eight (14.5%) collapsed the epidemic into one integrated area, whereas only twelve (21.8%) retained the full trajectory. Across six outcome domains, all 29 study-level comparisons favoured the temporally richer method and none reported a null or negative result, an asymmetry indicating probable reporting bias. Certainty was high for visual-assessment findings, moderate for sensing and dynamic modelling, and low for field-realised genetic gain. The phenotyping bottleneck has migrated from data acquisition to data representation.
Why it matches plant phenotyping methods植物病害フェノタイピング手法の動的評価を中心に、既存指標と代替手法を体系的に比較した方法論レビューである。
titleA Systematic Review of Dynamic Disease Phenotyping in Plant Pathology
1. Morphological traits such as floral area and body size are fundamental to ecological research, serving as inputs for studies of pollinator–plant interactions, habitat quality, and biodiversity monitoring. However, accurately measuring these traits from images remains challenging, particularly in complex field conditions where existing tools exhibit reduced accuracy and limited generalizability across taxa. 2. We present EcoMorph, a modular morphological measurement system that leverages the Segment Anything Model 3 (SAM3) to quantify traits across diverse ecological contexts. Unlike task-specific segmentation models requiring domain-specific training data, SAM3's prompt-based architecture enables segmentation of arbitrary biological structures from natural-language prompts, using the same underlying model across flowers, insects, and other targets without retraining. From the resulting segmentations, EcoMorph extracts three classes of measurement: area, linear dimensions, and object count. 3. We validated EcoMorph across two ecological scales. At the intermediate scale, EcoMorph-derived floral area agreed closely with manual ImageJ measurements (R² = 0.935, n = 74) under simple-background conditions and (R² = 0.928, n = 58) under complex-background conditions, with valid predictions for 95% of images. At the fine scale, EcoMorph-derived insect body area was strongly correlated with hand-measured intertegular distance (r = 0.810, n = 349), capturing body-size variation across species from the small Bombus impatiens to the large Xylocopa virginica. Object counts matched manual counts almost exactly for well-separated insects in an insect box (R² = 1.000, n = 12). 4. By combining prompt-based segmentation with modular measurement, EcoMorph enables high-throughput quantification of area, size, and count from heterogeneous image sources without taxon-specific training. This generality supports a broad range of ecological applications, including pollinator and plant trait research, biodiversity and abundance monitoring, and allometric biomass estimation.
Why it matches plant phenotyping methods植物の花の面積を画像から定量する汎用手法を開発し、手動測定と検証しているため、植物フェノタイピング手法が中心である。
abstractWe present EcoMorph, a modular morphological measurement system that leverages the Segment Anything Model 3 (SAM3) to quantify traits across diverse ecological contexts.
Unmanned aerial vehicle (UAV) photogrammetry offers a cost-effective approach to tree-level detection, however, Structure-from-Motion (SfM) outputs are sensitive to processing choices and site conditions, which can alter canopy representation and reduce individual-tree detection accuracy. Here, we systematically evaluate how SfM reconstruction quality and depth-map filtering influence RGB-only individual-tree detection under controlled acquisition conditions. Objectives were to (i) identify an optimal SfM-derived point-cloud configuration for delineating individual trees, and (ii) implement and test a segmentation workflow (local-maxima treetop detection plus Dalponte2016 in lidR) for detecting and counting trees. We assessed RGB-only SfM for individual-tree detection (ITD) across thirteen 1.21-ha loblolly pine ( Pinus taeda ) plots located in two counties in the state of Alabama in the southeastern United States; eight even-aged plantations and five mixed pine-hardwood stands, while holding image acquisition parameters constant. Using Agisoft Metashape Professional (Agisoft LLC, St. Petersburg, Russia), dense-cloud quality (Lowest, Low, Medium, High, Ultra High) and depth-map filtering (Disabled, Mild, Moderate, Aggressive) were varied in a 5 × 4 full-factorial design; assessment metrics included point-cloud density, canopy-surface completeness, canopy-height-model (CHM) agreement with field heights, and ITD precision/recall/F1. We identified a single high-resolution configuration (Ultra High + Disabled) by screening parameter sets for structural accuracy and suppression of false peaks. Using this configuration, CHMs matched field heights in Washington County, Alabama (R 2 = 0.96; RMSE = 0.44 m; bias = − 0.01 m) and in Cullman County, Alabama (R 2 = 0.44; RMSE = 1.14 m; bias = − 0.09 m); pooled performance was R 2 = 0.98; RMSE = 0.54 m; bias = − 0.01 m. ITD accuracy at the primary 3 m match radius yielded a precision of 0.03; recall = 0.29; F1 = 0.05 in the even-aged plantations (Washington) and a precision of 0.03; recall = 0.12; F1 = 0.05 in mixed pine–hardwood stands (Cullman); pooled F1 = 0.05. The selected parameters and workflow are reproducible and transferable, provide insight into RGB-SfM ITD performance, and indicate when lidar remains preferable for crown delineation.
Why it matches plant phenotyping methodsRGB-SfMによる個体樹の検出・樹高推定と、SfM設定およびセグメンテーションワークフローの系統的評価が研究の中心であり、植物の樹冠構造・樹高という形態形質を抽出する方法を検証している。
abstractwe systematically evaluate how SfM reconstruction quality and depth-map filtering influence RGB-only individual-tree detection
Reproduction assets foundThe paper's Code availability statement deposits the authors' SfM/ITD processing scripts publicly on OSF (DOI 10.17605/OSF.IO/UXBCZ). Phenotype/field datasets are only available on request, so they are not public assets.Code · publicThe workflow and processing scripts used in this study are publicly available through the Open Science Framework
(OSF) repository: Singh and Narine, [32]. Code Repository for Optimizing SfM Parameters for RGB-Only Individual-Tree
Detection in Loblolly Pine and Mixed Pine-Hardwood Stands. https://doi.org/10.17605/OSF.IO/UXBCZ.Open asset ↗10.17605/OSF.IO/UXBCZpdf-page:12 lines:1-70Plant phenotyping relevance match · UnverifiedCrossref · checked 5 Sept 2026
Accurate field crop counting supports phenotyping, crop monitoring, and yield-related analysis, but real field images often contain scale variation, target overlap, cluttered vegetation, shadows, and visually similar backgrounds. These factors make density-map regression difficult because weak target responses and accumulated false responses in non-target regions can both bias the final count. This study proposes Light-FCCNet, a compact multi-scale framework for accurate field crop counting under a limited parameter budget. The framework integrates three coordinated components: a lightweight feature pyramid aggregation module for compact scale-aware representation, a multi-attention fusion module for refining fused features and suppressing background-induced responses, and an FCC loss that combines pixel-level density regression, count consistency, and structural similarity. Light-FCCNet is evaluated on three public field crop counting datasets, namely GWHD, MTC, and URC. In the canonical ablation trajectory consisting of baseline, baseline_p1, baseline_p1_p2, and full configurations, the full model achieves the lowest errors, with MAE values of 13.28 on GWHD, 18.10 on MTC, and 61.23 on URC, corresponding to reductions of 18.0%, 25.3%, and 35.2% relative to the baseline. Compared with representative general counting baselines and our implementation of TasselNetV2++ under the same experimental protocol, Light-FCCNet obtains the lowest MAE on all three datasets while using only 0.92 M parameters. These results indicate that its advantage comes from the coordinated design of compact pyramid aggregation, attention-guided feature refinement, and counting-oriented supervision, rather than from any isolated module alone.
Why it matches plant phenotyping methods作物個体数を圃場画像から推定する新規CNN手法を開発し、複数データセットとアブレーションで性能検証しており、表現型取得・推定が研究の中心である。
abstractThis study proposes Light-FCCNet, a compact multi-scale framework for accurate field crop counting under a limited parameter budget.
Abstract Biotic stress is a major, yet under-quantified, driver of global soybean yield losses, and field-based phenotyping under pest pressure remains a critical bottleneck for crop improvement. Using multi-temporal data from soybean genotypes grown under insecticide-protected and unprotected conditions in Brazil, we present a UAV-based, large-scale and non-invasive framework for evaluating genotype performance under natural pest pressure. We introduce a three-dimensional metric that jointly captures productivity, feature-level similarity as a proxy for tolerance, and phenological response through days to maturity. This unified formulation enables field-based quantification of pest resilience and replaces labor-intensive and often unreliable direct pest collection and counting. To operationalize this framework, we integrate vegetation indices and self-supervised visual embeddings into a common representation space linking feature stability, performance response and phenological development. This approach enables robust identification of genotypes that maintain feature integrity, minimize developmental delay and sustain yield under pest pressure, with genotypic differences peaking during the pod-fill (R3–R4) and grain-fill (R5.1–R5.5) stages. Overall, this work establishes a scalable, field-ready paradigm for quantifying crop resilience to biotic stress and provides a practical pathway to accelerate breeding for stable yields under real-world agricultural conditions.
Why it matches plant phenotyping methodsUAVによる大規模な圃場フェノタイピング基盤と、植生指数・視覚埋め込みを統合した新しい耐虫性表現型の定量手法が研究の中心である。
abstractwe present a UAV-based, large-scale and non-invasive framework for evaluating genotype performance under natural pest pressure
Reproduction assets foundThe paper explicitly states that the analysis code is publicly available in the authors' GitHub repository (jianglong26/soybean-insect-resistance), which directly reproduces this paper's phenotyping pipeline (orthomosaic processing, VI/DINOv3 feature extraction, similarity analysis, genotype ranking). The paper also声明sCode · public540 The code used for analysis is available at https://github.com/jianglong26/Open asset ↗pdf-page:16 lines:1-45Plant phenotyping relevance match · UnverifiedCrossref · checked 5 Sept 2026
Accurate, high-throughput quantification of rice panicles is important for yield estimation and breeding-oriented rice phenotyping. However, unmanned aerial vehicle (UAV)-based panicle detection remains challenging because flight-altitude variation produces large target-scale changes. Flooded paddy backgrounds, leaf occlusion, and illumination fluctuations further obscure small panicle targets. To address these challenges, we constructed a composite multi-altitude dataset covering UAV imagery acquired from 3 to 20 m under varying field conditions. We then propose Panicle-DETR, a lightweight detection and counting framework based on a frequency-aware Cross Stage Partial (CSP) backbone. Rather than treating the Fast Fourier Transform (FFT) as a filter by itself, the proposed FasterFD module uses frequency-domain representations with learnable frequency-response reweighting to enhance panicle-related texture cues and reduce redundant background responses. A Lossless Feature Encoder is designed to preserve fine spatial information for small targets across altitude-induced scale changes, while a composite metric loss based on Normalized Gaussian Wasserstein Distance (NWD) and Inner-IoU improves localization for adherent and overlapping panicle clusters. On the composite dataset, Panicle-DETR achieved a Precision of 90.97%, a Mean Absolute Error of 4.28, and an R2 of 0.957 for single-frame panicle counting. With 13.78 M parameters and 53.0 GFLOPs, the framework achieved 16.9 FPS with 1.96 GB peak GPU memory in a battery-powered notebook benchmark, supporting its potential for resource-constrained field-side UAV image analysis.
Why it matches plant phenotyping methodsUAV画像からイネ穂の検出・計数という植物形質推定を中心に、マルチ高度データセットと専用解析フレームワークを開発・評価しているため。
abstractAccurate, high-throughput quantification of rice panicles is important for yield estimation and breeding-oriented rice phenotyping.
Orthoimagery derived from unmanned aerial vehicles (UAVs) has become a valuable data source for crop-growth monitoring. Individual plant-level (IPL) information enables high-throughput analyses by capturing plant-to-plant variability within fields. However, reliable IPL-based analysis requires accurate extraction of plant-specific regions, which remains challenging in soybean cultivation due to weed interference and canopy overlap. This study proposed an automatic preprocessing framework for IPL soybean growth monitoring that integrates deep-learning-based semantic segmentation with a furrow-guided region of interest (ROI) generation strategy using UAV imagery. A segmentation model was developed using combinations of RGB and multispectral orthoimagery, and a furrow line detection algorithm was designed to generate IPL ROIs aligned with crop rows. The ensemble model combining U-Net, DeepLabV3+, and SegFormer achieved the most stable performance (F1-score up to 0.94 and IoU up to 0.89). The furrow-guided ROI generation algorithm also accurately estimated crop counts, showing strong agreement with manual observations (R² = 0.90 and RMSE = 6.35). The generated IPL ROIs enabled accurate quantification of growth-related features, with strong agreement between automatically generated and manually delineated ROIs (R² > 0.90). Overall, the proposed preprocessing framework provides a practical and scalable solution for UAV-based high-throughput phenotyping in soybean and other ridge-based cropping systems.
Why it matches plant phenotyping methodsUAV画像から個体単位の植物領域を抽出し、成長形質を定量化する前処理・セグメンテーション手法が研究の中心であるため。
abstractThis study proposed an automatic preprocessing framework for IPL soybean growth monitoring that integrates deep-learning-based semantic segmentation with a furrow-guided region of interest (ROI) generation strategy using UAV imagery.
Reproduction assets foundThe authors state that the complete implementation of their IPL soybean preprocessing pipeline (semantic segmentation + furrow line detection) is publicly available on Zenodo. The annotated sample dataset, however, is only available upon request from the corresponding author, so it is not a public asset.Code · publicThe complete implementation of
this pipeline is publicly available at https://doi.org/10.5281/zenodo.21095307.Open asset ↗zenodo · 10.5281/zenodo.21095307pdf-page:7 lines:1-62Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 15 Sept 2026
Abstract Flower-visiting insect populations are declining since the 1990s, especially because of the decrease of floral resources in agricultural settings. Mass flowering crops can help increase resource availability, and plant breeding can be directed towards selecting varieties attracting more flower-visiting insects. This requires the implementation of an automated high-throughput phenotyping tool for assessing the attractiveness of plant genotypes to flower-visiting insects. In this study, ( i ) we present a procedure to take standardized images of sunflower heads with camera traps continuously at day and night in the field; ( ii ) we trained two versions of a deep learning model, named PolliCrop, to automatically detect and identify the three insect classes visiting the most sunflower (non- Bombus bees, bumble bees, lepidopterans); ( iii ) we assessed and validated the ability of PolliCrop to correctly predict the true visitation frequencies of the insect classes on three sunflower genotypes; ( iv ) we presented two statistical approaches to compare the insect visitation frequencies between plant genotypes, one including weather variables, and the other one without. One PolliCrop version yielded satisfying performance to correctly detect the three insect classes. In particular, it correctly predicted the insect visitation frequencies on two sunflower genotypes in a range of ±10%. The other PolliCrop version can be useful in certain contexts of images and objectives. PolliCrop can be extended in the future to other crop species by training PolliCrop on new images captured in these crops. The field experimental design to set up for comparing the attractiveness between genotypes is also discussed.
Why it matches plant phenotyping methods植物遺伝型の花への訪花昆虫誘引性を推定する画像取得・深層学習パイプラインを開発し、訪花頻度の予測性能も検証しており、表現型取得法が中心である。
abstractwe present a procedure to take standardized images of sunflower heads with camera traps continuously at day and night in the field
0. Morphological traits such as floral area and body size are fundamental to ecological research, serving as inputs for studies of pollinator–plant interactions, habitat quality, and biodiversity monitoring. However, accurately measuring these traits from images remains challenging, particularly in complex field conditions where existing tools exhibit reduced accuracy and limited generalizability across taxa. We present EcoMorph, a modular morphological measurement system that leverages the Segment Anything Model 3 (SAM3) to quantify traits across diverse ecological contexts. Unlike task-specific segmentation models requiring domain-specific training data, SAM3’s prompt-based architecture enables segmentation of arbitrary biological structures from natural-language prompts, using the same underlying model across flowers, insects, and other targets without retraining. From the resulting segmentations, EcoMorph extracts three classes of measurement: area, linear dimensions, and object counts. We validated EcoMorph across two ecological scales. At the intermediate scale, EcoMorph-derived floral area agreed closely with manual ImageJ measurements (R 2 = 0.935, n = 74) under simple-background conditions and (R 2 = 0.928, n = 58) under complex-background conditions, with valid predictions for 95% of images. At the fine scale, EcoMorph-derived insect body area was strongly correlated with hand-measured intertegular distance (r = 0.810, n = 349), capturing body-size variation across species from the small Bombus impatiens to the large Xylocopa virginica . Object counts matched manual counts almost exactly for well-separated insects in an insect box (R 2 = 0.9997, n = 12). By combining prompt-based segmentation with modular measurement, EcoMorph enables high-throughput quantification of area, size, and abundance from heterogeneous image sources without taxon-specific training. This generality supports a broad range of ecological applications, including pollinator and plant trait research, biodiversity and abundance monitoring, and allometric biomass estimation.
Why it matches plant phenotyping methods画像から花の面積など植物形態形質を抽出する汎用システムを開発し、手動測定との一致で検証しており、植物フェノタイピング手法が中心である。
abstractWe present EcoMorph, a modular morphological measurement system that leverages the Segment Anything Model 3 (SAM3) to quantify traits across diverse ecological contexts.
Reproduction assets foundThe paper's Data and code availability statement provides a public Zenodo deposit containing the validation datasets and code used for the EcoMorph phenotyping measurements (floral area, insect morphometrics, counts), plus a public web deployment of the EcoMorph software itself.Code · publicValidation datasets and code are available here on Zenodo
https://zenodo.org/records/20980236.Open asset ↗Zenodo · 20980236pdf-page:2 lines:1-54Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
This paper presents a method to estimate apple tree flower cluster count using image analysis techniques. The main research question is how accurately flower clusters on apple trees can be counted using a camera-based approach and weakly-supervised learning, compared to traditional visual estimation. A camera is used to capture images of blooming apple trees from two sides. These images are processed by a weakly-supervised model based on a ResNet feature extractor and a feature pyramid network serving as the feature aggregator. The model is trained and validated using reference data obtained through manual flower cluster counts in the orchard and from estimations based on camera images. The model was trained using field-validated data and visually-estimated data, enabling a comparative evaluation. The proposed model surpassed conventional image segmentation methods in estimating both manually counted and visually estimated flower clusters. The proposed method achieves a relative error of 10.26% in estimating flower cluster quantities, demonstrating its effectiveness and improved accuracy over traditional approaches. Its reliance on field-validated reference data adds to its robustness and practical relevance.
Why it matches plant phenotyping methodsリンゴ樹の花房数という植物形質を、カメラ画像と弱教師あり学習で推定する手法の開発・検証が研究の中心であるため。
abstractThis paper presents a method to estimate apple tree flower cluster count using image analysis techniques.
Abstract Background Stomata and pavement cells are fundamental components of the leaf epidermis, jointly regulating gas exchange, water loss, and leaf surface expansion. Stomata size, aperture, density, and pavement cell morphology are critical parameters for assessing plant transpiration efficiency, epidermal growth dynamics, and adaptive responses to environmental constraints. Despite their biological importance, quantifying stomatal and pavement-cell traits remains seldom not generalized, simple, and fast enough . Manual or semi-automated approaches limit large-scale phenotyping and restrict the integration of epidermal morphology into crop-improvement pipelines aimed at developing climate-resilient varieties with optimised stomatal patterning. To address such limitations, we developed Stomatalia , a deep learning-based platform designed to automate and standardise the quantification of stomatal and pavement cell traits. The algorithm was trained on epidermal images of cultivated and wild potato and tomato genotypes grown under optimal and abiotic-stress conditions. Stomatalia automatically detects stomata and pavement cells and extracts a broad range of morphological and integrative epidermal parameters, enabling high-throughput phenotyping within a unified workflow. Results Prior to platform development, we optimised a rapid, minimally-destructive leaf-printing protocol that generates negative impressions of the leaf surface within 40–100 s. Transparent positive prints were subsequently produced and imaged under a light microscope at scale settings ranging from 20 to 200 μm. The resulting images are analysed using Stomatalia’s user-friendly web-based interface, which runs an instance-segmentation deep learning algorithm to detect, count, and calculate stomatal and pavement cell parameters. The platform outputs structured files containing raw measurements, derived integrative traits, and associated metadata, facilitating downstream statistical and physiological analyses. Algorithm evaluation on independent datasets demonstrated high performance within the validated dicot imaging domain, with F 1 -scores ranging from 0.86 to 0.94 depending on image scale, species, and resolution, and high segmentation overlap for both stomata and pavement cells. The generality of stomatal detection was also tested on spring onion, chickpea, balsam poplar, and wheat in cross-species feasibility tests, although performance was more variable in monocots, and pavement-cell segmentation remained species- and architecture-dependent. Benchmarking against another publicly available app further showed that, under the tested web interface settings and image types, Stomatalia exhibited closer agreement with manual counts and substantially faster processing times. The practical performance of Stomatalia was further tested in a proof-of-concept trial with potato plants subjected to optimal irrigation and a long, gradual drought. The platform reliably quantified epidermal traits despite variations in leaf morphology and image quality, supporting the integrated interpretation of stomatal and pavement-cell responses under stress. Conclusions We developed Stomatalia as a robust, user-friendly deep learning platform for automated, high-throughput analysis of bright-field leaf epidermal images across varying magnifications and resolutions. Stomatalia facilitates rapid, reproducible, and coordinated phenotyping of stomatal and pavement cells by integrating methodological standardisation, computational automation, and multi-trait extraction in a single analytical workflow. Its strongest current application is the analysis of high-quality dicot leaf-print images, particularly in species and imaging conditions similar to those used for model training and validation. Cross-species and benchmark analyses further define its current scope: stomatal detection can be transferred to some additional epidermal architectures, whereas robust pavement-cell segmentation in monocots or highly divergent species will require further annotation and model retraining. Within these defined boundaries, Stomatalia provides a flexible and extensible framework for studying stomatal and pavement cell morphology and environmental plasticity, while also supporting broader efforts to dissect and optimise plant responses to abiotic stress.
Why it matches plant phenotyping methods気孔・舗装細胞の形態形質を画像から自動抽出する深層学習プラットフォームを開発し、独立データで性能評価・比較検証しているため、植物フェノタイピング手法が中心である。
abstractwe developed Stomatalia , a deep learning-based platform designed to automate and standardise the quantification of stomatal and pavement cell traits.
Reproduction assets foundThe authors publicly deposited the paper's test image datasets (raw/input leaf-print images, detection outputs, manual ground-truth counts, exported datasets) and the model file on Figshare, and provide a public Google Colab demo for running the Stomatalia algorithm. Both are paper-specific, public, and actionable.Dataset · publicThe test datasets and model file used in this work are available through the following link:
https://doi.org/10.6084/m9.figshare.32532672. The test_sets.zip archive contains the test image datasets
(cross-species and benchmark analysis), including raw/input images, detection output images, manual
ground-truth counts and exported datasets.Open asset ↗Figshare · 10.6084/m9.figshare.32532672pdf-page:25 lines:1-75Plant phenotyping relevance match · UnverifiedCrossref · checked 5 Sept 2026
Crop targets in UAV aerial images are typically characterized by small scale, dense distribution, severe mutual occlusion, and complex backgrounds, which often lead to low detection accuracy and large counting errors for existing deep learning models. To address these issues, this study proposes an improved YOLOv12-based crop detection and counting model, named YOLO-DC. By introducing an attention mechanism (LGCB-AM) and a multi-scale detection head (MS-DH), the proposed model effectively enhances local texture extraction, global modeling, foreground–background contrast, and boundary perception for dense small objects. Subsequently, a series of comparative experiments, ablation studies, and transfer experiments were conducted on the wheat and rice datasets. The results show that YOLO-DC achieves a favorable balance among detection accuracy, counting error, and model efficiency and overall outperforms the other comparison models. Ablation studies further verify the effectiveness of the proposed design, showing that LGCB-AM is the key contributor to the performance improvement, while the boundary branch and repulsion branch play critical roles in dense-target discrimination. In addition, an appropriate module insertion strategy can effectively balance high-level semantic enhancement and feature fusion stability. Transfer experiments demonstrate that pretraining on the wheat dataset and fine-tuning on the rice dataset significantly outperform training from scratch, indicating strong cross-crop transfer potential. Overall, the proposed YOLO-DC provides an effective solution for high-precision crop detection and counting in agricultural scenarios.
Why it matches plant phenotyping methodsUAV画像から作物個体を検出・計数する手法を中心に、モデル開発、比較、アブレーション、転移検証を行っており、植物個体数という観測可能な形態・集団特性を抽出するため、植物フェノタイピング手法として適格です。
abstractthis study proposes an improved YOLOv12-based crop detection and counting model, named YOLO-DC.
To overcome the inefficiency and subjectivity of manual seedling surveys, this study presents a unsupervised framework for evaluating maize sowing quality and emergence uniformity via UAV-based remote sensing. Centimeter-level multispectral imagery was captured to reconstruct 3D point clouds using SfM and MVS techniques. At the algorithmic level, an improved unsupervised pipeline was developed: the Otsu method was employed for plant segmentation, followed by a Fourier Transform to extract 2D spatial frequency features for precise crop row identification and automated spacing measurement. Subsequently, the Combined Entropy Uniformity (CEU) index was developed using Shannon entropy, and a proxy for canopy closure (CCP) was derived using a porosity model, thereby enabling the simultaneous relative quantification of seedling height consistency, spatial distribution uniformity, and canopy geometric structure variability. At the application level, the framework was validated through field trials involving 19 precision planters of diverse configurations. Performance was assessed using indices such as qualified spacing, miss-sowing, and the Coefficient of Variation of Plant Spacing (PSCV). Results indicate that: (1) Vacuum-type planters exhibited optimal stability at speeds of 7–9 km/h, achieving an average qualified spacing rate of 76.7% and a PSCV of approximately 24%, whereas finger-pickup planters were more sensitive to seed size variation and mechanical vibration. (2) The results from the Generalized Additive Model (GAM) suggest a possible nonlinear relationship between seeding rate and certain uniformity indices, indicating that appropriately adjusting operational parameters could help balance operational efficiency and seeding quality; however, this trend requires further validation with larger sample sizes and repeated observations. (3) Point cloud CEU metrics and canopy structure proxies based on the Gap Fraction model showed statistical correlations with certain manually collected indicators, indicating that this method has the potential for rapid screening of seeding quality and relative evaluation of seedling population structure at the field scale under the current experimental conditions.
Why it matches plant phenotyping methodsUAV画像・3D点群から作物の出芽、草丈均一性、空間分布、群落構造を抽出する解析ワークフローを開発し、19種のプランターで検証しており、植物表現型取得法が中心である。
abstractthis study presents a unsupervised framework for evaluating maize sowing quality and emergence uniformity via UAV-based remote sensing.
Accurate yield estimation and crop load monitoring are essential for precision orchard management, supporting targeted fertilization, pruning, thinning, harvest planning, and marketing decisions. However, reliable in-situ monitoring remains challenging because commercial orchards are characterized by severe canopy occlusion, fruit overlap, heterogeneous tree architecture, variable illumination, and complex backgrounds. This review synthesizes advances in multi-modal sensing and deep learning for orchard yield estimation, breaking down the paradigm into intermediate fruit-counting or crop-load monitoring steps and supplementary spectral quality-assessment dimensions. First, yield-related indicators are summarized, including direct phenotypic traits such as fruit number, size, volume, and spatial distribution, as well as indirect structural and physiological proxies such as canopy volume, vegetation indices, flowering intensity, and spectral maturity attributes. Second, representative sensing devices and carrying platforms are reviewed, including red-green-blue (RGB) cameras, red-green-blue-depth (RGB-D) sensors, light detection and ranging (LiDAR), hyperspectral and multispectral systems, unmanned ground vehicles (UGVs), and unmanned aerial vehicles (UAVs). Third, the evolution of estimation methods is discussed, from traditional image processing and machine learning to object detection, instance segmentation, multi-object tracking, point-cloud analysis, remote-sensing regression, and multi-modal fusion. The review shows that no single sensor or algorithm can satisfy all orchard monitoring requirements. Ground-based vision and depth sensing are more suitable for fine-scale fruit counting and sizing, whereas UAV and spectral sensing provide advantages for regional yield mapping and quality-enhanced assessment. Future research should emphasize occlusion-aware perception, robust cross-environment generalization, lightweight edge deployment, standardized benchmarks, and integrated quantity-quality monitoring frameworks for actionable crop load management.
Why it matches plant phenotyping methods果実数・サイズ・体積・空間分布などの植物形質を対象に、センシング機器と画像解析・深層学習による収量推定法を体系的にレビューしており、フェノタイピング手法が中心である。
abstractThis review synthesizes advances in multi-modal sensing and deep learning for orchard yield estimation, breaking down the paradigm into intermediate fruit-counting or crop-load monitoring steps and supplementary spectral quality-assessment dimensions.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 5 Sept 2026
Field / plotCountingObject detectionStress / disease detectionDisease symptoms / severity
This paper targets the practical needs of plant disease and pest object detection and counting in complex field environments and proposes a lightweight improved framework based on YOLOv10. A MEMBA-F multi-scale feature enhancement module is introduced on the Neck to strengthen representations of small targets and weak-texture lesions, and a CSCAF cross-scale context-aware fusion module is designed to adaptively align high-level semantics with low-level details via cross-scale attention and gated selection, suppress background interference, and improve localization stability. The proposed method is systematically compared with two-stage detectors, YOLO-series models, and Transformer-based detectors on three public datasets, and is further investigated through ablation studies, confusion matrix analysis, and Grad-CAM interpretability analysis. In addition, a density-binned counting evaluation is conducted to validate robustness from sparse to dense scenarios. Experimental results demonstrate that the proposed method achieves superior performance in Precision, Recall, mAP@50, and mAP@50-95, and significantly reduces counting errors in dense scenes under deployable inference cost, providing reliable support for precision plant protection monitoring and decision making.
Why it matches plant phenotyping methods植物病害の病変と害虫を画像から検出・計数するYOLOv10改良法を開発し、複数データセット、比較実験、アブレーション、密度別評価で性能検証しており、植物の病害状態の取得方法が中心である。
abstractThis paper targets the practical needs of plant disease and pest object detection and counting in complex field environments and proposes a lightweight improved framework based on YOLOv10.
PHENET's Use Case 1 on plant health is validating sensors and imaging methods for the assessment of wheat ear diseases, with two AI-powered applications now reaching proof-of-concept stage. The first, FUSASEYD, addresses Fusarium Head Blight (FHB), a major fungal disease in winter wheat. Using RGB field images and a deep learning instance segmentation model (YOLOv11), the application detects and quantifies FHB symptoms on wheat ears, offering an automated alternative to time-consuming expert visual scoring. GEVES has developed both a PC interface and a smartphone application to visualise model predictions in the field. Validation in French registration trials is planned for the 2026 campaign. A companion article by V. Cadot et al. is currently under review in the Journal of Experimental Botany special issue on Plant Phenomics & Enviromics Across Scales. The second, COYL (Counting Orange and Yellow Larvae), tackles a practical challenge faced by breeders, and rapidly counting wheat blossom midge larvae, both Sitodiplosis mosellana and Contarinia tritici, to characterise variety susceptibility. Using smartphone RGB images and YOLOv-based object detection, the best-performing model achieved high accuracy and successfully distinguished between the two visually similar species. An online counting application has been developed, currently accessible to Walloon Agricultural Research Centre members. The labelled COYL-1 dataset is publicly available at https://doi.org/10.5281/zenodo.19402333 for community use. A companion article by Antoine Deryck et al. is under submission at Plant Phenomics Journal.
Why it matches plant phenotyping methodsRGB画像と深層学習によるコムギ穂の病徴検出・定量化を開発し、専門家評点の自動化とセンサー／画像法の検証を目的とするため、植物フェノタイピング手法が中心である。
abstractPHENET's Use Case 1 on plant health is validating sensors and imaging methods for the assessment of wheat ear diseases
PHENET's Use Case 1 on plant health is validating sensors and imaging methods for the assessment of wheat ear diseases, with two AI-powered applications now reaching proof-of-concept stage. The first, FUSASEYD, addresses Fusarium Head Blight (FHB), a major fungal disease in winter wheat. Using RGB field images and a deep learning instance segmentation model (YOLOv11), the application detects and quantifies FHB symptoms on wheat ears, offering an automated alternative to time-consuming expert visual scoring. GEVES has developed both a PC interface and a smartphone application to visualise model predictions in the field. Validation in French registration trials is planned for the 2026 campaign. A companion article by V. Cadot et al. is currently under review in the Journal of Experimental Botany special issue on Plant Phenomics & Enviromics Across Scales. The second, COYL (Counting Orange and Yellow Larvae), tackles a practical challenge faced by breeders, and rapidly counting wheat blossom midge larvae, both Sitodiplosis mosellana and Contarinia tritici, to characterise variety susceptibility. Using smartphone RGB images and YOLOv-based object detection, the best-performing model achieved high accuracy and successfully distinguished between the two visually similar species. An online counting application has been developed, currently accessible to Walloon Agricultural Research Centre members. The labelled COYL-1 dataset is publicly available at https://doi.org/10.5281/zenodo.19402333 for community use. A companion article by Antoine Deryck et al. is under submission at Plant Phenomics Journal.
Why it matches plant phenotyping methodsRGB画像と深層学習によりコムギ穂の病徴を検出・定量する手法の開発とセンサー／画像手法の検証が中心であり、植物病害表現型の取得に該当する。
abstractPHENET's Use Case 1 on plant health is validating sensors and imaging methods for the assessment of wheat ear diseases
Manually counting sunflower seeds on capitula is labor-intensive, requiring approximately one person-hour per head, and can be inconsistent for densely packed heads. Existing phenotyping approaches often depend on laboratory-based equipment, limiting their accessibility. In this study, we developed a benchtop image-based pipeline for rapid, non-destructive estimation of developed and aborted seeds on intact dried sunflower heads. A dataset of 1093 sunflower capitula was imaged under fixed indoor lighting, and individual seeds were annotated as developed or aborted. A YOLOv8m one-stage object detector was trained and evaluated using a counting-focused protocol, in which a single confidence threshold was selected on the validation set and then applied unchanged to an independent test set of 109 images. The baseline model was compared with recent YOLO variants and different augmentation strategies. On the test set, the model achieved a mean absolute count error of 61.3 seeds per image, a mean relative error of 12.0%, and an mAP50 of 0.18 at the locked confidence threshold of 0.15. Only 13.8% of test images had relative errors below 2%. Larger YOLO models and augmentation variants did not improve performance. These findings show that the proposed system provides approximate, non-destructive seed-count estimation under controlled imaging conditions, while highlighting the need for improved localization in dense regions and domain adaptation for fresh heads or field conditions. The annotated dataset and trained model weights are made available to support reproducible research.
Why it matches plant phenotyping methodsヒマワリ頭花の発達・不稔種子数という植物形質を、画像取得とYOLOによる推定パイプラインで定量化する手法を開発・評価しており、方法が研究の中心である。
abstractwe developed a benchtop image-based pipeline for rapid, non-destructive estimation of developed and aborted seeds on intact dried sunflower heads.
Reproduction assets foundThe authors state the source code is available on GitHub and the CVAT-annotated dataset is available via a public share link; the GitHub repository URL is explicitly provided and matches an allowed URL. The dataset link itself is not given, so only the code/checkpoint repository qualifies as an actionable public asset.Code · publicThe developed system is available as a Telegram bot [ 19 ] and the source code is available on GitHub [ 20 ]. The CVAT annotated dataset is available via a public share link.Open asset ↗lines:84-103Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 13 Sept 2026
Modern agriculture operates at an unprecedented crossroads, it must simultaneously accelerate crop yields to feed an expanding global population and adapt to the severe, fluctuating pressures of climate change, structural soil degradation, abiotic water deficits, and evolving biological threats. Historically, selecting resilient crop varieties and implementing field-scale management strategies relied extensively on destructive, labor-intensive, and fundamentally subjective visual metrics. This manual processing approach has long been recognized as the primary operational bottleneck in agricultural advancement.To bridge the gap between rapidly expanding genomic data and actual field performance, the systematic, non-destructive quantification of structural and functional plant traits, plant phenotyping, has emerged as a transformative frontier. By integrating high-throughput engineering, multi-scale remote sensing, deep learning, and advanced molecular biology, modern phenotyping transitions crop science away from qualitative estimation toward highly reproducible, multidimensional data frameworks. This Research Topic presents new advances in advanced 3D reconstruction and deep semantic segmentation at the seedling stage; amodal fruit segmentation, morphological extraction, and early water-stress diagnostics; high-throughput in-field seedling counting and dynamic density modeling; multimodal foundation models, network pruning, and intelligent phytoprotection; aerial and spaceborne remote sensing for canopy analysis and weed monitoring; plant physiology, functional spectroscopy, and functional genomics under abiotic stress; and automated diagnostics for real-time orchard scouting and vineyard management.Automating the characterization of complex spatial layouts under controlled or greenhouse environments is essential for early variety selection and early-stage structural evaluation. Several contributions within this volume provide key breakthroughs in navigating overlapping tissues, severe occlusions, and low-contrast edge regions. showcases how substituting standard convolutions with deformable convolutions enables deep neural networks to accurately isolate the main stem of mature, high-density crops like soybeans. This architecture overcomes the traditional challenges of color mimicry and severe occlusion by pods and leaves, achieving an outstanding mIoU of 90.58% and providing reliable indices for lodging resistance and structural yield modeling (R 2 = 0.9746).Accurately extracting fruit morphology under commercial greenhouse conditions remains heavily constrained by overlapping crop structures, foliage cover, and variable shadows. Simple semantic masks typically fail when a target fruit is partially blocked, leading to a loss of key volumetric data.To resolve the challenge of hidden boundaries, Li, Yin, et al. (2025) developed CGA-ASNet, a specialized RGB-D amodal segmentation network driven by a Contextual and Global Attention (CGA) module designed to restore occluded tomato regions. Trained on a high-fidelity synthetic greenhouse dataset (Tomato-sim) generated via NVIDIA Isaac Sim's Replicator Composer and optimized with a mean coordinate fusion algorithm for real-world validation, this architecture expands the network's receptive field to predict the complete, hidden circular forms of occluded tomatoes, achieving an F@0.75 score of 94.2 and an amodal mIoU of 82.4%. This proves that simulation-to-real (Sim2Real) domain pathways can successfully decode full physical volumes under dense commercial canopies.Complementing this structural restoration, Yang, Li, et al. (2025) designed an integrated diagnostic framework to identify early water stress dynamics in greenhouse tomatoes. Built upon an optimized YOLOv11n core, their system integrates adaptive kernel convolutions (AKConv) into the network backbone's C3k2 modules and implements a recalibration feature pyramid detection head based on the specialized P2 small-target layer. This combination achieved a 5.4% increase in mAP50-95 for identifying fine phenotypic parts. By applying automated geometric analysis to the extracted bounding boxes, the system extracts plant heights and petiole count with low relative errors, feeding these phenotypic parameters into a Random Forest classification routine that flags water-stressed plants with 98% accuracy to guide targeted, automated drip irrigation.Accurate plant stands during early vegetative stages represent the foundational metric required to establish true field emergence rates, validate seed vigor across diverse breeding blocks, and perform early yield predictions.To solve the challenges of small targets, extreme spatial density, and adjacent leaf overlap, Zang et al. (2025) designed DM_IOC_fpn, a wheat seedling counting framework that balances local and global contextual features. By structuring a point-annotated dataset and embedding a densityenhanced encoder module, their network balances micro-scale spatial limits with macro-scale canopy structures. Optimized through a combined loss function tracking counting, classification, and regression parameters, this architecture achieved low error scores (RMSE = 2.91; MAE = 2.23), outperforming standard object-detection benchmarks in complex field environments.At the same time, scaling up to real-time aerial monitoring required major reductions in model complexity to support resource-constrained edge computers on autonomous aerial platforms. Feng, Nie, and Li (2025) engineered an ultra-lightweight YOLOv8n variant tailored for real-time maize seedling counting from high-speed UAV RGB overflights. By reparametrizing RepConv with HGNetV2, they constructed a lean Rep_HGNetV2 backbone, integrated a Bidirectional Feature Pyramid Network (BiFPN) for multi-scale feature alignment, and implemented a Task Dynamically Aligned Detection Head (TDADH). This architecture compressed total model parameters by 47% and reduced weight sizes to 3.5 MB while maintaining a 96.5% detection accuracy and an ultra-fast processing speed of 146.3 FPS, paving the way for low-cost, real-time field scouting.Automated phytoprotection requires machine-vision architectures capable of generalizing across highly diverse species, complex field conditions, and varying computational boundaries. A significant subset of the published papers addresses these challenges through foundation model adaptation, multi-modal alignment, and efficient network compression.A major paradigm shift presented in this collection involves moving away from task-specific training and toward foundation model adaptation. Chen, Ruan, et al. (2026) introduce a novel architecture integrating the DinoV3 foundation model with a Unet framework to achieve robust leaf lesion segmentation across diverse species (such as coffee and black gram). By incorporating a Spatial Prior Module (SPM), their approach surpassed standard benchmark networks by over 10.5% in IoU while reducing inference times by approximately 93.6%, demonstrating that highparameter foundation models can be highly optimized for resource-constrained edge devices in real-time scouting.To solve the perennial problem of limited training data for rare or emerging crop diseases, Cooper et al. ( 2026) developed an ingenious synthetic data generation pipeline. Combining 3D procedural leaf modeling in Blender with diffusion-based disease synthesis (Stable Diffusion fine-tuned with LoRA and ControlNet), they synthesized highly accurate plant disease images with perfect groundtruth annotation masks. When deployed in low-resource data settings, combining these synthetic pipelines with restricted real-world datasets consistently drives significant improvements in downstream segmentation tasks. To tackle specific, complex pathologies, Xu, Chang, et al. (2025) developed the TSSC deep learning model, which embeds three-neighbor channel attention paired with a complementary squeeze-and-excitation mechanism. This specific architecture minimizes structural degradation risks while pushing classification accuracy to 99.61% for highly complex pea leaf pathologies. Similarly, Feng, Liu, et al. (2025) tackled overlapping leaf occlusions and small lesion footprints in citrus groves with YOLO-Citrus, an optimized framework integrating C3K2-STA, ADown modules, and a Wise-Inner-MPDIoU loss function to strike a balance between edge computational constraints and field deployment.UAVs and high-resolution satellite imagery have expanded the operational scale of phenotyping from individual pots to vast breeding blocks and commercial fields, allowing researchers to capture macro-dynamic parameters over time.In complex canopy systems that defy standard top-down aerial sensing, such as single-staked white Guinea yams, Iseki et al. (2026) demonstrated the distinct advantage of utilizing multi-angle (combined nadir and oblique) UAV imaging configurations. When coupled with support vector regression, this method captures complementary canopy-structure information to model shoot biomass trajectories (R 2 = 0.79) across multiple years and management zones. These nondestructive, time-series datasets enabled the fitting of genotype-specific Richard's growth curves using Bayesian inference, isolating valuable genetic variations in early growth allocation.To capture full-season vertical physiological changes over large scales, Li, Yue, and Luo (2025) developed a hybrid CNN-LSTM-Attention (CLA) model designed to estimate the full-period Leaf Area Index (LAI) in rice using multi-temporal UAV multispectral imagery. By using the CNN layer to extract instantaneous spatial features, the LSTM block to process seasonal time-series intervals, and a self-attention mechanism to weight critical growth transitions, their platform achieved a high coefficient of determination (R 2 = 0.92) and kept relative root mean square errors (RRMSE) below 9%. This network minimized soil background noise during early vegetative stages (LAI values 1-
Why it matches plant phenotyping methods植物フェノタイピングの技術動向を扱うEditorialであり、画像解析、UAVセンシング、深層学習、形質抽出などの方法が中心的に整理されている。
Three dimensional (3D) instance segmentation is essential for precision characterization of tree architecture at the branch level, which supports both tree fruit crop breeding and the development of robotic systems for orchard management. Existing methods usually use sparse convolution-based operation, which requires a coordinate quantization preprocess to generate sparse tensors, risking the loss of geometric details for fine-grained downstream phenotyping tasks. To overcome this challenge, we developed the dynamic sparse point-voxel transformer (DSPVFormer) model for the efficient and accurate 3D instance segmentation of high-resolution point clouds for dormant apple trees. Our hybrid DSPVFormer architecture maximizes the use of raw point features by dynamically mapping and aggregating the raw point features into the sparse voxel embeddings, capturing strong geometric features that may be discarded during quantization. Evaluations demonstrate that DSPVFormer achieved statistically significant improvements over baseline models on most instance segmentation metrics, which are further translated into more accurate phenotyping evaluation including branch counting and pruning map generation. These advances directly benefit downstream applications in plant phenotyping and robotic pruning for tree crops such as apples. Meanwhile, experimental results on phenotyping tasks suggested that phenotyping-specific evaluation metrics should be prioritized over upstream computer vision performance metrics to realize the full potential of high-throughput phenotyping for real-world applications.
Why it matches plant phenotyping methodsリンゴ樹の3D点群から枝レベル形質を抽出するセグメンテーション手法を開発・評価し、枝数や剪定マップへの応用まで検証しており、植物フェノタイピング手法が中心である。
abstractwe developed the dynamic sparse point-voxel transformer (DSPVFormer) model for the efficient and accurate 3D instance segmentation of high-resolution point clouds for dormant apple trees.
Reproduction assets foundThe paper states that its data and code (including the DSPVFormer analysis pipeline built on Plant Segmentation Studio) are publicly available at the authors' PSS GitHub repository. The COS dataset of 98 dormant apple tree point clouds is also referenced as accessible via this repository/statement. Other URLs (spconv,mCode · publicThe data and code are available at the PSS GitHub repository: https://github.com/perrydoremi/PlantSegStudio .Open asset ↗https://github.com/perrydoremi/PlantSegStudiolines:383-408Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
In-season estimation of wheat grain yield potential is critical for crop management and advancing breeding efforts. Spike and spikelet counts serve as key indicators directly linked to yield potential, yet their assessment still relies on manual counting which is both labor-intensive and error-prone. High-resolution digital (RGB) imagery combined with deep learning-based object detection methods has substantially advanced automatic wheat spike detection and counting. However, precise spikelet-level phenotyping remains largely underexplored. This study evaluates two recent YOLO variants, YOLOv11 and YOLOv12, for wheat spike and spikelet detection and counting using oriented bounding boxes (OBB), and introduces a new large-scale benchmark dataset comprising 48,521 spike and 60,404 spikelet instances with OBB annotations. For spike detection, the pre-trained YOLOv11 achieved superior accuracy (mAP@0.5 = 95.8%, Pearson r = 0.993) with shorter training and inference times compared to YOLOv12. For spikelet detection, the non-pretrained YOLOv11 demonstrated higher accuracy (mAP@0.5 = 99.0%), while counting performance was comparable across models. These results establish OBB-based YOLO detection as a robust and scalable approach for AI-driven wheat phenotyping.
Why it matches plant phenotyping methods小麦の穂・小穂という収量関連形質の画像ベース検出・計数手法を比較評価し、大規模ベンチマークデータセットも構築しているため、フェノタイピング手法が中心である。
abstractThis study evaluates two recent YOLO variants, YOLOv11 and YOLOv12, for wheat spike and spikelet detection and counting using oriented bounding boxes (OBB), and introduces a new large-scale benchmark dataset comprising 48,521 spike and 60,404 spikelet instances with OBB annotations.
Reproduction assets foundThe paper openly states its supporting data (spike/spikelet imagery with OBB annotations) is available on Zenodo, and the underlying models are deployed on the authors' public WheatAI cloud platform.Dataset · publicData availability
The data supporting the findings of this study are openly available at: https://doi.org/10.5281/zenodo.20215489 .Open asset ↗zenodo · 10.5281/zenodo.20215489lines:219-266Code / dataset availability confirmedOpenAlex · Europe PMC · checked 15 Sept 2026
The number of petals in an inflorescence is an important phenotypic indicator for quality evaluation and cultivar identification of cut chrysanthemums ( Chrysanthemum morifolium Ramat.). Current manual measurement methods are time-consuming, error-prone, and poorly suited to the complex geometry of chrysanthemum flowers, which limits their utility for large-scale phenotyping and breeding programs. Although image-based phenotyping has advanced rapidly, automated and reliable methods for petal counting in densely packed or partially obscured inflorescences remain underdeveloped. Here, we developed a deep learning-based framework for automatic extraction of petal number in cut chrysanthemums. Images from multiple varieties were collected to construct a representative dataset, and petal density maps were generated through manual annotation with Gaussian kernel function. We employed a Congested Scene Recognition Network (CSRNet) enhanced with a Squeeze-and-Excitation (SE) channel attention mechanism (SE-CSRNet) for petal density estimation. Spearman correlation analysis revealed strong agreement between visible and actual petal counts (Spearman’s r=0.953, p<0.0001). Compared with the original CSRNet, SE-CSRNet reduced mean absolute error (MAE) and root mean squared error (RMSE) by 5.2% and 7.4%, respectively. Further optimization using regression fitting revealed that random forest achieved the best performance (MAE = 4.24, RMSE = 5.06, R 2 = 0.967), indicating reliable stability and satisfactory generalization under the conditions evaluated in this work. Application of the optimized model to two cut chrysanthemum varieties confirmed its practicality by successfully detecting reductions in petal number under high-temperature stress. Our results demonstrate that integrating dataset construction, deep learning–based density estimation, and machine learning optimization enables efficient and accurate prediction of petal number in cut chrysanthemums.
Why it matches plant phenotyping methods花弁数という植物形質を画像から自動抽出する深層学習手法を開発し、データセット構築、性能比較、検証、実用適用まで行っており、表現型取得手法が研究の中心である。
abstractHere, we developed a deep learning-based framework for automatic extraction of petal number in cut chrysanthemums.
Reproduction assets foundThe article states that some data (the chrysanthemum petal-counting dataset and related materials) will be available at the authors' public GitHub repository (qwsdfgz/petalscount), with other data available from the corresponding author upon reasonable request. The repository URL is explicitly provided by the authors,但Dataset · publicnctional components of bud-leaves and flowers in edible chrysanthemum (Chrysanthemum morifolium Ramat)
Horticulturae 11 5 2025 448
10.3390/horticulturae11050448
Appendix A
Supplementary data
The following is the Supplementary data to this article.
Multimedia component 1
Data availability
Some data will be available at this URL: https://github.com/qwsdfgz/petalscount . Other data are openly available from the corresponding author upon reasonable request.
Appendix A
Supplementary data to this article can be found online at https://doi.org/10.1016/j.plaphe.2026.100238 .Open asset ↗qwsdfgz/petalscountlines:602-636Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Single-molecule fluorescence in situ hybridization (smFISH) has emerged as a powerful tool for studying gene expression dynamics with unparalleled precision and spatial resolution in a variety of biological systems. Recent advancements have expanded its application to encompass plant studies, yet there remains a need for a simple and robust smFISH method adapted to plant tissue sections. Here, we present an optimized smFISH protocol, termed cryo-smFISH, for visualizing and quantifying single mRNA molecules in plant tissue cryosections. This method exhibits remarkable sensitivity, enabling the detection of low-expression transcripts, including long non-coding RNAs. By integrating a deep learning-based algorithm into our image analysis pipeline, our method enables precise assignment of RNA abundance in nuclear and cytoplasmic compartments. The method also enables robust integration with immunofluorescence, as cryosectioning enhances antibody penetration. This allows for the sequential visualization and quantification of both RNAs and endogenous proteins within the same cells. Finally, this study demonstrates the use of smFISH to validate single-cell RNA sequencing (scRNA-seq) expression patterns in plant tissues. By extending smFISH to plant cryosections, plant scientists will be able to exploit the full potential of quantitative transcript analysis at cellular and subcellular resolution.
Why it matches plant phenotyping methods植物組織向けcryo-smFISHプロトコルと画像解析法を開発し、RNA量を細胞・細胞内区画で定量する手法が研究の中心である。分子測定ではあるが、植物組織の状態を定量する方法として技術的貢献が明確。
abstractHere, we present an optimized smFISH protocol, termed cryo-smFISH, for visualizing and quantifying single mRNA molecules in plant tissue cryosections.
Reproduction assets foundThe authors deposit all data underlying graphs/heatmaps plus custom R/Python scripts and Cellpose segmentation models in a public GitHub repository specific to this paper. Third-party tools (FISH-quant, DeconvolutionLab2, Stellaris Designer) are generic and excluded.Code · publicAll custom code, including R/Python scripts and Cellpose segmentation models, is available at https://github.com/xuezhang911/zhang_et_al_smFISH_cyrosections .
Funding
This work was supported by Vetenskapsrådet (2023-03895), the Novo Nordisk Foundation (NFF24OC0093553 and NNF25OC0100533), and the Carl Tryggers Stiftelse (CTS 18- 325).
Acknowledgments
We thank A. Menkis for initial technical support with cryostat operation and Alexandre Berr for scientific feedback. We also thank memOpen asset ↗zhang_et_al_smFISH_cyrosectionslines:122-152Dataset · publictic ( Bolger et al., 2014 ). The raw gene-count matrix was obtained using the pseudoalignment software Kallisto ( Bray et al., 2016 ). RNA-seq reads were normalized as transcripts per million (TPM).
Data and code availability
The supplemental information and all data underlying the graphs and heatmaps presented are available at https://github.com/xuezhang911/zhang_et_al_smFISH_cyrosections .Open asset ↗zhang_et_al_smFISH_cyrosectionslines:106-121Code · publicech.com/stellaris-designer . For mRNA detection, the coding sequence of the target gene was entered into the program, which automatically generated a set of probes complementary to the target mRNA.
The sequences of the probes were then subjected to quality control using an automated local blast R script, available on GitHub at: https://github.com/xuezhang911/zhang_et_al_smFISH_cyrosections/tree/main/smFISHprobes . The smFISH probes used in this study and their respective fluorophores are shown in Supplemental Table 3 . The probes were diluted in Tris-EDTA buffer to a final stock concentration of 25 μM.
Cryo-smFISH
Sample preparationOpen asset ↗zhang_et_al_smFISH_cyrosectionslines:75-85Plant phenotyping relevance match · UnverifiedEurope PMC · OpenAlex · Crossref · checked 13 Sept 2026
Accurate tree counting from remote sensing data is essential for forest inventory, biomass estimation, carbon accounting, and ecological monitoring. However, existing approaches predominantly rely on airborne RGB imagery and often struggle in complex forest scenes where neighboring crowns exhibit highly similar textures and colors and where overlapping crown boundaries become ambiguous. To address this limitation, the LiDAR-derived Canopy Height Model (CHM) is introduced as a complementary modality that provides explicit cues on canopy height variation and vertical structure to support RGB-based analysis. Building on this, we propose BCAR-Net, a broker-guided RGB and depth (RGB-D) multimodal framework that couples bidirectional cross-modal interaction, adaptive tri-branch fusion, and auxiliary reconstruction within a two-stage optimization scheme. Specifically, a bidirectional cross-attention U-Net generates an intermediate broker RGB-D representation from paired RGB images and depth maps through symmetric bidirectional cross-attention between the two modalities and direction-aware gating. The original RGB image, depth map, and broker representation are then jointly encoded by three weight-sharing branches and adaptively aggregated by a spatial fusion gate for density-map regression. To regularize the fused latent feature, a multi-scale cross-attention reconstruction decoder provides auxiliary RGB and depth reconstruction supervision by querying multi-scale BCA-UNet encoder features through 2D cross-attention, and a reconstruction-oriented first stage replaces externally generated fused-image supervision, yielding a task-consistent optimization scheme. Experiments on the NEONTreeEvaluation benchmark show that BCAR-Net consistently outperforms single-modality settings and direct RGB-D concatenation multimodal baseline. Additional experiments on a public UAV RGB-LiDAR dataset provide a small-scale supplementary evaluation under a different acquisition setting, where BCAR-Net achieves modest but consistent improvements over RGB-only and depth-only baselines. These results demonstrate that the proposed framework offers an effective but computationally cautious solution for tree counting in complex forest environments.
Why it matches plant phenotyping methodsRGB画像とLiDAR由来データから樹木数を推定する深層学習手法を開発し、複数ベンチマークで比較評価しており、植物個体の計測手法が研究の中心である。
abstractwe propose BCAR-Net, a broker-guided RGB and depth (RGB-D) multimodal framework
Accurate detection of tobacco plants in complex field environments is critical for precision agriculture, crop monitoring, and yield estimation. Traditional manual counting methods are time-consuming, labor-intensive, and susceptible to environmental and subjective factors. In this study, we propose an improved YOLO11-based framework for automated tobacco plant detection, specifically designed to address challenges such as scale variation, dense distribution, and background interference. The framework integrates four key modules: the Edge-Enhanced Feature Stem (EEFS) to strengthen low-level feature extraction, the Multi-Scale Kernel Interaction (MSKI) to capture multi-scale contextual information, the Adaptive Weighted Feature Fusion (AWFF) to optimize feature aggregation, and the Global–Local Synergistic Attention (GLSA) to enhance feature discrimination by jointly modeling local details and global context. A comprehensive UAV-based tobacco dataset was constructed, encompassing multiple lighting conditions, collection heights, and observation angles. Experimental results demonstrate that the proposed method significantly outperforms the YOLO11 baseline and achieves superior performance compared to mainstream YOLO variants. Ablation studies and heatmap visualizations confirm the effectiveness of each module. Furthermore, the model exhibits robust performance under multi-dimensional environmental perturbations, including varying illumination, scale, and camera angles. The proposed framework provides a practical and efficient solution for automated tobacco plant counting, offering potential applications in UAV-based precision agriculture and large-scale crop monitoring.
Why it matches plant phenotyping methodsUAV画像からタバコ個体を自動検出・計数する手法を開発し、専用データセット、比較実験、アブレーション、頑健性評価まで行っており、植物個体数の取得方法が中心です。
abstractwe propose an improved YOLO11-based framework for automated tobacco plant detection
To address the bottlenecks of low efficiency, poor consistency, and inadequate compatibility with high‑throughput phenotyping pipelines inherent in manual field‑based seed counting during soybean breeding, this study developed and validated an enhanced automatic soybean seed detection and counting model, YOLO‑Soy, tailored for complex field environments. Built on a YOLO11n backbone, the model integrates a Zoom multi‑scale feature fusion module, a C2PSA self‑attention enhancement module, a ScalSeq hierarchical feature sequence aggregation module, and a soybean-specific detection head. These additions systematically enhanced the saliency of tiny-seed features under dense occlusion and complex backgrounds and strengthened the capacity for foreground-background separation. Experiments were conducted using two‑year field imagery (2024–2025) and a year-stratified leave-one-year-out cross-validation strategy for training and validation. Ablation study revealed that the four improved modules are functionally complementary, forming a comprehensive pipeline of interference mitigation, scale adaptation, precise feature fusion, and detection output transformation. A single module exhibited limited effect when acting independently, whereas multi-module synergy produced substantial gains. Test-set results demonstrated that the seed counts predicted by the model were highly consistent with manual ground truth, achieving a coefficient of determination ( R ²) of 0.934, a mean relative error of 2.446%, a mean average precision (mAP@0.5) of 0.737, and an inference speed of 58.78 FPS. These metrics satisfy the requirements for real-time field detection. The findings indicated that YOLO-Soy can accelerate the seed‑counting step in variety selection processes, greatly reducing manual workload and subjective errors.
Why it matches plant phenotyping methods圃場画像からダイズ種子数を自動推定するYOLOモデルを開発し、交差検証・アブレーション・精度評価で検証しており、植物表現型取得法が研究の中心である。
abstractthis study developed and validated an enhanced automatic soybean seed detection and counting model, YOLO‑Soy, tailored for complex field environments
Mungbean (Vigna radiata (L.) R. Wilczek) is a vital source of digestible proteins and is well-suited for the plant-based protein industry. In this study, we analyzed pod morphological traits in the Iowa Mungbean Diversity (IMD) panel of 372 genotypes (2022-2023) using image-analysis-based phenotyping on 2,418 pod images. Pod morphological traits were extracted using deep learning image analysis, achieving excellent agreement with manual measurements (r > 0.96 for pod length (PL) and seed-per-pod (SPP)). Four complementary genome-wide association studies models identified 65 significant SNPs (-log10(P) ≥ 5.56) associated with pod curvature, length, width, and SPP traits. A significant SNP (5_35265704) on chromosome 4 was linked to pod dimensional traits, length, width, and curvature. A candidate gene, Virad04G0076900, located 15.6 kb from this SNP, is part of the GH3 gene family and has an Arabidopsis ortholog (AT4G27260) known for influencing organ elongation, pod, and seed development. Another SNP, 5_210437 on chromosome 6, has been found to be significantly associated with both PL and SPP. A candidate gene, Virad06G0002400 (36.5 kb from this SNP), encodes a potassium transporter and shares homology with the Arabidopsis gene HAK5 (AT4G13420), known to influence pod growth. Image-based measurements achieved genomic prediction accuracies ranging from 0.61 to 0.85 across various traits, demonstrating comparable accuracy to manual methods for linear traits and up to 22% improvement for complex shape traits. These results highlight the potential of deep learning-assisted phenomics integrated with genomic tools to accelerate selection for improved pod architecture in mungbean breeding programs across the Midwestern United States and globally.
Why it matches plant phenotyping methods深層学習による画像解析でマメ pod の形態形質を抽出し、手動測定との一致度を検証しており、画像ベース表現型取得が研究の中心です。
abstractusing image-analysis-based phenotyping on 2,418 pod images
Reproduction assets foundThe paper's image-based pod phenotyping and genomic analysis scripts are explicitly stated to be publicly available on the authors' GitHub repository. Raw phenotypic data (image-based and manual measurements, BLUEs, BLUPs, GP results) are only in supplementary files without a direct public URL, so they are not listed; Code · publicAll analysis scripts used in this study are publicly available at GitHub: https://github.com/vboddepalli89/Image-based-pod-phenotyping .Open asset ↗https://github.com/vboddepalli89/Image-based-pod-phenotypinglines:395-459Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Published31 May 2026International Journal for Research in Applied Science and Engineering TechnologyCited by 0 · OpenAlex ↗
Despite numerous efforts to incorporate emerging innovations into the agricultural domain for increasing crop yield and actively managing the state of the fields, it remains difficult for the industry to implement cutting-edge technologies in practice. This paper proposes AgroVision – a web-based intelligent multimodal system for comprehensive analysis of crops around the world. Designed using a three-layer scalable architecture, the system includes four modules – CNN-based growth stages and plant diseases recognition, AI Chatbot with LLM capabilities and RAG support, as well as the video analysis tool for detecting plant density and weeds. The key technology behind the core image analysis functionality of AgroVision is represented by the efficient Vision Mamba (ViM) architecture, which allows for analysing multiple tasks simultaneously using only one image uploaded by the user. Based on the extensive dataset called "New Plant Diseases Dataset" containing over 87 thousand images divided into 38 classes, the ViM model demonstrates exceptional results achieving weighted average F1-Score of 97.1%. Considering that the inference latency of the model does not exceed 25-40 milliseconds, the system can be deployed at the edge, providing an easy-to-use solution for farmers.
Why it matches plant phenotyping methods植物の成長段階、病害、植物密度を画像から推定するウェブ型解析プラットフォームを提案しており、表現型取得・推定が研究の中心である。
abstractThis paper proposes AgroVision – a web-based intelligent multimodal system for comprehensive analysis of crops around the world.
Blueberry production has rapidly expanded over the past decade, accompanied by growing demand for efficient and accurate methods to monitor the flowering and fruiting phases of blueberry development, which has a direct impact on yield potential. Accurate determination of blueberry phenology enables growers to make data-driven decisions on freeze protection applications and harvest windows. In addition, objective phenology data of blueberry mapping populations will provide high-quality phenotype data for the discovery of genetic mechanisms regulating blueberry flowering and fruiting times. Traditional approaches, such as manual counting and visual ratings, are labor-intensive and subjective in capturing variation across genotypes. Recent progress in computer vision and deep learning has enabled automated flower detection, but most existing studies on blueberries remain restricted to narrow flowering windows or close-up images, limiting their application at the bush level and across the seasonal development. In this study, we developed BerryFlowerNet, a customized YOLO-based model to detect and count blueberry flower clusters from bud to green fruit stages. A comprehensive dataset was collected on three dates using a field phenotyping robot, covering five flowering stages. The integration of CFNet, a custom module fusing shallow spatial features, and PIoU loss improved the detection performance. Additionally, the Slicing Aided Hyper Inference algorithm was employed to address small-object detection in bush-level images. Experimental results demonstrated that BerryFlowerNet outperformed the baseline YOLO model and three additional detectors, achieving an average mAP0.5 of 0.644 across five independent training runs. The model achieved an accuracy of 0.88 when predicting blueberry flowering stages, indicating its effectiveness and accuracy. Additionally, the results of the bush-level image analysis showed the capability of the model to capture genotype-level differences in flowering dynamics. Overall, this approach offers new opportunities for growers and breeders to determine blueberry phenological development that is critical for optimizing on-farm management strategies and advancing precision phenotyping to facilitate the development of climate-resilient blueberries.
Why it matches plant phenotyping methodsブルーベリーの花房検出・計数と開花ステージ推定を行うCNNおよびフィールド表現型ロボットの開発・評価が研究の中心であり、植物の生育状態を直接推定する実質的な表現型手法である。
abstractIn this study, we developed BerryFlowerNet, a customized YOLO-based model to detect and count blueberry flower clusters from bud to green fruit stages.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · Crossref · checked 5 Sept 2026
Rice is a significant food that plays a vital part in delivering nutrition to the world's population. Hence, approaches for assessing rice yield have received considerable study. The amount of rice seedlings (density) is a main agronomic module. It is related to harvest and also plays a significant part in the survival rate. Unmanned Aerial Vehicles (UAVs) are prepared with lightweight sensors, which creates a substantial effect in the field of crop phenotyping. The UAV was effectively used to measure germination rates and density in an accurate and effective method that would otherwise be laborious and expensive to obtain when compared to manual valuation. In image processing, mainly over the applications of deep learning (DL) models, there was a notable academic search for the value of UAV images for varied agricultural monitoring tasks. This work develops a Rice Seedlings for Assessing Germination Rates and Density using Aerial Images with Hierarchical Deep Network (RSAGRD-AIHDN) model. The goal of this paper is to assess germination rates and seedling density in rice fields using remote sensing (RS) or UAV-based imaging techniques for improved crop establishment monitoring. To accomplish that, the image pre-processing stage is initially applied with dual stages, such as image acquisition and pre-processing, to ensure high-quality and consistent inputs. Furthermore, the RSAGRD-AIHDN model employs the ConvNeXt method for the feature extraction process. For rice seed detection and classification, the RSAGRD-AIHDN model implements ensemble models, namely stacked autoencoder (SAE), bidirectional temporal convolution network (BiTCN), and Deep Q-Learning (DQL). The experimental assessment of the RSAGRD-AIHDN method is performed under the aerial dataset of rice seedlings. The experimentation of the RSAGRD-AIHDN method portrayed a superior accuracy value of 98.68% over existing approaches.
Why it matches plant phenotyping methodsUAV画像と深層学習モデルを用いて、イネの発芽率と苗密度という植物形質を推定する手法を開発・評価しており、表現型取得が研究の中心である。
abstractThe UAV was effectively used to measure germination rates and density in an accurate and effective method
Field / plotMultimodalWhole plant / canopy / plot / fieldClassificationCountingGrowth / development / phenology
Phenological monitoring of Actinidia chinensis is critical for optimising operational costs and yield prediction. However, current manual assessment methods are time-consuming, making them impractical for large-scale precision agriculture applications. Most existing phenological datasets focus exclusively on image data without spatial validation. The Multi-Modal Actinidia chinensis Phenology Dataset is composed of (i) 1 665 annotated images of phenological stages from bud to fruit set and (ii) georeferenced videos with systematic manual ground truth of spatial stage distributions. The dataset employs an adapted 17-class BBCH system that consolidates visually similar stages, excludes problematic categories, and introduces generic structural classes to address practical annotation difficulties. Additionally, the data is organised hierarchically across various plant structures, genders, and phenological stages. The annotated images offer versatility for a range of applications, including training data for computer vision models to detect phenological stages. Furthermore, the georeferenced videos facilitate the validation of automated counting algorithms. This combined approach enables plant-level detection accuracy and provides an illustrative methodology for spatial validation that users can extend to additional orchards, promoting the development and benchmarking of automated phenological monitoring systems for precision agriculture applications in kiwifruit production.
Why it matches plant phenotyping methodsキウイフルーツの生育段階を対象とした注釈画像・地理参照動画データセットであり、自動フェノロジー検出と空間検証のためのベンチマーク基盤が中心である。
abstractThe Multi-Modal Actinidia chinensis Phenology Dataset is composed of (i) 1 665 annotated images of phenological stages from bud to fruit set and (ii) georeferenced videos with systematic manual ground truth of spatial stage distributions.
Reproduction assets foundThe paper describes a public multi-modal Actinidia chinensis phenology dataset (annotated images, georeferenced videos, ground-truth counts) deposited on Zenodo, plus authors' MIT-licensed preprocessing scripts on GitHub. CVAT and FiftyOne are generic third-party tools and excluded.Dataset · publicThe Multi-Modal Actinidia chinensis Phenology Dataset described in this Data Descriptor is publicly available
at Zenodo: https://doi.org/10.5281/zenodo.17371025.Open asset ↗Zenodo · 10.5281/zenodo.17371025pdf-page:12 lines:1-92Code · publicCustom scripts for dataset preparation are publicly available under the MIT License at https://github.com/Open asset ↗GitHubpdf-page:12 lines:1-92Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
Field / plotFlowerClassificationCountingFruit / seed / panicle traits
As a crucial agricultural crop in China, litchi exhibits a biennial bearing pattern with alternating high-yield and low-yield cycles, known as on-year and off-year respectively. Research has identified unstable floral initiation as the primary cause of irregular fruiting in mid-to-late maturing cultivars. Rapid and accurate quantification of female to male flower ratios during the flowering phase enables targeted management strategies to optimize floral development and enhance fruit-setting rates. This study proposes Flower Quantification and Gender Recognition Network (FQGR-Net), a three-branch neural network architecture for simultaneous classification and counting of female and male flowers. Through module-level optimization, FQGR-Net improves both counting accuracy and computational efficiency, achieving average MAE of 8.498 and RMSE of 13.209 across categories in experiments conducted on the self-constructed dataset. Comparative experiments with other deep neural network models on public datasets show the proposed method achieves optimal performance. A regression analysis between predictions and ground truth produces R2 values of 0.930 and 0.971 for female and male flower quantification respectively. A dedicated litchi flower phenotyping analyzer was developed to address the technological gap in automated floral census systems. Field trials demonstrated over 80% accuracy in female/male flower counting.
Why it matches plant phenotyping methods雌雄花の画像ベース計数・性別認識手法と専用フェノタイピング解析器を開発し、データセットおよび野外試験で性能評価しているため、植物形質取得法が中心である。
abstractThis study proposes Flower Quantification and Gender Recognition Network (FQGR-Net), a three-branch neural network architecture for simultaneous classification and counting of female and male flowers.
Reproduction assets foundThe authors explicitly state that the code and data for this litchi flower quantification/gender recognition study are publicly downloadable from their GitHub repository. The Roboflow datasets are cited third-party comparison datasets, not paper-specific assets, and the litchi dataset itself is only available upon (un)Code · publicThe code and data mentioned in the article can be downloaded from https://github.com/Mazhe-02/Morphology-based-Litchi-Flower-Quantification-and-Gender-RecognitionOpen asset ↗Mazhe-02/Morphology-based-Litchi-Flower-Quantification-and-Gender-Recognitionlines:583-591Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
Grain number is a primary agronomic trait for targeted yield improvement, with the prospect of enhanced grain production leading to greater food security. Given the complex polygenic nature of the grain number trait, large sample sizes are essential for effective QTL identification. The implementation of trained computer vision models for grain detection offers a timely and cost-effective solution for rapid QTL isolation. In this study, we trained a grain detection model using Ultralytics' You Only Look Once (YOLOv11) framework. Training was completed on 1000 images of barley spikes, derived from a doubled haploid (DH) population descended from Hindmarsh and RGT Planet. The trained model, termed BarleyGC, achieved satisfactory accuracy metrics (mAP50-95 = 71.9%, recall = 96.7%, precision = 97.1%). Phenotypic characterisation of the DH population was completed with BarleyGC on a distinct collection of 973 images. The Pearson correlation coefficient (r) between model and manual-derived counts for the trait of grain number per spike was 0.895 ( p n = 153 DH lines), revealed a QTL peak at position 224.959 cM on the genetic map (LOD = 3.14), named qGN-2H. The QTL region contained 21 candidate genes-including HORVU2Hr1G092290 (HORVU.MOREX.r3.2HG0184740), encoding the six-rowed spike 1 ( Vrs1 ) gene-a well-characterised major regulator of row-type divergence and lateral spikelet development. Our study demonstrates the power of the YOLOv11 framework for grain quantification, with BarleyGC capable of grain detection directly from images of intact spikes in two-rowed barley varieties-thus achieving accelerated sample processing for the grain number trait.
Why it matches plant phenotyping methods大麦穂の画像から穀粒数を推定するYOLOv11モデルを開発・検証し、手動計数との比較およびQTL解析に応用しており、植物表現型取得法が研究の中心である。
abstractThe implementation of trained computer vision models for grain detection offers a timely and cost-effective solution for rapid QTL isolation.
Quantitative pollen viability analysis is a critical but labor-intensive step in plant reproductive biology. Existing deep-learning Segment Anything Models (SAM) fail to reliably segment viable pollen in Alexander-stained anthers. To address this, we fine-tuned an existing Cellpose-SAM model for pollen segmentation. We integrated it into PAT (Pollen Analysis Tool), a cross-platform desktop application. PAT features instance segmentation with interactive quality control, an in-app model retraining module, and publication-ready statistical outputs. We deployed PAT in an EMS suppressor screen of semi-sterile Arabidopsis smg7-6 mutants, enabling efficient candidate prioritization for whole genome sequencing and mapping candidate mutation. This screen led to the identification of a point mutation in CAP-D2 ( capd2-2 ), a Condensin I subunit, that rescues the smg7-6 meiotic phenotype. Notably, mutation in a Condensin II subunits (CAP-D3 and CAP-H2) does not confer rescue. Further characterization suggests the capd2-2 allele is hypomorphic, showing no defects in vegetative growth, chromocenter compaction, or transposable element silencing. Collectively, we demonstrate that accessible AI tools have the potential to bridge gaps in plant phenotyping and accelerate the pace of biological discovery. Highlight We combined AI-powered image analysis with an easy-to-use desktop app to automate plant pollen counting, then used it to identify a new genetic suppressor of meiotic defects.
Why it matches plant phenotyping methods花粉の生存性を画像から自動定量するセグメンテーション手法とソフトウェアPATの開発が中心であり、植物表現型の取得・抽出手法に該当する。
abstractWe integrated it into PAT (Pollen Analysis Tool), a cross-platform desktop application.
Accurate early-stage estimation of rice plant density is essential for precision crop management. However, current remote sensing methods face limitations in spatial resolution, revisit frequency, and sensitivity under sparse canopy conditions, highlighting the need for scalable, high-resolution UAV-based approaches. This study presents a UAV-based multispectral imaging framework for early-stage rice plant density estimation, proposing a scalable and cost-efficient solution for precision agriculture. Fractional vegetation cover derived from the Modified Soil Adjusted Vegetation Index (MSAVI) was used as the primary predictor variable in linear regression modelling. UAV imagery was acquired across varying flight altitudes (15–30 m) and crop growth stages (14–32 DAS). Five-fold cross-validation results shows that accuracy improved with crop development, with notable gains between 14 and 20 DAS. During the early vegetative stage, RMSE ranged from 39-41 plants/m2 and MAPE averaged ~30%, reflecting moderate predictive accuracy caused by sparse canopy cover and strong soil interference. As the crop progressed to early tillering, prediction error declined, with RMSE improving to approximately 30 plants/m2 and MAPE decreasing to about 29%. This improvement was attributed to denser canopy structure and stronger spectral separation between vegetation and background soil. Further analysis identified 18–25 DAS as the optimal developmental window for reliable plant density estimation, wherein models achieved high coefficients of determination (R² = 0.9139–0.9395) and the lowest RMSE (34 plants/m2). No significant differences were observed among flight altitudes, suggesting higher-altitude flights can maintain accuracy while improving operational efficiency and coverage.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像からイネの植物密度を推定する枠組みを開発・検証しており、植物形質の取得・推定手法が研究の中心である。
abstractThis study presents a UAV-based multispectral imaging framework for early-stage rice plant density estimation
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 5 Sept 2026
Introduction: Tiller production is a critical determinant of turfgrass canopy density and plant performance, yet manual tiller counting is too labor-intensive for large breeding programs. Methods: To address this limitation, we evaluated 770 plants from an interspecific bentgrass hybrid population and developed three automated approaches for tiller quantification: a classical edge-based segmentation pipeline and two deep-learning models, Faster R-CNN and YOLOv8. Using a large annotated image dataset, we assessed each method's accuracy, robustness under occlusion, and computational efficiency. Results: Although two-stage detectors are often expected to provide superior precision for complex plant structures, the one-stage YOLOv8 model achieved the highest accuracy (R² = 0.97) and processed images substantially faster than Faster R-CNN, while both the edge-based method and Faster R-CNN showed reduced performance in dense canopies. Discussion: These findings demonstrate that recall-oriented one-stage detection can outperform more complex two-stage models for phenotyping tasks involving fine, highly occluded structures. The resulting workflow provides a reliable, high-throughput solution for generating biologically meaningful tiller counts and offers a transferable framework for integrating image-derived phenotypes into genetic analyses and breeding pipelines across grass species.
Why it matches plant phenotyping methodsイネ科植物の分げつ数を画像から自動抽出する複数手法を開発・比較検証しており、植物表現型取得法が研究の中心である。
abstractdeveloped three automated approaches for tiller quantification: a classical edge-based segmentation pipeline and two deep-learning models, Faster R-CNN and YOLOv8.
Grain number estimation plays a crucial role in agriculture, serving as a key indicator for crop yield and quality assessment. With advances in computer vision, automatic grain detection has become a significant research area, where deep learning methods have shown remarkable promise. This study proposes a vision transformer model called Swin Transformer, which leverages hierarchical attention mechanisms across shifted windows to effectively capture both local and global features of grains in complex imagery. The model achieves the highest accuracy of 98%, outperforming baseline traditional CNN (ResNet-50) and DINO models in grain counting tasks. To support and validate model performance, explainable AI (XAI) techniques such as Grad-CAM and LIME are employed, highlighting the interpretability and focus of the model on relevant grain regions. Furthermore, a comprehensive empirical analysis is conducted using multiple statistical tests to evaluate the model's robustness and generalizability across various grain morphological parameters, establishing the Swin Transformer as a powerful and interpretable solution for intelligent grain counting in agricultural data analytics.
Why it matches plant phenotyping methods画像から穀粒数を推定する深層学習手法の開発・比較検証が研究の中心であり、植物の収量関連形質を測定するため、植物フェノタイピング手法として収録する。
abstractThis study proposes a vision transformer model called Swin Transformer
Reproduction assets foundThe paper's Data availability statement names a public Kaggle dataset of wheat grain counting images used for the study's grain counting experiments. No author analysis code or trained model checkpoints are disclosed.Dataset · publicThe dataset used and/or analyzed during the current study is publicly available at: https://kaggle.com/datasets/ociule/wheat-grain-counting-100-images.Open asset ↗kaggle · ociule/wheat-grain-counting-100-imageslines:1190-1253Plant phenotyping relevance match · UnverifiedEurope PMC · checked 14 Sept 2026
Accurate, high-throughput quantification of rice panicles plays a vital role in advancing precision yield prediction. However, transitioning to real-time, edge-deployable unmanned aerial vehicle phenotyping is often impeded by extreme spatial scale variations from altitude fluctuations and complex unstructured backgrounds. To address this, we constructed a comprehensive composite dataset specifically capturing multi-altitude and varying illumination field conditions. We then propose Panicle-DETR, a highly optimized precision phenotyping framework incorporating a frequency-aware CSP backbone. By projecting visual perception into the frequency domain, the architecture inherently suppresses low-frequency environmental noise and minimizes computational redundancy. Furthermore, a Lossless Feature Encoder prevents the irreversible pixel decimation of micro-targets across varying operational altitudes, while a composite metric loss explicitly disentangles heavily adhered panicle clusters. Evaluated on our composite dataset, Panicle-DETR achieved an outstanding detection Precision of 90.97% alongside robust agronomic counting stability, demonstrated by a Mean Absolute Error of 4.28 and an \( R^2 \) of 0.957. With a compact footprint of only 13.78 M parameters, this framework fundamentally overcomes the computational and spatial limitations of traditional vision models, establishing a highly reliable paradigm for autonomous, onboard agricultural monitoring.
Why it matches plant phenotyping methods米の穂を画像から検出・計数する軽量なUAV表現型解析フレームワークを開発し、複合データセット上で性能評価しており、表現型取得・抽出手法が中心である。
abstractwe constructed a comprehensive composite dataset specifically capturing multi-altitude and varying illumination field conditions.
Background: This study investigates the construction and optimization of the You Only Look Once (YOLO) deep learning model for high-precision identification of suitable tobacco leaves.(2) Methods: Using tobacco fields in Xiaoxin Street, Niulanjiang Town, Songming County, Kunming as the study area, a total of 1200 UAV images collected during the planting, growth, and harvesting stages were employed as the training dataset to train object detection models such as YOLO v3.After 200 training iterations, the recognition performance of each model was compared and analyzed.(3) Results: YOLO v5 and YOLO v7 were selected as baseline models, and a channel attention mechanism was integrated to develop the improved YOLO v5-EN model.Ablation experiments were conducted by incorporating the attention module, dynamic rectified linear unit (DReLu) activation function, and a feature refinement module.YOLO v7 en was designed as a backbone network, and metrics such as precision, recall, and accuracy were comprehensively evaluated to assess the performance of both the baseline and improved models in identifying the number of tobacco plants.Compared to the baseline, the improved YOLO v5 model demonstrated a 0.36% increase in precision and a 1.55% increase in recall, achieving an overall recognition accuracy of 91.41%.The improved YOLO v7 model achieved a precision of 99.16% and a mean average precision (map) of 95.86%.These results indicate that the enhanced YOLO v5 model with channel attention effectively addresses the issues of missed and false detections in tobacco plant recognition.Furthermore, the improved YOLO v7 model, integrated with collaborative optimization strategies and an enhanced backbone, significantly improves the performance and efficiency of the detection model, particularly in terms of accuracy and processing speed for complex visual tasks.(4) Conclusions: The improved YOLO models significantly enhance the accuracy of tobacco plant count recognition and offer a practical solution for efficient tobacco plant statistics, serving as a reference for intelligent agriculture.
Why it matches plant phenotyping methodsUAV画像からタバコ植物数を推定するYOLOモデルの改良・比較・アブレーション評価が中心であり、植物個体数という形態的状態の抽出手法を開発している。
titleEnhanced YOLO Architecture with Attention Mechanism for Accurate Tobacco Plant Counting from UAV Images
The tobacco plant counting is an important aspect in tobacco production management, traditional manual methods are time-consuming, labor-intensive and inaccurate, failing to meet the efficiency demands of modern agriculture. To enhance the accuracy and efficiency of tobacco plant counting in the field environment, this study utilizes high-resolution remote sensing imagery collected by drones to construct a sample dataset and proposes an improved YOLOv8-based target detection model (YOLOv8-CSD). YOLOv8-CSD model, based on YOLOv8, incorporates a coordinate attention mechanism (CA) to improve the extraction ability of the model to tobacco plant features. It also optimizes the feature pyramid network (FPN) and adds a small target detection layer to enhance the detection ability for the small target tobacco plants. Additionally, the shape intersection over ratio (SIoU) loss function is used to accelerate model convergence, and the slice-assisted hyper inference (SAHI) strategy is introduced to improve the accuracy and inference efficiency of small target detection by slicing high-resolution images. The experimental results show that the YOLOv8-CSD model achieves a precision of 97.96%, a recall rate of 97.93%, and an average accurate mean (mAP0.5) of 99.32%, significantly outperforming the original YOLOv8 and other 5 commonly used target recognition models. In addition, the efficiency of YOLOv8-CSD model is only lower than YOlOv11, indicating good overall performance. The YOLOv8-CSD model has good adaptability and robustness in tobacco plant detection at different growth stages, with a low missed detection rate, and the YOLOv8-CSD model can effectively meet the requirements of tobacco plant counting in complex field scenarios.
Why it matches plant phenotyping methodsUAV画像からタバコ個体を検出・計数する画像解析手法を開発し、複数モデルとの性能比較で検証しており、植物形態・個体数の取得が研究の中心である。
abstractproposes an improved YOLOv8-based target detection model (YOLOv8-CSD)
Abstract Sugarcane ( Saccharum spp.), a C4 plant, is a vital renewable biofuel and sugar source for industries worldwide. However, synchronizing flowering between parental lines often poses challenges for breeders, hindering effective crossbreeding efforts. This study aimed to develop a high‐throughput phenotyping (HTP) strategy to evaluate flowering‐related traits using vegetation indices (VIs) and other metrics alongside artificial intelligence (AI)‐based prediction methods. A total of 154 genotypes were planted in an augmented block design at the IAC sugarcane breeding station in Serra Grande‐BA, Brazil. Raw RGB (Red, Green, Blue) images were captured using a DJI Mavic 3 Enterprise drone during the plant cane (PC) and first ratoon (FR) crop seasons. These images were processed to create orthomosaics and compute metrics/vegetation index; subsequently, machine learning (ML) and deep learning pipelines for systematic analysis were developed. A convolutional neural network (CNN) model achieved promising results, with an accuracy rate of up to 84% in the flowering detection task. Additionally, flower counts from the CNN model showed a moderate correlation with field data, evidenced by an R 2 value of 0.72 at the onset and an R 2 value of 0.29 at the conclusion of the flowering season for the PC. This resulted in an overall average regression R 2 of 0.46 with a root mean square error (RMSE) of 13.80. Furthermore, an artificial neural network classification model reached a notable accuracy of 0.87 in differentiating genotypes based on their flowering response (early‐flowering vs. late‐flowering), utilizing VIs and digital model‐based metrics as input parameters. The ML regression model demonstrated performance levels of R 2 = 0.51 and RMSE = 8.06 for days to flag leaf emergence in PC and R 2 = 0.52 and RMSE = 7.93 for days to flowering in FR. These results highlight the potential of HTP strategies, utilizing orthomosaics and AI, to accelerate data collection and analysis, offering significant insights for breeding programs in sugarcane.
Why it matches plant phenotyping methodsドローンRGB画像、オルソモザイク、植生指数、AIを組み合わせた開花形質のハイスループット取得・予測手法を開発し、精度検証まで実施しており、フェノタイピング手法が研究の中心である。
abstractThis study aimed to develop a high‐throughput phenotyping (HTP) strategy to evaluate flowering‐related traits using vegetation indices (VIs) and other metrics alongside artificial intelligence (AI)‐based prediction methods.
Rice is a strategic commodity in supporting national food security. However, its productivity remains hindered by manual growth monitoring processes, climate change challenges, and limited human resources. This final project develops a seedling detection and counting system using the YOLO (You Only Look Once) algorithm, with aerial imagery input acquired from UAV (Unmanned Aerial Vehicle), presented through an interactive web-based dashboard. The dataset is enhanced with MIRV (mirror vertical) and MIRH (mirror horizontal) augmentation techniques to improve training data diversity. All experiments were conducted on three models: YOLO11n, YOLOv10n, and YO- LOv8n. Evaluation shows that the YOLO11n configuration using AdamW and a learning rate of 0.01 achieves mAP@50 of 0.592 and precision of 0.852. The system supports data-driven agronomic decision-making to anticipate crop failure risks, thus assisting large-scale rice field owners in monitoring seedling effectively and efficiently.
Why it matches plant phenotyping methodsUAV画像とYOLOによるイネ苗の検出・計数手法およびWebシステムの開発が中心で、苗数という植物状態を定量化しているため。
abstractThis final project develops a seedling detection and counting system using the YOLO (You Only Look Once) algorithm, with aerial imagery input acquired from UAV (Unmanned Aerial Vehicle), presented through an interactive web-based dashboard.
Image-based plant phenotyping depends on dense structural understanding of crops, yet pixel-level annotation remains expensive across species, organs, growth stages, and field conditions. General-purpose vision foundation models offer a natural route to label efficiency, but their web-scale pretraining objectives transfer weakly to agricultural imagery, where semantics are often determined by fine organ geometry inside repetitive, texture-dominated scenes. We introduce SPROUT, a diffusion foundation model for multi-crop plant phenotyping. SPROUT learns from 2.6 million unlabeled open-field images (MCD-2.6M) using a pixel-space Diffusion Transformer, and selects transferable features with a label-free effective-rank criterion over denoising timesteps. This design shifts pretraining from crop-based invariance to structure-preserving denoising, making the representation better aligned with dense phenotyping tasks. We evaluate SPROUT across dense phenotyping tasks, including organ segmentation, crop-weed parsing, depth estimation, and counting. SPROUT consistently improves over strong web-pretrained baselines, with the largest gains on dense structural prediction, and shows favorable label and compute efficiency compared with general-purpose and crop-specific foundation models. The source code and MCD-2.6M dataset are publicly available.
Why it matches plant phenotyping methods植物フェノタイピング向けの拡散基盤モデルを開発し、複数の作物・器官に対する画像ベースの構造推定タスクで評価しているため、表現学習法とデータセットが中心的な方法論的貢献である。
abstractWe introduce SPROUT, a diffusion foundation model for multi-crop plant phenotyping.
The existing fruit tree yield prediction methods mainly rely on fruit period images or long-term meteorological and soil data, which make it difficult to meet the needs of early yield prediction. In addition, the flowering period images contain complex spatial distribution and severe overlap between flowers, which makes it challenging to directly extract stable structural indicators related to yield. Most existing research has focused on simple statistical indicators such as the number of flowers, while the spatial clustering structure of flowers and their relationship with yield have not been fully explored. Therefore, this article proposes an early apple yield prediction based on flowering stage image thinning simulation characteristics. In this study, blossom images and fruit maturity yield data from 100 apple trees were collected, with flower mask images extracted through standardized image processing. First, the traditional DBSCAN clustering algorithm was enhanced by integrating a KDTree acceleration structure and an adaptive multi-scale mechanism, forming the adaptive multi-scale clustering algorithm (AMS-DBSCAN) to achieve efficient identification of flower clusters and individual flowers. Based on this, two flower thinning simulation strategies based on density and spatial uniformity were designed to model artificial thinning rules and construct multi-dimensional, interpretable phenotypic features. Then, the original statistical features were fused with strategy-generated features and optimized using Lasso. We compared multiple models including XGBoost, BPNN, and SVR for yield prediction. The experimental results showed that XGBoost achieved good predictive performance under the hybrid feature set (R 2 = 0.856, RMSE = 3.098), which was further improved to R 2 = 0.900 after feature optimization with Lasso. The results demonstrate that the proposed method enables reliable early yield estimation, providing a new reference for precision management and early decision-making in fruit tree cultivation.
Why it matches plant phenotyping methods花画像から花群・個体を抽出し、間引きシミュレーション由来の解釈可能な表現型特徴を構築する画像解析手法が研究の中心であり、収量推定に技術的に応用・評価されている。
abstractflower mask images extracted through standardized image processing
Pesticides are widely used in agriculture to control weeds, insects, and diseases that threaten crop yields. However, their extensive use raises concerns about environmental impacts, particularly in aquatic ecosystems, which are vulnerable to contamination through runoff and leaching. To assess the toxicity of pesticides to aquatic plants, we applied an optimized automated duckweed (Wolffia globosa) frond-counting tool based on the StarDist technique. Using this method, we tested twenty-eight commonly used pesticides, including herbicides, fungicides, and insecticides effects on duckweed growth. The herbicide paraquat showed the strongest growth inhibition (IC 50 50 = 384.2 ppb). Simazine and pendimethalin exhibited moderate toxicity, while glyphosate, triclopyr, and glufosinate showed lower toxicity. Surprisingly, metamifop did not inhibit duckweed growth up to the highest tested concentration (10 6 ppb). Isoprothiolane was the only fungicide tested that exhibited significant toxic effects on duckweed (IC 50 ≈ 924.3 ppb). All others, including azoxystrobin, hexaconazole, difenoconazole, picoxystrobin, tebuconazole, and cyproconazole, only inhibited plant growth at unnaturally high concentrations. Interestingly, cyazofamid promoted duckweed growth under the test conditions. Among 12 insecticides tested, 8 exhibited relatively low toxicity to duckweed (IC 50 > 10 5 ppb). Cypermethrin, carbofuran, fenpropathrin, and nitenpyram showed very low toxicity, with IC 50 values exceeding 10 6 ppb. Our results both enhance understanding of agrochemical toxicity and demonstrate the utility of automated, high-throughput quantification of W. globosa growth, providing a rapid and effective approach for pesticide toxicity assessment in aquatic environments.
Why it matches plant phenotyping methodsStarDistによるウキクサ葉状体の自動カウントと成長定量が、農薬毒性評価のための中心的な方法として用いられているため。
abstractwe applied an optimized automated duckweed (Wolffia globosa) frond-counting tool based on the StarDist technique
Cellulose synthase complexes (CSCs) play a central role in plant cell wall formation. Their dynamic behavior at the plasma membrane leads to the deposition of cellulose microfibrils into the apoplastic space, thereby shaping the architecture and mechanical properties of the cell wall. Although previous imaging studies have provided important insights into CSC dynamics and localization, standardized and reproducible workflows for quantitative measurements of CSC speed and density remain limited. Here, we present a reproducible live-cell imaging and analysis workflow for quantifying the speed and density of fluorescently labeled CSCs at the plasma membrane in Arabidopsis thaliana . The protocol integrates optimized spinning-disk confocal imaging, surface-based projection of z-stack recordings, automated detection of diffraction-limited CSCs foci, and kymograph-based speed measurements using freely available tools in Fiji. While selected steps, such as region of interest definition and parameter selection for spot detection or trajectory analysis, remain user-guided, these decisions are constrained to well-defined stages within an otherwise standardized pipeline, thereby reducing variability and improving reproducibility across experiments. The workflow has been validated across multiple tissues, reporter lines, genetic backgrounds, and perturbation conditions in Arabidopsis and enables robust comparative analysis of CSC dynamics. Beyond CSCs, this workflow is expected to be adaptable to other fluorescently labeled proteins that appear as diffraction-limited foci at or near the plasma membrane. Key features • Enables accurate CSC speed and density measurements during both primary and secondary cell wall formation using spinning-disk confocal time-lapse imaging. • Combines surface-projection, kymograph analysis, and high-throughput particle detection to quantify CSC dynamics even in crowded or low-signal plasma membrane regions. • Provides a standardized analysis workflow validated across multiple Arabidopsis genotypes, including inducible systems and mutant backgrounds that possess altered cell wall biosynthesis. • Applicable to any fluorescently labeled diffraction-limited foci at or near the plasma membrane, extending the workflow beyond CSCs.
Why it matches plant phenotyping methods植物細胞内のセルロース合成酵素複合体の速度・密度という観測可能な状態を、ライブイメージングと自動解析で定量する再現可能な手法を開発・検証した研究であり、方法が中心です。
abstractHere, we present a reproducible live-cell imaging and analysis workflow for quantifying the speed and density of fluorescently labeled CSCs at the plasma membrane in Arabidopsis thaliana .
Doubled haploid (DH) technology can fast-track crop breeding. Haploid induction yields haploids with only one set of genomes, which are usually sterile. Haploid fertility (HF) is the ability of haploid plants to set seed, and it is a critical bottleneck in DH pipelines. Genetic mechanisms to restore HF hold immense potential in DH crop breeding, yet its phenotyping remains manual, destructive, and inconsistent. While recent advances in imaging and machine learning have improved throughput for general plant traits, no curated image dataset exists for Arabidopsis thaliana that explicitly represents HF. Here, we present AutoSiQ, a dataset and baseline deep learning pipeline for automated HF quantification. AutoSiQ includes high-resolution scanned inflorescences annotated with a seven-class ontology encompassing green siliques, green fertile siliques, mature siliques, fertile siliques, cracked fertile siliques, cracked siliques, and flowers. This multi-class annotation scheme preserves biologically meaningful information beyond binary fertile/non-fertile distinctions, enabling reliable fertility estimation and future phenotyping applications. We release baseline object detection models (YOLOv5), trained using the AutoSiQ dataset, and evaluate their performance across confidence thresholds. Model predictions strongly correlate with manual counts, achieving R² up to 0.94 for total silique number estimation. We further demonstrate AutoSiQ's utility for automated haploid fertility rate (HFR) estimation and genotype discrimination between two contrasting genotypes (WT and bmf2 mutant). A longitudinal analysis identifies ~60 days after sowing (DAS) as the optimal harvest time for maximizing mature silique counts by balancing between the number of immature buds and silique shattering. By releasing both the dataset and baseline code, AutoSiQ provides a reproducible and extensible foundation for high-throughput fertility phenotyping in haploid Arabidopsis .
Why it matches plant phenotyping methodsハプロイド稔性を画像から定量するデータセットと深層学習パイプラインを開発・評価しており、植物フェノタイピング手法が中心である。
abstractHere, we present AutoSiQ, a dataset and baseline deep learning pipeline for automated HF quantification.
Reproduction assets foundThe paper's AutoSiQ dataset (annotated scanned Arabidopsis inflorescence images with seven-class silique ontology and manual fertility counts) is publicly deposited on Zenodo per the data availability statement. The YOLOv5 GitHub repository is a generic third-party library, not an authors' code asset.Dataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://zenodo.org/records/17905566 .Open asset ↗zenodo · 17905566lines:367-402Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Plant density at the wheat emergence stage is a fundamental structural attribute of agroecosystems, exerting strong control on early competition, resource use efficiency, and yield formation. While UAV-based counting approaches have been widely explored for visually distinct crops such as maize and cotton, accurate and scalable estimation of wheat seedlings remains challenging due to their small size, high spatial density, and spectral similarity to soil and residue backgrounds. Moreover, existing RGB-based UAV and ground imaging approaches face an inherent trade-off between spatial resolution, spectral sensitivity, and operational efficiency.Here, we propose MS²‑Net (Multi-altitude, Multispectral Seedling Network), a high-throughput Earth-observation framework that integrates multi-altitude multispectral UAV observations with deep learning to enable robust estimation of wheat plant density at the emergence stage. Field experiments were conducted across three major wheat-growing regions in China (Henan, Hebei, and Shaanxi), covering approximately 1,500 plots spanning large variability in sowing density, genotype, and early growth conditions. Multispectral UAV imagery (blue, green, red, red-edge, and near-infrared) was acquired at four flight altitudes (12, 15, 20, and 40 m), enabling systematic evaluation of the trade-off between spatial detail and mapping efficiency. High-resolution smartphone images collected synchronously at plot level provided accurate reference plant counts for model training and validation.All UAV data were radiometrically calibrated to surface reflectance and used to derive conventional vegetation indices (NDVI, GNDVI, NDRE, OSAVI, and a red-edge chlorophyll index) for spectral interpretability. Wheat plant density was estimated using a deep regression framework built on an EfficientNet-B6 backbone and enhanced with spectral-aware adaptation, spatial attention, and scale-consistent feature learning, allowing MS²-Net to exploit both multispectral information and multi-scale spatial patterns. Across five-fold cross-validation over regions and flight altitudes, MS²-Net achieved robust density estimation (R² = 0.86, RMSE = 37.20 plants m⁻², averaged across sites and flight altitudes), with red-edge and near-infrared bands contributing substantially to model stability across observation scales.Results demonstrate that multi-altitude multispectral UAV observations provide a practical balance between spatial resolution, spectral sensitivity, and survey efficiency, outperforming both ground-based imaging and RGB-only UAV approaches for early wheat stand assessment. By enabling rapid, field-scale and spectrally informed plant density mapping, MS²-Net provides a scalable pathway for operational agroecosystem monitoring, high-throughput phenotyping, and precision crop management under real field conditions.
Why it matches plant phenotyping methodsマルチスペクトルUAV画像と深層学習によってコムギの出芽期植物密度を推定する手法を開発・検証しており、植物形質の取得・抽出が研究の中心である。
abstractwe propose MS²‑Net (Multi-altitude, Multispectral Seedling Network), a high-throughput Earth-observation framework that integrates multi-altitude multispectral UAV observations with deep learning to enable robust estimation of wheat plant density at the emergence stage.
Precision agriculture is an essential approach to improving productivity, sustainability, and resilience in modern farming systems amid global changes. Drone-based monitoring represents a significant part of this agricultural transition because it can collect data across extensive regions and across multiple crop growth stages. Perennial cash crops are associated with farmers’ livelihoods and with links to global supply chains. Yield estimates for crops like cocoa and cashew are challenging because of the difficulty of measuring yields under very high canopy cover, erratic fruiting, and inefficient conventional in-field methodologies that involve excessive human error, labour, and time constraints. This paper proposes a hybrid deep learning model, GradTabViTNet, for multi-stage crop condition and yield estimation. The architecture integrates Vision Transformers (ViT) for geospatial attribute extraction, TabNet for attention-based tabular data analysis, CSRNet for accurate object counting in dense canopy environments, and Grad-CAM to enable interpretability by marking key regions in drone images. Classification and counting features are then combined using LightGBM regression to accurately estimate yield. Experimental evaluation using the cashew and cocoa datasets demonstrates that the proposed GradTabViTNet model outperforms existing methods, achieving 99.25% accuracy of 99.10%, precision 98.40%, recall, and 99.27% an F1-score of. The fusion of aerial monitoring with interpretable deep learning methods creates an extensible, stable approach for crop yield prediction, enabling sustainable agriculture, enhanced decision-making for farmers, and stronger food security through data-driven management of high-value perennial crops.
Why it matches plant phenotyping methodsドローン画像から樹冠下の作物状態と収量を推定する深層学習手法を提案し、カシューナッツ・カカオデータセットで既存手法と比較評価しているため、植物形質取得・推定法が中心である。
abstractThis paper proposes a hybrid deep learning model, GradTabViTNet, for multi-stage crop condition and yield estimation.
Rice early tillering characteristics are key indicators for high-yield breeding, with tiller number and tillering rate as core parameters. High-throughput, temporal, and precise monitoring of tiller numbers via drone digital imagery provides quantitative support for tillering trait screening in breeding, serving as an important auxiliary tool for smart breeding. However, during the early tillering stage, complex backgrounds (e.g., water bodies, soil) and small, dense breeding plots pose challenges to high-throughput rice plant extraction and accurate tiller number estimation. To address this, this study proposes a rice tiller number estimation method based on an improved Swin-UNet model and multi-feature fusion. A PSO-optimized XGBoost model was constructed for tiller number estimation by integrating selected features. Experimental results show that the improved Swin-UNet model achieved a segmentation accuracy of 92.5% (7.2% higher than U-Net), and the PSO-XGBoost model, using 12 features (10 morphological and 2 color), yielded R²=0.85 and RMSE = 0.35. Application verification on 576 untrained breeding plots generated tiller number thematic maps, providing data support for germplasm tillering trait identification and advancing smart breeding.
Why it matches plant phenotyping methodsドローン画像からイネの分げつ数を抽出・推定する画像解析手法を開発し、セグメンテーション精度と推定性能を検証しているため、植物表現型計測が研究の中心である。
abstractthis study proposes a rice tiller number estimation method based on an improved Swin-UNet model and multi-feature fusion.
Abstract To address the issues of detail loss and matching difficulties in fruit tree 3D reconstruction caused by complex branch–leaf morphology, fruit occlusion, and illumination variations, this paper proposes an end-to-end cross-scale collaborative attention multi-view stereo network, termed MSA-MVSNet, for high-quality 3D reconstruction of orchard trees, while integrating semantic segmentation for fruit counting. A multi-scale feature enhancement module is designed to adaptively fuse deep semantic features and shallow fine-grained details through a spatial–channel collaborative attention mechanism, thereby enhancing the network’s capability to represent multi-scale structures such as trunks, branches, and leaves. Multi-branch dilated convolutions are introduced to enlarge the receptive field, and deformable convolutions are incorporated to adaptively capture the irregular geometric shapes of fruits, improving modeling robustness. In addition, a feature matching transformer is introduced to strengthen long-range global contextual correlations within and across images via intra-attention and inter-attention mechanisms, thereby improving matching stability in low-texture and repetitive-texture regions.To validate the effectiveness of the proposed method, experiments are conducted on self-collected real orchard dataset and public benchmark datasets. The results demonstrate that MSA-MVSNet outperforms baseline models by 8.2% in terms of 3D reconstruction quality. Finally, by combining depth filtering with the semantic segmentation results of YOLOv11-Seg, a semantic-guided fruit reconstruction and counting framework is constructed. This framework achieves an overall counting F1-score of 92.8% on the self-collected dataset with varying scene sparsity and 93.5% on the public Fuji-sfm dataset, demonstrating its effectiveness and generalization capability.
Why it matches plant phenotyping methods果樹の3D再構成と果実カウントという植物形質取得を目的に、マルチビュー再構成ネットワークとセグメンテーション統合手法を開発・検証しており、フェノタイピング手法が中心である。
abstractthis paper proposes an end-to-end cross-scale collaborative attention multi-view stereo network, termed MSA-MVSNet, for high-quality 3D reconstruction of orchard trees, while integrating semantic segmentation for fruit counting.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 5 Sept 2026
This study addresses the challenge of automated high-throughput phenotyping of wheat spike characteristics using modern computer vision and deep learning methods. Accurate estimation of spikelet number is a key indicator of plant productivity, yet traditional manual counting approaches are labor-intensive, slow, and difficult to scale to large breeding datasets. To overcome these limitations, we propose a spikelet detection strategy based on simplified point annotations, where an expert marks only the centers of spikelets rather than drawing detailed segmentation masks or bounding boxes. This significantly reduces annotation time and lowers the overall cost of preparing training datasets for machine learning models. To determine the most effective way of utilizing such simplified annotations, three computational methods were explored: segmentation of binary masks using a U-Net architecture, density regression based on two-dimensional Gaussian distributions optimized via Kullback-Leibler divergence, and detection of fixed-size bounding regions using the YOLOv8 object detection framework. The models were evaluated on dedicated test datasets using both quantitative metrics (MAE, MAPE) and spatial localization metrics (Precision, Recall, F1 score). The results demonstrate that U-Net-based approaches provide consistently high accuracy in spikelet localization and counting while maintaining robustness to annotation imperfections. In contrast, the YOLOv8-based method showed reduced performance, likely due to the geometric mismatch between fixed-size boxes and the natural elongated shape of spikelets. Overall, the proposed methodology highlights the effectiveness of combining minimalistic point-level annotation with advanced segmentation models for automating phenotyping workflows. This approach has the potential to accelerate breeding programs, enhance the efficiency of large-scale phenotypic data collection, and support further development of robust computer-vision tools for plant science applications.
Why it matches plant phenotyping methodsコムギの穂の小穂数をRGB画像から自動推定する画像解析・深層学習手法を開発し、複数モデルを定量評価しており、フェノタイピング手法が研究の中心である。
abstractThis study addresses the challenge of automated high-throughput phenotyping of wheat spike characteristics using modern computer vision and deep learning methods.
Accurate flower-load assessment is critical for informed thinning strategies in orchard management. UAV-based deep learning automated counting offers efficiency advantages, yet precise counting is heavily dependent on abundant annotated data, which is scarce and costly to obtain in agricultural settings. While semi-supervised learning alleviates dependency on manual annotation, its application to UAV-based orchard imagery faces challenges: complex backgrounds and small target sizes, which undermine pseudo-label reliability. To address these challenges, this study proposes a two-stage framework to achieve separate counting of apple flowers at different phenological stages. First, a color-SAM flower extractor (CSAM-FE) is proposed to preprocess images using a strategy combining color thresholding with the Segment Anything Model (SAM), suppressing background noise and extracting high-quality flower clusters, thereby providing purified inputs for the subsequent counting network. Second, an uncertainty-guided semi-supervised flower counting network (USCount-Net) is proposed for accurate stage-specific flower counting with limited labeled data. The USCount-Net incorporates two key components: an adaptive pseudo-label filtering (PLF) mechanism based on frequent forward uncertainty estimation (FFUE) is designed to dynamically suppress noisy gradient backpropagation, mitigating error propagation from unreliable pseudo-labels; and a noise-sensitive adaptive gated fusion (AGF) module is introduced to fuse cross-scale features without redundancy, addressing significant scale variations across phenological stages and observation angles. Comparative experiments on a self-built apple flower counting dataset demonstrate that USCount-Net achieves lower MAE and RMSE than state-of-the-art methods at 10%, 30%, and 50% labeling ratios. The results demonstrate that the proposed methodology serves as methodological support for rapid and precise apple flower counting in low-annotation agricultural scenarios.
Why it matches plant phenotyping methodsリンゴ花の画像抽出・計数手法と半教師あり解析ネットワークを開発し、データセット上で比較評価しているため、植物表現型取得が中心である。
abstractthis study proposes a two-stage framework to achieve separate counting of apple flowers at different phenological stages.
Reproduction assets foundThe paper's Data availability statement explicitly provides public access to the authors' USCount-Net source code on GitHub and the self-built apple flower counting dataset (UAV images, annotations, flower cluster images) on Google Drive.Code · publicThe source code is publicly available at https://github.com/haohuihui5019/USCount-Net . And the source dataset can be accessed at https://drive.google.com/drive/folders/1KP8H0qIuct56hWre5GV6ZJnzwOpen asset ↗USCount-Netlines:681-780Dataset · publicThe source code is publicly available at https://github.com/haohuihui5019/USCount-Net . And the source dataset can be accessed at https://drive.google.com/drive/folders/1KP8H0qIuct56hWre5GV6ZJnzwOpen asset ↗lines:681-780Plant phenotyping relevance match · UnverifiedEurope PMC · OpenAlex · checked 15 Sept 2026
TeaField / plotNeRF / 3D Gaussian SplattingLiDAR / point cloudLeafWhole plant / canopy / plot / fieldCountingObject detection2D/3D reconstructionSegmentation
Accurate perception of tea buds is a fundamental prerequisite for intelligent and precise tea harvesting planning. However, in real tea plantation environments, reliable harvesting-oriented perception at the planning level remains highly challenging due to the small size of tea buds, severe occlusion, complex background clutter, and the lack of accurate three-dimensional spatial information. To address these challenges, we propose TeaNeRF, an integrated three-dimensional visual perception pipeline designed for harvesting-oriented tea bud analysis. Instead of treating detection, segmentation, and spatial analysis as independent tasks, TeaNeRF integrates sequential two-dimensional recognition, monocular depth estimation, and neural radiance field reconstruction into a coherent perception pipeline, allowing accurate spatial understanding of tea buds in complex natural scenes. It should be noted that the proposed integration is conducted at the perception-output level, where multiple modular components are connected through fixed interfaces, rather than through joint optimization or an end-to-end trainable formulation. The proposed framework combines an enhanced YOLO-based detector, prompt-guided segmentation, and monocular depth priors to guide NeRF-based three-dimensional reconstruction. By incorporating depth supervision and semantic-aware neural fields, TeaNeRF generates dense and geometrically consistent point clouds with reliable semantic separation. Quantitative evaluations show consistent improvements in reconstruction fidelity, as reflected by increased PSNR and reduced LPIPS across multiple tea tree scenes. Based on the reconstructed semantic point cloud, a three-dimensional clustering and geometric fitting strategy is further developed to enable tea bud counting and harvesting-oriented candidate point estimation at the perception level. Experiments conducted on a real-world dataset of 4,700 tea plantation images demonstrate that TeaNeRF improves detection accuracy (mAP@50 = 91.7%), segmentation quality (IoU = 0.640), and overall three-dimensional perception performance. Case-level counting results on representative tea trees indicate that the proposed 3D semantic point cloud-based approach can provide feasible tea bud counting behavior and consistent spatial guidance cues for downstream harvesting planning. By providing structured three-dimensional spatial information, including tea bud locations, counts, and harvesting-oriented candidate points, TeaNeRF offers practical perception-level outputs for downstream planning in automated tea harvesting systems.
Why it matches plant phenotyping methods茶芽の検出・セグメンテーション・3D再構成を統合し、茶芽の計数と3D位置推定を行う知覚パイプラインが研究の中心であり、単なる収穫対象の局在化を超えた器官形質の抽出を含む。
abstractwe propose TeaNeRF, an integrated three-dimensional visual perception pipeline designed for harvesting-oriented tea bud analysis.
With the increasing cost of labor, smart agriculture has emerged as a key trend for the future of agricultural development. This paper presents an integrated approach for tomato maturity clas-sification and yield estimation using both RGB and multispectral images. The proposed approach consists of three main components: tomato detection, tomato tracking and counting, and maturity classification of tomatoes. YOLOv8 combined with OSNet is first employed to detect tomatoes, while StrongSORT is then adopted to track consistent identities across image sequences. For maturity classification, multiple vegetation indices, including NDVI, GNDVI, and GRRI, are first transformed using principal component analysis, followed by classification using support vector machines, k-nearest neighbors, and neural networks. Tomatoes are categorized into three ma-turity levels: immature, almost mature, and mature. Results demonstrate that the proposed ap-proach can effectively estimate yield of tomatoes at each maturity stage. This capability provides practical support for harvest planning and labor allocation in precision agriculture.
Why it matches plant phenotyping methodsRGB・マルチスペクトル画像からトマトの成熟度と収量を推定する画像解析ワークフローが中心で、果実状態および収量という植物形質を直接評価している。
abstractThe proposed approach consists of three main components: tomato detection, tomato tracking and counting, and maturity classification of tomatoes.
In wheat breeding, the number of spike grains is a key indicator for evaluating wheat yield, and timely and accurate detection of wheat spike grain is of great practical significance for yield estimation. However, in actual field production, the counting of spike grain still relies on manual counting after threshing, which poses problems such as complex measurement processes, time-consuming and laborious. At present, achieving automated and intelligent detection of wheat spike grain still faces significant challenge. Therefore, the focus of this study is to use the most advanced computer vision technology for fast and automatic detection of wheat spike grain. During the wheat filling stage, a total of 936 wheat spike grain images were collected, and these images were expanded through data augmentation to ultimately obtain 3700 wheat spike grain images. According to the partition ratio of the small scale dataset, 80% of the 3700 images are used for training, 10% for validation, and the remaining 10% for testing. This study selected six state-of-the-art deep learning models: YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, YOLOv8x, and Faster R-CNN. In all wheat spike grain test, YOLOv8n showed high precision, recall, mAP50, and mAP50-95, with values of 96.8%, 96.8%, 98.9%, and 58.4%, respectively. The precision of other models was 96.7% for YOLOv8m, 96.5% for YOLOv8s, 96.3% for YOLOv8l, 96.2% for YOLOv8x, and 95.7% for Faster R-CNN. YOLOv8n not only has a lower number of parameters, FLOPs, inference time, model size, and GPU memory usage, as well as higher detection precision in wheat spike grain counting tasks, fully meet the spike grain counting requirements of wheat breeding. The multi-scale feature fusion and lightweight computing of YOLOv8n help improve model performance, and its performance is better compared to other deep learning models. This study designed and implemented a WeChat mini program for wheat spike grain counting, so as to achieve automatic detection and counting of wheat spike grains, which provided valuable reference for grain detection, counting, and yield estimation of other crops.
Why it matches plant phenotyping methods小麦穂粒数という植物形態・収量関連形質を、画像と深層学習で自動取得・計数する手法が研究の中心であり、複数モデルの性能比較とアプリ実装も行っているため。
abstractthe focus of this study is to use the most advanced computer vision technology for fast and automatic detection of wheat spike grain
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 5 Sept 2026
Introduction: Accurate counting and spatial localization of soybean seeds-particularly Seeds Per Plant (SPP)-are critical for yield estimation and cultivar evaluation. In field environments, however, complex backgrounds, pod occlusion, and uneven grain filling make high-precision counting challenging, and traditional methods often struggle to balance accuracy and robustness. Methods: To address these challenges, this study proposes SoyCountNet, a deep learning framework for automatic soybean seed counting and localization at the single-plant level under field conditions. The model is built on a self-constructed field-based phenotyping platform and optimized using the lightweight Point-to-Point Network (P2PNet). For feature extraction, a VGG19_BN backbone and a Super Token Sampling Vision Transformer (SViT) module are employed to enhance local feature representation and global contextual understanding. During feature fusion, the Efficient Channel Attention (ECA) mechanism strengthens seed-related features while suppressing interference from leaves, stems, and soil. Furthermore, an improved loss function that combines point-distance constraints with overlap penalties enhances both counting precision and spatial consistency. Results: Experimental results demonstrate that SoyCountNet outperforms existing approaches on the field soybean dataset. It achieves a mean absolute error (MAE) of 4.61, a root mean square error (RMSE) of 6.03, and a coefficient of determination (R²) of 0.94. The model demonstrates consistent performance across the tested soybean cultivars, providing reliable SPP estimates within the evaluated dataset. Discussion: These findings indicate that SoyCountNet offers a reliable and scalable solution for precise soybean seed counting and localization in complex field environments. Its lightweight architecture allows deployment on intelligent agricultural platforms, supporting high-throughput phenotyping, yield prediction, and precision breeding, while providing a foundation for the future development of intelligent and sustainable agricultural technologies.
Why it matches plant phenotyping methods単一個体の種子数（SPP）を画像から自動計数・位置推定する手法を開発し、圃場データで性能評価しているため、植物表現型取得が中心である。
abstractthis study proposes SoyCountNet, a deep learning framework for automatic soybean seed counting and localization at the single-plant level under field conditions
To address the inefficiency and high cost of manual counting of tobacco leaves, this study proposes a UAV-based method for automatic leaf counting in field-grown tobacco using 3D point clouds and an improved PointNext network. Although UAV imagery has been applied to crop phenotyping, most existing UAV-based leaf-counting methods still rely on 2D images or hand-crafted features and rarely exploit 3D point clouds with dedicated leaf-level segmentation, which limits accuracy and robustness under leaf overlap, variable viewing angles, and complex field backgrounds. In this work, oblique UAV photogrammetry is used to reconstruct individual plants into 3D point clouds, and a segmentation network, SRW-PointNext, is developed by integrating an SCSA attention mechanism and a Residual-SegHead to enhance feature extraction and segmentation performance, while a re-weighted loss alleviates class imbalance. Leaf point clouds are then clustered using MeanShift to obtain leaf counts. Experiments on field-grown tobacco demonstrate that the proposed method achieves a point-cloud segmentation precision of 92.09%, a MIoU of 76.13%. Compared with the original PointNext baseline, SRW-PointNext increased MIoU and overall precision by 3.34% and 2.42% respectively. The final accuracy rate of leaf counting was 92.61%, effectively achieving accurate and stable leaf counting under actual field conditions, and providing technical support for digital management, yield estimation and seedling breeding in tobacco production.
Why it matches plant phenotyping methodsUAV三次元画像と改良セグメンテーション手法により圃場タバコの葉数を推定する方法を開発・検証しており、表現型取得が研究の中心である。
abstractthis study proposes a UAV-based method for automatic leaf counting in field-grown tobacco using 3D point clouds and an improved PointNext
Reproduction assets foundThe paper reports a UAV-based tobacco leaf counting method with an annotated 1000-plant point cloud dataset and SRW-PointNext code, both explicitly declared publicly available at author-provided Zenodo and GitHub URLs matching the allowed list.Dataset · publicData supporting the reported results can be found at: https://zenodo.org/records/15130271 .Open asset ↗zenodo · 15130271lines:531-564Code · publicThe code used in this study is available at: https://github.com/Nan20377/SRW-Pointnext.git .Open asset ↗github · Nan20377/SRW-Pointnextlines:531-564Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
Published21 Feb 2026International Journal of Applied Earth Observation and GeoinformationCited by 2 · OpenAlex ↗
• Proposes a UAV remote sensing framework for ratoon sugarcane seedling counting. • Achieves precise segmentation of canopy using unsupervised learning methods. • Integrates multiple feature methods and models; KBest-F with GBR excels (R 2 = 0.7641). • Reveals the core feature contribution mechanism based on SHAP analysis. • Compares deep learning methods and proves the reliability of the work. Accurate and efficient monitoring of seedling emergence is critical for early-stage crop management and yield forecasting in sugarcane production. To meet this practical demand for precise field phenotyping, this study developed a high-throughput phenotyping framework leveraging unmanned aerial vehicle (UAV) remote sensing data and machine learning. This framework addresses the critical agricultural challenges of inefficient manual counting and the need for plot-scale monitoring in sugarcane production by enabling high-throughput sugarcane seedling number prediction through the integration of UAV-acquired RGB and multispectral imagery. Specifically, the sugarcane canopy was accurately segmented from the background using K-means clustering, a step that enabled the extraction of canopy area and the generation of a mask for obtaining canopy-level average features (including vegetation indices and texture features). These features together form a comprehensive feature set. Subsequently, six different feature selection methods were used to optimize the feature set, and eight machine learning models were combined for training and evaluation. The results showed that the combination of Gradient Boosting Regression (GBR) and KBest-F feature selection method yielded the optimal prediction performance, with a coefficient of determination (R 2 ) of 0.7641, a root mean square error (RMSE) of 19.42, and a mean absolute error (MAE) of 15.93. Further analysis identified canopy area, the Normalized Difference Red Edge Index (NDRE), red edge contrast, and green entropy as core predictive features. They collectively contribute over 60% of total feature importance, and their synergistic effects support accurate seedling number estimation. This framework offers an efficient, scalable tool for plot-scale seedling monitoring, with substantial potential for precision field management of high-density crops.
Why it matches plant phenotyping methodsUAV画像、キャノピー segmentation、特徴抽出、機械学習を統合し、サトウキビ苗数という植物状態を圃場スケールで推定する高スループット表現型計測フレームワークが研究の中心である。
abstractthis study developed a high-throughput phenotyping framework leveraging unmanned aerial vehicle (UAV) remote sensing data and machine learning
We present a novel method for accurately counting winter wheat tillers based on RGB images from hand-collected samples. An efficient sample preparation method assembles wheat tillers into bundles from which individual tillers are robustly detected automatically, using classical image analysis. A custom-made user interface ('TillerCounter' program) allows adjusting the automatic detections interactively, which leads to highly accurate tiller counts comparable to the ground truth obtained by manual counting. The key contributions of our work include:1.An efficient method for imaging straw tillers based on bundle assembly.2.An extensive study of the obtained image quality and comparison with the ground truth data from manual counting.3.Demonstration of the approach's high accuracy using correlation analysis (Pearson correlation coefficient R = 0.973 compared to ground truth) and error analysis (root mean squared relative errors below 5 %).
Why it matches plant phenotyping methods小麦分げつ数という植物形態形質を、画像取得・古典的画像解析・専用ソフトウェアで自動推定し、手動計数を基準に精度検証しているため、フェノタイピング手法が中心です。
abstractWe present a novel method for accurately counting winter wheat tillers based on RGB images from hand-collected samples.
Reproduction assets foundThe paper's authors publicly released the TillerCounter GUI source code on GitHub, which implements the Hough-transform-based tiller counting analysis used in this study. The paper also cites original image/count data at Zenodo (10.5281/zenodo.14446564), but no Zenodo URL is present in the allowed URL list, so only theCode · publicThe source code of the TillerCounter GUI is given at https://github.com/agroscope-ch/TillerCounterGui. Original data is given at Zenodo repository: 10.5281/zenodo.14446564Open asset ↗agroscope-ch/TillerCounterGuihtml-lines:163-195Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Automated seed phenotyping has wide applications in research and agriculture and relies on easy-to-use platforms and pipelines. Seed phenotyping in the model species Arabidopsis thaliana poses a significant challenge due to the large number of tiny seeds produced by individual plants, which are difficult to manually separate and count. Manual counting methods are time-consuming and prone to user bias, particularly for subtle phenotypic changes. To address these limitations, we developed Samplify, a scalable, automated pipeline for seed segmentation and classification by integrating classical image processing techniques with Meta's Segment Anything Model. Samplify effectively segments Arabidopsis seeds, even in dense clusters where conventional methods fail. To demonstrate its versatility, we quantified the seed abortion occurring in interploidy crossings in Arabidopsis, often referred to as 'triploid block'. Samplify includes a random forest classifier trained on a set of computed seed shape features that enable the categorization of seeds into normal, partially collapsed, and fully collapsed seeds, automating the manual classification process. The tool, designed as a command-line application, significantly reduces manual annotation workload. Our validation across multiple datasets demonstrates high segmentation and classification reliability, making Samplify a valuable resource for the plant research community.
Why it matches plant phenotyping methods種子画像のセグメンテーション・分類による表現型抽出パイプラインを開発し、複数データセットで信頼性を検証しているため、植物フェノタイピング手法が中心である。
abstractwe developed Samplify, a scalable, automated pipeline for seed segmentation and classification
Wheat emergence rate and emergence uniformity are key indicators for evaluating seed vigor and sowing quality, and they play an important role in wheat growth and yield formation. Traditional methods for measuring emergence rate and evaluating emergence uniformity rely on manual assessment, which is inefficient, highly subjective, and unable to meet the demand for large scale, high efficiency, and precise acquisition of wheat emergence data. In this study, RGB images and a two-stage deep learning algorithm were used to extract and analyze seedling traits of 420 wheat varieties under two nitrogen levels, and the results were applied to genome wide association studies to elucidate the genetic basis. The two-stage algorithm integrates a Bidirectional Feature Pyramid Network, small object detection layer, large size image input, and FasterNet to improve detection and instance segmentation speed and accuracy. The proposed method achieved an emergence rate accuracy of 0.929, with R 2 = 0.914 and RMSE = 2.448 compared to manual measurements, and required less than 0.2 s per image for analysis. By employing this two-stage algorithm for processing and analysis, varieties (e.g., Gao8901 and ShiYou20) that consistently exhibited high emergence rates and uniformity under multiple nitrogen treatments were identified. Furthermore, genome-wide association study identified the major loci qEmergence rate-3A and qUniformity-6B governing seedling emergence rate and uniformity, which likely enhance wheat seedling traits by modulating energy supply or related signaling molecules. The emergence-rate and uniformity data generated by the two-stage algorithm significantly accelerated the discovery of relevant genes and enabled the identification of wheat varieties with high emergence rate and uniformity, providing valuable insights and practical references for high-quality breeding and gene mining.
Why it matches plant phenotyping methodsRGB画像と二段階深層学習による出芽率・均一性の自動取得手法を開発し、手動測定との精度比較および大規模品種適用を行っており、表現型取得法が研究の中心である。
abstractTraditional methods for measuring emergence rate and evaluating emergence uniformity rely on manual assessment, which is inefficient, highly subjective, and unable to meet the demand for large scale, high efficiency, and precise acquisition of wheat emergence data.
Reproduction assets foundThe authors openly provide test code, base models, and sample test data for the WS-YOLO two-stage wheat seedling phenotyping pipeline in a public GitHub repository. Raw phenotype datasets are only available upon request, so they do not qualify as public assets.Code · publicThe test code, base models, and sample test data are openly available in the GitHub repository: https://github.com/AIWheatLab/WheatSeedling.Open asset ↗AIWheatLab/WheatSeedlinghtml-lines:375-402Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
Introduction Key rice panicle traits (grain number, panicle length, grain dimensions, maturity) determine yield and quality, and high-precision/high-throughput measurement is critical for rice breeding. Traditional methods are. Methods A dataset of 5300 rice panicle images (loose/normal/dense types; milk/dough/full maturity/over-ripe stages) was constructed, with 3290 for training, 940 for validation, and 470 for testing. A deep learning pipeline integrating. Results The panicle length extraction achieved R²=0.9583, RMSE=5.69 mm. Grain counting R² values were 0.9799 (loose), 0.9551 (normal), 0.9278 (dense). Grain length R²=0.8823, grain width MAPE=6.64%. OPG-YOLOv8. Discussion This study provides a comprehensive, automated tool for rice panicle phenotyping, addressing occlusion challenges and bridging the gap between advanced models and breeding applications.
Why it matches plant phenotyping methodsイネ穂の画像から粒数・穂長・粒形などの形質を抽出する深層学習パイプラインを開発・評価しており、フェノタイピング手法が研究の中心です。
titleDeep learning-based methods for phenotypic trait extraction in rice panicles.
In recent years, excessive tillering caused by high temperatures during early growth has contributed to rice quality deterioration in warm regions of Japan. Accurate determination of midseason drainage timing is essential but remains difficult due to year- and cultivar-dependent variability. In this study, we developed a smartphone-based web application that estimates rice tiller number from canopy images and diagnoses the optimal timing of midseason drainage by comparing estimated tiller numbers with cultivar-specific target values. The system operates entirely on a smartphone using HTML5 canvas-based pixel extraction, JavaScript computation, and Google Apps Script-based backend processing. Field experiments conducted in Chiba Prefecture using three rice cultivars showed a strong linear relationship between estimated and observed tiller numbers (R 2 = 0.9439). The root mean square error (RMSE) was 42.6 tillers m -2 , with a consistent negative bias (-34.6 tillers m -2 ), indicating systematic underestimation. Considering typical tiller increase rates near midseason drainage (12.0-24.3 tillers m -2 day -1 ), these errors correspond to approximately 1-3 days of growth progression, which is acceptable for timing-based decision-making. Although the system does not aim to provide precise absolute tiller counts, it reliably captures relative growth-stage dynamics and supports threshold-based diagnosis. The proposed approach enables rapid, on-site decision support using only a smartphone, contributing to labor-saving and improved water management in rice production.
Why it matches plant phenotyping methodsスマートフォン画像からイネの分げつ数を推定する手法とWebアプリを開発し、圃場で精度検証しているため、植物表現型取得が研究の中心である。
abstractwe developed a smartphone-based web application that estimates rice tiller number from canopy images
The center-pivot irrigation system is a highly efficient water-saving agricultural system. However, it typically operates solely based on predefined paths, speeds, and water volumes, and cannot assess the true needs of the crops due to its inability to monitor their growth status. To address this limitation, we deployed cameras on the center-pivot irrigation system to establish a mobile phenotyping platform, and developed a maize tassel detection model based on the YOLOv11 architecture. Several key improvements were implemented to address the complex morphology and challenging feature extraction of maize tassels: The ODConv (Omni-Dimensional Convolution) module was introduced to more comprehensively capture dynamic tassel features through a dynamic convolution strategy; To suppress interference from complex field environments on detection results, the SE (Squeeze-and-Excitation) attention mechanism was embedded to enhance effective feature responses and reduce noise impact; The BiFPN (Weighted Bi-directional Feature Pyramid Network) structure was adopted to strengthen multi-scale feature fusion capabilities, further improving the model’s detection performance for tassels at various scales; To tackle the difficulty of tassel recognition and localization in field environments, the C3K2 module was fused with the CSAM (Cross-Slice Attention Mechanism) to construct a C3K2-CSAM module, achieving more precise tassel identification and localization. Experimental results demonstrate that the OSBC-YOLO model achieved a precision of 91.8 % and a mean average precision (mAP) of 85.7 %, while reducing the number of parameters, FLOPs, and memory usage by 8.1 %, 28.1 %, and 7.3 % respectively compared to the original model. Field tests conducted on the center-pivot irrigation system revealed an error rate of 5.87 % between the number of tassels detected by the model and the ground truth count. These results verify that the OSBC-YOLO model can effectively perform tassel counting in practical operating environments and possesses the capability for recognition of crop phenotypic traits. This system provides critical data support for intelligent variable-rate irrigation decisions, promoting the transformation of center-pivot irrigation systems from traditional “single-function operation” to an integrated “perception-decision-execution” mode.
Why it matches plant phenotyping methodsトウモロコシ雄穂の検出・計数を行う移動型画像表現型解析プラットフォームとモデルを開発・検証しており、植物形態形質の取得が中心的貢献である。
abstractwe deployed cameras on the center-pivot irrigation system to establish a mobile phenotyping platform, and developed a maize tassel detection model based on the YOLOv11 architecture
Tea shoot density monitoring is crucial for quality control and yield optimization in plantations. This study developed a UAV-based multispectral imaging framework integrating machine learning for automated tea shoot assessment. We collected 3,122 ground-truth samples across four categories (tea shoots, mature leaves, dry leaves, soil) using portable spectrometry, transformed five spectral bands into 23 vegetation indices, and applied feature selection to identify optimal predictors. A two-stage classification approach was implemented: Stage-1 eliminated non-tea background with 100% accuracy; Stage-2 classified growth stages using MLP, SVM, and XGBoost algorithms. MLP achieved superior performance with 97% F1-score for tea shoots and 96% for mature leaves, outperforming SVM (87%) and XGBoost (93%). Field validation across three plantations revealed distinct temporal patterns: reverse J-shape (early budding), bell-shape (active germination), to J-shape distributions (mature canopy). Spatial uniformity was quantified using the Tea Shoot Density Index (TSDI), with the most uniform plot showing Is = − 0.876 and NNI = 1.654. This framework enables rapid plantation-wide assessment, reducing manual sampling time from days to hours while providing actionable insights for precision fertilization and irrigation management, advancing sustainable tea production.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像と機械学習により茶芽の密度・生育段階を推定する手法を開発し、圃場検証も行っており、フェノタイピング手法が研究の中心である。
abstractThis study developed a UAV-based multispectral imaging framework integrating machine learning for automated tea shoot assessment.
Aiming at the hysteresis problem of traditional contact-type mechanical throughput detection method during combine harvester operation, this study proposes a multimodal data-driven online throughput prediction method. By building a multimodal sensor system that integrates vehicle-mounted cameras, GPS, grain moisture content, and feeding auger power sensors, a throughput prediction framework based on wheat ear biomass characteristics was established: Firstly, the WEC-MVFF wheat ear online counting density map estimation model is designed, and MobileViT is used to build a feature extraction backbone network. The multi-scale fusion module and the centralized conversion module are combined to realize the collaborative extraction of shallow texture features and deep semantic features of dense wheat ears in the field. Secondly, divide the image into regions of interest and establish a throughput prediction model based on multimodal information of wheat ear number (image) − moisture content (sensor) − travel speed. At the same time, a detection model based on feeding auger power is constructed as a comparison benchmark. Field tests show that the WEC-MVFF model maintains an average counting accuracy of more than 90 % under different wheat ear density, travel speed and light intensity conditions. The model’s online counting advantage is verified through ablation study and comparative tests with other counting models. The throughput prediction method achieved an MAE of 0.70 kg/s and 0.77 kg/s in test area 1 and 2 respectively, the first-order difference fluctuation was stable in the range of ±0.50 kg/s, and the prediction frame rate of 10-13fps met the real-time requirements. Compared with the single-modal image prediction method, the accuracy of the multimodal method considering moisture content was improved by 0.80 kg/s and 0.75 kg/s in the two test areas, respectively. Compared with traditional mechanical quantity detection methods, it has higher accuracy and stability while achieving early prediction, providing reliable feedforward information support for the intelligent control of harvesters.
Why it matches plant phenotyping methods小麦穂の画像計数とマルチモーダルセンサを用いて、穂密度・バイオマス特性および収量流量を推定する手法を開発・検証しており、植物形質の取得・推定が研究の中心である。
Semantic reconstruction of agricultural scenes plays a vital role in tasks such as phenotyping and yield estimation. However, traditional approaches based on manual scanning or fixed camera setups remain a major bottleneck, while active-mapping methods based solely on occupancy grids are too coarse for accurate trait estimation. To address this gap, we propose an active 3D reconstruction framework for horticultural environments using a mobile manipulator. The system integrates OctoMap with 3D Gaussian Splatting to enable accurate and efficient target-aware mapping. A low-resolution OctoMap provides probabilistic occupancy information for informative viewpoint selection and collision-free planning, while 3D Gaussian Splatting leverages geometric, photometric, and semantic information to optimize 3D Gaussians for high-fidelity scene reconstruction. We further introduce a robust mapping strategy that mitigates semantic segmentation and depth noise, together with a background pruning method that reduces memory and computational cost. We validate our framework across simulated, laboratory, and real greenhouse scenes, showing consistent improvements across three state-of-the-art Gaussian Splatting backbones. In simulation, where ground-truth geometry is available, our approach outperforms occupancy-based mapping in both reconstruction accuracy and runtime efficiency: compared with a 0.01m-resolution OctoMap, it doubles the fruit-level F1 score under noisy conditions while achieving up to a threefold reduction in runtime. Beyond simulation, novel-view synthesis quality also improves consistently in laboratory and real greenhouse environments, with PSNR and mIoU improving by up to 1.5 dB and 18%, respectively. Finally, the reconstructed semantic maps enable fruit counting and volume estimation with accuracies approaching 80%.
Why it matches plant phenotyping methods園芸ロボット向けの3D再構成・能動マッピング手法を開発し、果実の計数・体積推定という植物形質の取得に適用・検証しているため、フェノタイピング手法が中心的です。
titleOctoSplat: Hybrid OctoMap-Gaussian Splatting for Active Semantic Mapping and Phenotyping with Horticultural Robots
Reproduction assets foundThe paper's supplementary material is hosted on the authors' public project page (jrcuaranv.github.io/octosplat), and the authors state that all code and data are publicly available. The SimSense repository is a third-party depth-sensor simulator tool, not a paper-specific asset.Code · publicAll code and data are publicly available to facilitate reproducibility.Open asset ↗lines:59-163Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
The legume-rhizobium symbiosis is a cornerstone of sustainable agriculture due to its ability to facilitate biological nitrogen fixation. Still, real-time visualization and quantification of this interaction remain technically challenging, especially across different host backgrounds. In this study, we systematically evaluate the efficacy of the nitrogenase system nifH promoter (P nifH ) in driving expression of distinct fluorescent reporters; superfolder yellow fluorescent protein (sfYFP), superfolder cyan fluorescent protein (sfCFP), and various red fluorescent proteins (RFPs) within root nodules of determinate ( Lotus japonicus-Mesorhizobium japonicum ) and indeterminate ( Pisum sativum-Rhizobium leguminosarum ) systems. We show that P nifH -driven sfYFP and sfCFP yield strong, uniform, and reproducible fluorescence in nodules of both systems, facilitating reliable quantification of nodulation traits and strain occupancy. In contrast, RFPs including monomeric (mScarlet-I, mRFP1, mARs1) and multimeric (AzamiRed1.0) variants exhibited weak or inconsistent signals in pea. Notably, fluorescent labeling did not impair rhizobial competitiveness for root nodule occupancy, and P nifH -driven sfYFP and sfCFP reporters enabled robust multiplexed imaging in single-root and split-root assays. In the lotus, mScarlet-I worked robustly and facilitated a tripartite strain labeling system. Complementing our molecular toolkit, we established a deep learning-based analytical pipeline for high-throughput, automated quantification of nodulation traits, validated against standard ImageJ analysis. Altogether, our results identify P nifH -driven sfYFP and sfCFP as robust, broadly applicable reporters for legume-rhizobium symbiosis studies, while highlighting the need for optimized red fluorophores in some contexts. The integration of validated promoter-reporter constructs with state-of-the-art computational approaches provides a scalable framework for dissecting the spatial and competitive dynamics of plant-microbe mutualisms. Importance The legume-rhizobium symbiosis is central to sustainable agriculture through its capacity for biological nitrogen fixation, yet tools for real-time, quantitative visualization of this interaction remain limited. Here, we demonstrate that the nifH promoter (P nifH ) effectively drives expression of superfolder yellow (sfYFP) and cyan (sfCFP) fluorescent proteins in both determinate ( Lotus japonicus-Mesorhizobium japonicum ) and indeterminate ( Pisum sativum-Rhizobium leguminosarum ) nodules. These reporters enable robust, reproducible fluorescence without impairing rhizobial competitiveness, supporting multiplexed imaging and quantitative nodulation analyses. By contrast, red fluorescent proteins exhibited host-dependent variability, underscoring the need for improved red fluorophores. Integration of validated promoter-reporter constructs with a deep learning-based image analysis pipeline establishes a scalable framework for high-throughput assessment of nodule occupancy and symbiotic dynamics. This work provides a practical molecular and computational toolkit for dissecting plant-microbe mutualisms across diverse host systems.
Why it matches plant phenotyping methods植物根粒の蛍光可視化・定量法を複数宿主で評価し、深層学習画像解析パイプラインを標準法と比較検証しており、植物形質（根粒形成・占有）の取得手法が中心的です。
abstractwe established a deep learning-based analytical pipeline for high-throughput, automated quantification of nodulation traits, validated against standard ImageJ analysis
High-resolution UAV photogrammetry has become a key technology for precision agriculture, enabling centimeter-level crop monitoring and point-level plant localization. However, point-level maize localization in UAV imagery remains challenging due to (1) extremely small object-to-pixel ratios, typically less than 0.1%, (2) prohibitive computational costs of quadratic attention on ultra-high-resolution images larger than 3000 x 4000 pixels, and (3) agricultural scene-specific complexities such as sparse object distribution and environmental variability that are poorly handled by general-purpose vision models. To address these challenges, we propose the Additive Kolmogorov-Arnold Transformer (AKT), which replaces conventional multilayer perceptrons with Pade Kolmogorov-Arnold Network (PKAN) modules to enhance functional expressivity for small-object feature extraction, and introduces PKAN Additive Attention (PAA) to model multiscale spatial dependencies with reduced computational complexity. In addition, we present the Point-based Maize Localization (PML) dataset, consisting of 1,928 high-resolution UAV images with approximately 501,000 point annotations collected under real field conditions. Extensive experiments show that AKT achieves an average F1-score of 62.8%, outperforming state-of-the-art methods by 4.2%, while reducing FLOPs by 12.6% and improving inference throughput by 20.7%. For downstream tasks, AKT attains a mean absolute error of 7.1 in stand counting and a root mean square error of 1.95-1.97 cm in interplant spacing estimation. These results demonstrate that integrating Kolmogorov-Arnold representation theory with efficient attention mechanisms offers an effective framework for high-resolution agricultural remote sensing.
Why it matches plant phenotyping methodsUAV画像から個体位置を抽出する手法を開発し、個体数と株間距離という植物群落形質を推定しており、データセット構築と技術評価も中心的である。
abstractTo address these challenges, we propose the Additive Kolmogorov-Arnold Transformer (AKT)
High-throughput, low-cost phenotyping remains a critical bottleneck in wheat breeding, genetics, and crop management. This is particularly evident in the measurement of complex yield components (i.e., spike and spikelet counts), disease and grain-quality traits related to Fusarium Head Blight (FHB) and Fusarium-Damaged Kernels (FDK), and microscale physiological traits such as density and size of stomata and aperture. We introduce WheatAI (wheatai.net), an AI-powered web application designed to bridge the gap between advanced computer vision, AI and deep learning models, and high-throughput phenotyping (HTP) and practical agricultural applications. WheatAI v1.0 provides an accessible, browser-based interface that supports multiscale data ingestion from smartphones, Unmanned Aerial Vehicles (UAVs), and portable microscopes. The core functionalities of the platform include plot- and field-scale assessment via UAV- and smartphone-based wheat spike detection and counting, as well as smartphone-based spikelet counting. Additionally, it offers grain quality assessment through FDK ratio estimation and kernel morphometric measurements, such as length, width, and area, derived from smartphone images of kernel samples. For leaf-level analysis, WheatAI provides microscale phenotyping through automated stomatal counting, size, and aperture measurement from digital microscopy images. The system supports both single-image and bulk processing via a guided upload-and-run workflow. This platform is designed to reduce labor costs and rater subjectivity while accelerating field-to-lab decision cycles. By providing standardized, image-based outputs, WheatAI enables breeders, agronomists, and producers to implement high-throughput selection and precision scouting at scale.
Why it matches plant phenotyping methodsWheatAIは、画像から収量構成要素、病害関連形質、穀粒形態、気孔形質を抽出する高スループット植物フェノタイピング基盤そのものであり、方法・ソフトウェアの開発が中心です。
abstractWe introduce WheatAI (wheatai.net), an AI-powered web application designed to bridge the gap between advanced computer vision, AI and deep learning models, and high-throughput phenotyping (HTP) and practical agricultural applications.
ABSTRACT Stomata are microscopic pores that play a vital role in transpiration and gaseous exchange from leaf surfaces in plants. The stomatal density and size directly influence photosynthesis and hydrodynamics capacity. Conventional approaches for counting and determining stomatal density is labour-intensive and lack scalability. Although there are several AI-based stomata finder tools that were published in the last decade, existing models were trained on model plants like wheat, barley and Arabidopsis . Stomata in such model plants are generally elliptical, but applying a universal model to all plant species is not feasible due to their diverse morphological characteristics. Previous studies have suggested using the stomatal index to quantify the ratio between epidermal cells and total stomatal count. However, this approach can be difficult to apply consistently, as epidermal cell shape and size vary across plant species. Instead, we propose measuring stomatal density based on the number of stomata per total imaged pixel area in the captured images. In this study, a comparison between YOLOv12 and RF-DETR models were made for real-time stomata detection in normal and difficult-to-image and out-of-focus occluded images. The in-house training dataset consisted of images of 300 rice,100 barley and 50 sugarcane leaves that were captured against a dark background. YOLOv12 outperformed RF-DETR with higher mAP50:95 score. The models were trained with image augmentation for 300 epochs and YOLOv12 achieved a peak mean average precision of 98.5% and exceled at detecting stomata across abaxial and adaxial surfaces of leaves of both monocot and dicot plants. StomaQuant has also been shown to be effective for both epidermal peel and ethanol decolorised samples. Thus, StomaQuant can be used to effectively and efficiently estimate the stomatal density and size in a wide range of host plant species.
Why it matches plant phenotyping methods気孔の検出・密度・サイズ推定を目的とする深層学習画像解析手法を開発し、複数モデルおよび困難画像で性能比較・検証しており、植物表現型取得が研究の中心である。
titleStomaQuant: Deep Learning-Based Quantification for Stomatal Trait Assessment
AppleCottonPearField / plotNeRF / 3D Gaussian SplattingFruitCounting2D/3D reconstructionSegmentation
Rigorous crop counting is crucial for effective agricultural management and informed intervention strategies. However, in outdoor field environments, partial occlusions combined with inherent ambiguity in distinguishing clustered crops from individual viewpoints poses an immense challenge for image-based segmentation methods. To address these problems, we introduce a novel crop counting framework designed for exact enumeration via 3D instance segmentation. Our approach utilizes 2D images captured from multiple viewpoints and associates independent instance masks for neural radiance field (NeRF) view synthesis. We introduce crop visibility and mask consistency scores, which are incorporated alongside 3D information from a NeRF model. This results in an effective segmentation of crop instances in 3D and highly-accurate crop counts. Furthermore, our method eliminates the dependence on crop-specific parameter tuning. We validate our framework on three agricultural datasets consisting of cotton bolls, apples, and pears, and demonstrate consistent counting performance despite major variations in crop color, shape, and size. A comparative analysis against the state of the art highlights superior performance on crop counting tasks. Lastly, we contribute a cotton plant dataset to advance further research on this topic.
Why it matches plant phenotyping methodsNeRFと3Dインスタンスセグメンテーションを用いて作物個体・器官数を推定する画像ベース表現型計測手法を開発・検証しており、方法が研究の中心である。
abstractwe introduce a novel crop counting framework designed for exact enumeration via 3D instance segmentation.
Reproduction assets foundThe paper contributes a public infield cotton plant dataset (8 plants, ~150 iPhone images each, ground-truth boll counts, SAM instance masks) and states that source code, dataset, and multimedia are available at the authors' public project page, which is an allowed URL. The spectacularai GitHub URL is a generic third-pDataset · publicthat incorporates crop visibility
and mask consistency, enabling robustness against occlusions and annotation
discrepancies.
•
We release a public infield cotton plant dataset designed for 3D
rendering and cotton boll counting tasks.
The source code, dataset, and multimedia material associated with this project
can be found at
https://robotic-vision-lab.github.io/cropnerf .
II Related Work
II-A Image-Based Techniques
Image-based methods typically employ object detection to identify crops within
images. For example, Chen et al. [ 4 ] utilized multiple
convolutional neural networks (CNNs) to map input images to total fruit counts.
Similarly, Häni et al. [ 5 ] formulated crop counting as a
multOpen asset ↗lines:108-187Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Jan 2026Computers and Electronics in Agriculture.
Maize ear traits are critical indicators for elucidating yield formation mechanisms and are widely used in genetic studies. Traditional two-dimensional (2D) phenotyping suffers from planar analysis constraints, occlusions in single-view imaging, and limited robustness to mixed textures or curved ears. To address these issues and support germplasm archiving and breeding research, we developed MaizeEar3DPheno (MEP3D), a 3D point cloud-based method for quantifying maize ear phenotypic traits. A structured-light 3D scanning system equipped with a motorized rotary platform was designed to acquire point clouds of 30 maize ears from three different varieties. Preprocessing involved axis alignment via PCA, uniform downsampling, and removal of non-kernel regions. Following preprocessing, key phenotypic traits, including ear length, diameter, and barren tip length, were calculated from the processed point clouds. MEP3D integrated directional erosion with density-based clustering to achieve robust kernel segmentation and counting. A spatial analysis algorithm was further developed to locate kernel row arrangements from geometric features. The results demonstrated that the proposed method achieved high-precision cross-variety kernel counting, with a mean absolute percentage error (MAPE) of 0.91%, and a coefficient of determination (R²) of 0.9917 across all maize ears. Kernel row quantification was fully consistent with manual measurements, allowing extraction of row inclination and average kernel number per row. Ear length, diameter, and barren tip length estimation achieved R² values of 0.9864, 0.9871, and 0.9670, respectively, demonstrating robustness. The generated high-fidelity 3D phenotypic data supports automated evaluation of ear and kernel traits and facilitates in-depth analysis of spatial morphological characteristics.
Why it matches plant phenotyping methodsトウモロコシ雌穂の3D点群取得・処理・形質抽出法を開発し、カーネル数や穂長などを手動測定と比較検証しており、フェノタイピング手法が研究の中心である。
abstractwe developed MaizeEar3DPheno (MEP3D), a 3D point cloud-based method for quantifying maize ear phenotypic traits.
ABSTRACT Three‐dimensional measurement technology based on point clouds can effectively solve the problem of plant occlusion and is a hot research direction for plant phenotyping methods. Rapid and low‐cost 3D reconstruction and accurate 3D point cloud segmentation are two major challenges in 3D phenotyping technology. Taking watermelon seedlings as an example, we proposed a new phenotyping method that uses the Instant‐NGP for 3D reconstruction and the improved PointNet++ for 3D point cloud segmentation. We performed dynamic downsampling and filtering based on the point cloud scale and designed different phenotypic measurement methods for hypocotyl and leaf point clouds. To overcome the difficulty of measuring the hypocotyl caused by slenderness, curvature and inclination, we proposed a segmented stem 3D point cloud skeleton extraction algorithm. The experimental results show that our method achieved satisfactory measurement results for the seedling phenotypes of four growth stages. The detection accuracy of the number of cotyledon leaves and the number of true leaves both exceed 95% and the coefficient of determination ( R 2 ) of leaf area, hypocotyl length and stem diameter phenotypes are all beyond 0.8. The proposed method provides a novel, efficient and precise 3D plant phenotyping solution, with good application and promotion value.
Why it matches plant phenotyping methods3D再構成、点群セグメンテーション、骨格抽出、形質測定を統合した植物フェノタイピング手法の開発が中心であり、精度評価も実施している。
abstractwe proposed a new phenotyping method that uses the Instant‐NGP for 3D reconstruction and the improved PointNet++ for 3D point cloud segmentation.
Accurate estimation of rapeseed emergence requires reliable counting and spatially precise localization under field conditions. The video-based system, YOLO_DBSCAN_Track_RTK, integrates lightweight YOLOv11n detection, DeepSORT object tracking, Horizontal Adaptive Distance DBSCAN (HAD-DBSCAN) for crop-row clustering, a dynamic boundary-drift counting mechanism, and RTK-GNSS-aided georeferencing. Seedling centers are extracted frame by frame, clustered along the row direction with an adaptively estimated neighborhood radius and centroid-based inter-frame matching, then tracked and counted within a drifting spatiotemporal window. A calibrated projection chain links image coordinates to world coordinates by fusing visual trajectories with RTK reference points, thereby generating centimeter-level geospatial identities. Field experiments conducted on 12 videos covering 120 rapeseed varieties at the East Anhui Experimental Station of Anhui Agricultural University demonstrate strong performance: the detector achieves an AP of 93.6 % with a processing speed of 110 FPS; HAD-DBSCAN reaches 0.967 clustering accuracy while preserving row integrity under uneven density and delayed emergence; the tracking module attains a tracking accuracy (Pₜᵣ) of 92.5 %, a tracking precision (Pₘₜ) of 93.1 %, an ID switch rate (WID) of 7.4 %, and a counting precision (Pc) of 92.8 %; Geolocation yields a mean error of 2.84 cm with quasi-normal residuals centered near zero. These results establish a unified framework for efficient seedling counting and multi-temporal plant-level monitoring, enabling growth analysis to support high-throughput phenotyping.
Why it matches plant phenotyping methods圃場画像から菜種幼苗の検出・追跡・計数・高精度位置推定を行う方法を開発し、性能検証しており、植物表現型取得が研究の中心である。
abstractThe video-based system, YOLO_DBSCAN_Track_RTK, integrates lightweight YOLOv11n detection, DeepSORT object tracking, Horizontal Adaptive Distance DBSCAN (HAD-DBSCAN) for crop-row clustering, a dynamic boundary-drift counting mechanism, and RTK-GNSS-aided georeferencing.
Highlights This article proposes ISCF, a novel method for precise soybean pod and seed counting using a segmentation followed by a classification strategy. Compared to YOLO, ISCF offers faster inference, higher accuracy, and a more efficient pipeline for real-time applications. The proposed method focuses on the practical value of the lightweight design, making it deployable on edge devices for real-world use. The proposed method applies to the automated counting of seeds or fruits of various crops in controlled indoor environments, demonstrating strong generalizability and adaptability. Abstract. Accurate counting of soybean pods and seeds is essential for yield prediction, crop management, and variety improvement. However, existing automatic methods under controlled indoor conditions often exhibit limited computational efficiency, insufficient accuracy, and limited practical deployment for reducing manual workload. To address this, we propose an Indoor Soybean Counting Framework (ISCF), a lightweight deep learning framework that decouples localization and classification into two independent stages to count soybean pods and seeds. ISCF first performs precise segmentation of soybean pods using the proposed Indoor Soybean Segmentation Network (ISSN), followed by classification of the number of seeds per pod using a MobileNetV3-based architecture. Optimized for lightweight design, ISCF is well-suited to real-time deployment on edge devices. Experimental results demonstrate the superior performance of ISCF in soybean pod and seed counting tasks, achieving an AP 50 of 99.5% for pod segmentation, a mean absolute error (MAE) of merely 0.72, and an R 2 of 0.9942 for pod counting, and an MAE of 3.79 and an R 2 of 0.9573 for seed counting. Moreover, ISCF generalizes well to datasets from four additional crop species, underscoring its potential for a broad range of indoor crop counting and phenotyping applications. Keywords: Image classification, Image recognition, Instance segmentation, Lightweight network, Plant phenotyping, Soybean counting.
Why it matches plant phenotyping methods植物の莢・種子数という収量関連形質を画像から自動抽出する軽量深層学習フレームワークを開発・評価しており、表現型取得手法が中心である。
abstractwe propose an Indoor Soybean Counting Framework (ISCF), a lightweight deep learning framework that decouples localization and classification into two independent stages to count soybean pods and seeds
Accurate estimation of rapeseed emergence requires reliable counting and spatially precise localization under field conditions. The video-based system, YOLO_DBSCAN_Track_RTK, integrates lightweight YOLOv11n detection, DeepSORT object tracking, Horizontal Adaptive Distance DBSCAN (HAD-DBSCAN) for crop-row clustering, a dynamic boundary-drift counting mechanism, and RTK-GNSS-aided georeferencing. Seedling centers are extracted frame by frame, clustered along the row direction with an adaptively estimated neighborhood radius and centroid-based inter-frame matching, then tracked and counted within a drifting spatiotemporal window. A calibrated projection chain links image coordinates to world coordinates by fusing visual trajectories with RTK reference points, thereby generating centimeter-level geospatial identities. Field experiments conducted on 12 videos covering 120 rapeseed varieties at the East Anhui Experimental Station of Anhui Agricultural University demonstrate strong performance: the detector achieves an AP of 93.6 % with a processing speed of 110 FPS ; HAD-DBSCAN reaches 0.967 clustering accuracy while preserving row integrity under uneven density and delayed emergence; the tracking module attains a tracking accuracy ( P tr ) of 92.5 %, a tracking precision ( P mt ) of 93.1 %, an ID switch rate ( W ID ) of 7.4 %, and a counting precision ( P c ) of 92.8 %; Geolocation yields a mean error of 2.84 cm with quasi-normal residuals centered near zero. These results establish a unified framework for efficient seedling counting and multi-temporal plant-level monitoring, enabling growth analysis to support high-throughput phenotyping.
Why it matches plant phenotyping methods画像検出・追跡・クラスタリング・RTK測位を統合し、圃場での rapeseed 苗の計数と個体位置推定を技術的に開発・検証しており、植物表現型取得が中心である。
abstractThe video-based system, YOLO_DBSCAN_Track_RTK, integrates lightweight YOLOv11n detection, DeepSORT object tracking, Horizontal Adaptive Distance DBSCAN (HAD-DBSCAN) for crop-row clustering, a dynamic boundary-drift counting mechanism, and RTK-GNSS-aided georeferencing.
Accurately estimating the number of cotton bolls is vital for plant phenotyping, offering essential insights for both breeders and growers. This trait offers valuable phenotypic information on plant productivity and supports crop management decisions to optimize yield and profitability for growers. Manual counting of bolls in the field, however, is impractical because it is labor-intensive and time-consuming. This study presented a video-based cotton boll counting approach that integrated a transformer-based detector (RT-DETR) with multi-object tracking techniques. To prevent double-counting bolls across frames, two motion estimation methods, FlowFormer and TAPIR were explored to predict the movement of bolls between adjacent frames and a two-stage association process combining Intersection over Union (IoU) and Euclidean distances was developed to track bolls across time. To further enhance counting accuracy, a virtual counting line was introduced to reduce ID switch errors. Experimental results demonstrated the effectiveness of the RT-DETR model, achieving an mAP0.5 exceeding 0.93 for dense boll detection. Furthermore, both FlowFormer and TAPIR can be used for tracking cotton bolls in the videos while the tracking performance of the FlowFormer-based method was slightly higher than that of the TAPIR-based method with an MOTA of 73.36 % and an IDF1 of 79.89 %. The tracking approach integrating RT-DETR and FlowFormer exhibited a relatively strong correlation between the predicted and the ground-truth boll number with an R² of 0.60 and an MAPE of 14.34 % on multi-plant plots. In single-plant plots, the approach achieved a high correlation with an R² of 0.97 and a MAPE of 10.33%. These findings indicated the potential of the proposed approach as an effective, automated tool to support breeding programs and yield assessments in cotton production. Both the code and dataset can be accessed at: https://github.com/UGA-BSAIL/Dense_cotton_boll_counting.
Why it matches plant phenotyping methods綿花のボール数という植物生産形質を、動画検出・追跡とロボット収集で自動推定する手法の開発・評価が研究の中心であり、mAP、MOTA、IDF1、R²、MAPEによる技術検証も行っている。
abstractThis study presented a video-based cotton boll counting approach that integrated a transformer-based detector (RT-DETR) with multi-object tracking techniques.
The accurate estimation of soybean (Glycine max) stand establishment is essential for evaluating crop emergence and informing early‐season management practices. Recent advances in unmanned aerial vehicle (UAV) imagery and computer vision offer opportunities to automate plant population assessments; however, limited information exists on their accuracy in soybeans. This study evaluated two commercial UAV‐based plant counting platforms, a point‐based (convolutional neural network-derived) and a line‐based (Hough transform-derived) approach across two growing seasons, three flight altitudes (15.2, 45.7, and 91.4 m), and seven plant removal treatments, including a control (no removal). UAV imagery was collected at 7‐ to 10‐day intervals from 10 to 36 days after planting (DAP), and predictions were compared to manual on‐ground counts. The point‐based method provided the highest accuracy (within ±12% of ground‐truth; R² = 0.81) when imagery was collected between 14 and 20 DAP at 15.2‐m altitude. Accuracy declined beyond 27 DAP as canopy overlap increased. The line‐based method remained more stable across altitudes and later growth stages but consistently overestimated plant counts, particularly in dense and narrow row canopies. Incorporating on‐ground calibration areas improved accuracy by an average of 28% and up to 48% for the line‐based approach in narrow rows. Row spacing and plant removal patterns had minimal effects on prediction error, although short repeating gaps were poorly detected by the line‐based method. Overall, UAV‐based plant counts in soybean are feasible and dependable when flights are timed during early vegetative growth and supported by calibration, providing a practical tool for in‐season management and field‐based crop monitoring.
Why it matches plant phenotyping methodsUAV画像とコンピュータビジョンによるダイズ個体数推定を中心に、複数手法の精度比較、飛行条件評価、地上校正の効果検証を行っているため、植物フェノタイピング手法研究に該当する。
abstractThis study evaluated two commercial UAV‐based plant counting platforms
Sieve elements in the phloem transport carbon and small molecules, such as RNA and phytohormones, throughout the plant body. Understanding the physical dimensions of sieve elements and phloem tissue is thus crucial for predicting how much carbon can be moved at any given time. Quantification of sieve element diameters and areas has previously been performed using transmission electron microscopy, scanning electron microscopy, and light microscopy, but sieve element identification is difficult because the phloem is a heterogeneous tissue. The recently identified LM26 antibody labels a pectin in the sieve element cell wall, allowing the identification of sieve elements and the measurement of their properties, such as diameter, relatively quickly and the quantification of their number in cross sections using image analysis software. Here, we describe methods for immunolabelling sieve elements in fresh or fixed tissue embedded in polyethylene glycol or methacrylate. The protocol is broadly adaptable to various fixation and sectioning methods, provided they do not alter the structure of pectins in the cell wall.
Why it matches plant phenotyping methods師部篩要素の同定、画像解析による直径・面積・数の定量を可能にする免疫標識プロトコルが中心で、植物形態形質の取得手法を提供している。
abstractallowing the identification of sieve elements and the measurement of their properties, such as diameter, relatively quickly and the quantification of their number in cross sections using image analysis software.
The study focuses on utilizing plant leaf characteristics for plant identification and disease detection. Leaves are pivotal for gathering information about plants. The proposed model uses computer vision and smart agricultural technologies to discern venation and texture features in various plant leaves. This research utilized a modified dataset derived from the Flavia leaf image dataset, comprising images of 32 plant species. The dataset was divided into two subsets (one with 1907 images and another with 1000 images) to differentiate between tuned and untuned image processing. Techniques such as GLCM, LBP, Gabor filters, Fractal Dimension, and box-counting were employed to extract leaf texture features, including venation patterns. The study conducted four experiments with training and testing splits of 70/30 and 80/20. A novel method combining SVM with fractal dimension analysis was benchmarked against six classifiers (Random Forest, KNN, DNN, Naïve Bayes, Decision Tree, andSVM), achieving an impressive accuracy of 88% and a Fractal Dimension of 1.8709. This research holds significant potential for advancing digital and modern agriculture, particularly in the early detection of plant diseases and accurate plant identification.
Why it matches plant phenotyping methods葉の輪郭・葉脈・テクスチャ特徴を画像から抽出する手法の開発と分類器比較が研究の中心であり、植物器官の観測可能な形態特徴を定量化しているため含める。
abstractThe proposed model uses computer vision and smart agricultural technologies to discern venation and texture features in various plant leaves.
Accurate and efficient leaf trait measurement is essential for plant phenotyping, agronomy, and ecological studies. In this work, we introduce Leaf Analyzer, a novel open-source, fully automated computer vision-based tool for high-throughput leaf morphological trait measurement such as leaf area, dimensions, perimeter, count, and percent damage. Unlike existing methods that rely on strong foreground-background contrast or controlled imaging conditions, Leaf Analyzer employs an unsupervised clustering approach based on the K-means++ clustering algorithm and a novel Leaf Background Separation (LBS) feature, which combines the L∗ and b∗ channels from CIEL∗a∗b∗ color space and the saturation channel from HSV color space. The proposed method and the LBS feature can effectively distinguish leaves from the background across varying lighting conditions, leaf colors, and camera orientations. To evaluate the performance of the new software, we conducted comprehensive quantitative and qualitative comparison experiments with two widely used software tools - Petiole Pro and LeafByte, demonstrating that Leaf Analyzer achieves superior accuracy and consistency, particularly under challenging imaging conditions. Additionally, we explore methods to further enhance measurement precision, including leaf flattening and the integration of supplementary leaf features such as texture features and color specific features. Beyond leaf trait measurement, we showcase the versatility of Leaf Analyzer in a range of applications, including nondestructive plant phenotyping, seed counting, root trait analysis, leaf area measurement for petri dish-grown plants, plant projected silhouette area or crown projection area estimation, leaf damage assessment, and broader plant science applications, making it a valuable tool for researchers working in laboratory and field environments.
Why it matches plant phenotyping methods葉形態形質を自動抽出するオープンソース画像解析ツールの開発と、既存ツールとの定量比較検証が研究の中心であるため。
abstractIn this work, we introduce Leaf Analyzer, a novel open-source, fully automated computer vision-based tool for high-throughput leaf morphological trait measurement such as leaf area, dimensions, perimeter, count, and percent damage.
Reproduction assets foundThe authors state that the Leaf Analyzer source code, installer files, and all data (including evaluation images) used in this study are publicly available on their GitHub repository.Code · publicThe Leaf Analyzer source code, platform-specific installer files, and all data used in this study are publicly available on our GitHub repository at https://github.com/squashking/Leaf-Analyzer .Open asset ↗squashking/Leaf-Analyzerlines:239-277Dataset · publicAll the images used in the evaluation have been published on our Github repository ( https://github.com/squashking/Leaf-Analyzer ).Open asset ↗squashking/Leaf-Analyzerlines:134-155Plant phenotyping relevance match · UnverifiedCrossref · checked 5 Sept 2026
Accurate stand count and growth stage detection are essential for crop monitoring, since traditional methods often overlook field variability, leading to poor management decisions. This study evaluated the performance of the YOLOv9-small model for detecting and counting corn plants under real field conditions. The model was tested across three soil background types, two flight heights (30 and 70 m), and four corn growth stages (V2, V3, V5, and V6). Unmanned aerial vehicle (UAV) imagery was collected from three distinct fields and cropped into 640 × 640 pixels. Datasets were split into training (70%), validation (20%), and testing (10%) datasets. Model performance was assessed using precision, recall, classification loss, and mean average precision of 50% and 50–90%. The results showed that the V3 and V5 stages yielded the highest detection accuracy, with mAP50 values exceeding 85% in conventional tillage fields and slightly lower performance in gray/red-brown conditions due to background interference. Increasing flight height to 70 m reduced accuracy by 8–12%, though precision remained high, particularly at V5, and performance was poorest for V2 and V6. In conclusion, YOLOv9-small is effective for early-stage corn detection, particularly at V3 and V5, with 30 m providing optimal results. However, 70 m may be acceptable at V5 to optimize mapping time.
Why it matches plant phenotyping methodsUAV画像とYOLOv9を用いてトウモロコシの個体数・生育段階を検出し、飛行高度や土壌背景別に性能評価しており、表現型取得手法の評価が中心である。
abstractThis study evaluated the performance of the YOLOv9-small model for detecting and counting corn plants under real field conditions.
The AI revolution, advanced Graphics Processing Units (GPUs), and open-source platforms have enabled Machine Learning (ML) and Deep Learning (DL) algorithms to rapidly and accurately extract phenotypic features from imagery. Such advancements have led to phenotypic digitization and made rapid yield forecasting possible. Yield predictions are critical to assess the merit of genotypes to propel cultivar development. This trial followed a three-replicated Randomized Complete Block Design (RCBD) with 36 diverse sorghum genotypes in 2023 at Ashland Bottoms, Kansas. The field images were captured 6 m above using a DJI M300 drone at 90° nadir and 45° oblique angles. This research trained YOLO and Faster R-CNN (Detectron2) models to harness yield attributes from UAS field and lab images. The YOLO models outperformed the Faster R-CNN in detecting sorghum panicles, achieving a mean average precision at 50 % IoU (mAP@0.50) scores of 0.92-0.98, compared to 0.61-0.89 for Faster R-CNN. Panicle detection from field imagery showed a linear correlation of 0.86 with ground truth field panicle counts. Lab imagery analyses measured panicle area, seed counts, and seed area with correlation coefficients of 0.79, 0.94, and 0.25 with respective ground truth observations. Support Vector Regression (SVR), Random Forest Regression (RFR), and Decision Tree Regression (DTR) were used to predict yield with correlation coefficients of 0.74, 0.71, and 0.78, respectively, and SHapley Additive exPlanation (SHAP) analysis revealed panicle seed count as the primary driver of yield prediction. We observed YOLO models are well-suited for extracting yield-predictive features from pertinent images. Such features can then be incorporated into ML regression models to predict yield per se performance with greater accuracy. The GitHub link is provided in the Data availability section.
Why it matches plant phenotyping methodsUAS・実験室画像から穂数、穂面積、種子数・面積などの植物形質を深層学習で抽出し、検出精度を検証して収量予測へ利用する方法が研究の中心である。
abstractThis research trained YOLO and Faster R-CNN (Detectron2) models to harness yield attributes from UAS field and lab images.
Reproduction assets foundThe authors explicitly state that scripts, fine-tuned models, datasets, and sample images for this sorghum yield-forecasting study are publicly available on GitHub, matching the allowed URL exactly.Code · publicThe scripts, fine-tuned models, datasets, and sample images pertinent to this manuscript are available on GitHub at https://github.com/mbari78/DL_for_Sorghum_Yield_Prediction.git .Open asset ↗https://github.com/mbari78/DL_for_Sorghum_Yield_Prediction.gitlines:244-299Plant phenotyping relevance match · UnverifiedEurope PMC · OpenAlex · checked 15 Sept 2026
Phalaenopsis orchids are one of Taiwan's key floral export products, and maintaining consistent quality is crucial for international competitiveness. To improve production efficiency, many orchid farms outsource the early flask seedling stage to contract growers, who raise the plants to the 2.5-inch potted seedling stage before returning them for further greenhouse cultivation. Traditionally, the quality of these outsourced seedlings is evaluated manually by inspectors who visually detect defects and assign quality grades based on experience, a process that is time-consuming and subjective. This study introduces a smart image-based deep learning system for automatic quality grading of Phalaenopsis potted seedlings, combining computer vision, deep learning, and machine learning techniques to replace manual inspection. The system uses YOLOv8 and YOLOv10 models for defect and root detection, along with SVM and Random Forest classifiers for defect counting and grading. It employs a dual-view imaging approach, utilizing top-view RGB-D images to capture spatial leaf structures and multi-angle side-view RGB images to assess leaf and root conditions. Two grading strategies are developed: a three-stage hierarchical method that offers interpretable diagnostic results and a direct grading method for fast, end-to-end quality prediction. Performance comparisons and ablation studies show that using RGB-D top-view images and optimal viewing-angle combinations significantly improve grading accuracy. The system achieves F1-scores of 84.44% (three-stage) and 90.44% (direct), demonstrating high reliability and strong potential for automated quality assessment and export inspection in the orchid industry.
Why it matches plant phenotyping methods胡蝶蘭苗の画像から欠陥・根・葉の状態を抽出し、品質を自動評価する画像ベースの表現型推定手法を開発・比較しており、手法が研究の中心である。
abstractThis study introduces a smart image-based deep learning system for automatic quality grading of Phalaenopsis potted seedlings
This Research paper develops an automated Plant Health Monitoring system that leverages Convolutional Neural Networks (CNNs) to perform simultaneous leaf- level disease classification and leaf counting from plant images. The proposed pipeline uses a CNN-based feature extractor feeding two task-specific branches: a classification head that identifies healthy versus diseased leaves (and the disease type) and a counting head that estimates leaf number via a regression/segmentation approach. Input images are preprocessed with augmentation and normalization to improve robustness to lighting, occlusion, and background variation. The model is trained on a curated set of annotated plant images and adapted for efficient inference using transfer learning and lightweight architectures suitable for edge deployment. Results show the approach provides reliable disease detection and accurate leaf counts, enabling timely alerts and actionable insights for precision agriculture. The system aims to reduce manual inspection effort, speed up diagnosis, and support better crop-management decisions.
Why it matches plant phenotyping methods植物画像から病害状態と葉数を推定するCNNベースの取得・解析パイプラインが研究の中心であり、植物表現型の計測手法として明示的に開発・評価されている。
abstractdevelops an automated Plant Health Monitoring system that leverages Convolutional Neural Networks (CNNs) to perform simultaneous leaf- level disease classification and leaf counting from plant images.
Field / plotMultimodalRootWhole plant / canopy / plot / fieldCountingObject detectionStress / disease detectionGrowth / time-series analysisGrowth / development / phenologyRoot system architecture
• Presents a full-process review of image-based high-throughput plant phenotyping (HTPP). • Covers recent advances in platforms, sensors, deep learning, and field-level applications. • Highlights emerging methods like Promptable models, Digital Twins, and weak supervision. • Discusses deployment challenges including data scarcity and model generalization. • Proposes future directions: multimodal fusion, uncertainty modeling, and lightweight design. With the rapid global population growth and increasing challenges in sustainable agriculture, high-throughput plant phenotyping (HTPP) has become a vital tool for advancing crop breeding and precision agriculture. This review provides a comprehensive overview of recent technological trends in image-based HTPP, focusing on the integration of advanced sensors, automated phenotyping platforms, and deep learning techniques. We summarize the evolution of imaging modalities, including 2D, 2.5D, and 3D sensors, and their respective applications in phenotype acquisition. We then examine the progress of deep learning-based models in core phenotyping tasks such as stress and disease detection, growth monitoring, organ counting, root system analysis, and postharvest quality assessment. Special attention is given to the emergence of Transformer architectures, multimodal fusion strategies, weakly supervised learning, and prompt-based foundation models. Despite significant advancements, current HTPP systems still face several challenges, including high costs, limited generalization in open-field conditions, and the need for large-scale annotated datasets. To address these, we discuss potential solutions such as transfer learning, synthetic data generation via digital twins, lightweight deployment for edge devices, and uncertainty estimation for model interpretability. By highlighting key developments and open problems, this review aims to guide future research toward scalable, robust, and intelligent plant phenotyping systems that can operate reliably in real-world agricultural environments.
Why it matches plant phenotyping methods画像ベース高スループット植物フェノタイピングのセンサー、プラットフォーム、画像解析技術を包括的にレビューしており、方法論が中心です。
abstractPresents a full-process review of image-based high-throughput plant phenotyping (HTPP).
Accurate maize tassel counting on individual ridge plays a critical role in advancing maize breeding programs by providing insights into crop adaptability and agronomic performance. The current manual method of processing male inflorescences has low accuracy and high labor intensity. Moreover, the existing algorithms cannot be directly applied to the counting of an individual ridge maize tassels. An automated system with UAV-based RGB imagery is presented for maize tassel counting. The object detection model was developed based on You Only Look Once version 8-medium (YOLOv8m), which detects tassels. A double-step Otsu threshold algorithm (DSOTSUTA) was designed to extract individual maize tassel ridge, which eliminated the interference of two adjacent ridges on both sides. Individual ridge tassel counting was implemented by Deep learning based Simple Online and Realtime Tracking (Deep SORT). It assigned identifiers (IDs) to each tassel and filtered abnormal IDs by analyzing the displacement increments of IDs in consecutive frames eliminating errors caused by ID switching. The object detection model achieved a mean Average Precision (mAP) of 91.6 %. The DSOTSUTA was tested on 5340 images and effectively extracted individual maize tassel ridges. The system achieved a root mean square error (RMSE) of 22.14 tassels per video, a mean absolute percentage error (MAPE) of 8.43 %, and an accuracy of 92.23 %, signifying a mean absolute percentage error (MAPE) of 8.43 % between the predicted tassel counts and ground truth observations. These results indicate that this automated system has the ability to enhance the accuracy of individual ridge maize tassel counting in breeding programs.
Why it matches plant phenotyping methodsUAV画像からトウモロコシ雄穂を自動検出・追跡・計数する手法を開発し、精度評価まで行っており、植物形質取得が研究の中心である。
abstractAn automated system with UAV-based RGB imagery is presented for maize tassel counting.
Peanuts rank as the seventh-largest crop in the United States with a farm value exceeding $1 billion. Conventional peanut yield estimation methods involve digging, harvesting, transporting, and weighing, which are labor-intensive and inefficient for large-scale research operations. This inefficiency is particularly pronounced in peanut breeding, which requires precise pod yield estimations of each plot in order to compare genetic potential for yield to select new, high-performing breeding lines. To improve efficiency and throughput for accelerating genetic improvement, we proposed an automated robotic imaging system to predict peanut yields in the field after digging and inversion of plots. A workflow was developed to estimate yield accurately across different genotypes by counting the pods from stitched plot-scale images. After the robotic scanning in the field, the sequential images of each peanut plot were stitched together using the Local Feature Transformer (LoFTR)-based feature matching and estimated translation between adjusted images, which avoided replicated pod counting in overlapped image regions. Additionally, the Real-Time Detection Transformer (RT-DETR) was customized for pod detection by integrating partial convolution into a lightweight ResNet-18 backbone and refining the up-sampling and down-sampling modules in cross-scale feature fusion. The customized detector achieved a mean Average Precision (mAP50) of 89.3% and a mAP95 of 55.0%, improving by 3.3% and 5.9% over the original RT-DETR model with lighter weights and less computation. To determine the number of pods within the stitched plot-scale image, a sliding window-based method was used to divide it into smaller patches to improve the accuracy of pod detection. In a case study of a total of 68 plots across 19 genotypes in a peanut breeding yield trial, the result presented a correlation (R 2 =0.47) between the yield and predicted pod count, better than the structure-from-motion (SfM) method. The yield ranking among different genotypes using image prediction achieved an average consistency of 84.8% with manual measurement. When the yield difference between two genotypes exceeded 12%, the consistency surpassed 90%. Overall, our robotic plot-scale peanut yield estimation workflow showed promise to replace the human measurement process, reducing the time and labor required for yield determination and improving the efficiency of peanut breeding.
Why it matches plant phenotyping methodsロボット撮像と画像解析により圃場区画の落花生莢数・収量を推定するワークフローを開発し、検出精度や手動測定との整合性を検証しており、フェノタイピング手法が中心である。
abstractwe proposed an automated robotic imaging system to predict peanut yields in the field after digging and inversion of plots.
Plant density is an important variable for management and phenotyping of small-grain cereal crops such as wheat and barley. While many image-based estimation methods exist to replace laborious manual counting, most of them rely on empirical relationships that may not generalize well to different sites, growth stages, species and varieties. In this study, we propose a novel small-grain cereal plant density estimation method that uses leaf tip density dynamics derived from submillimeter-scale images acquired at 45° view zenith angle. This method contained two steps. In the first step, a P2PNet deep learning detection model was trained to estimate leaf tip count in a surface of known area to get the leaf tip density. An occlusion correction method was then applied on this density, leading to an estimation error of about 20% at critical growth stages. In the second step, a wheat leaf dynamic model was used to simulate the evolution of leaf tip density over thermal time as functions of several variables, including mean time of plant emergence, phyllochron and plant density. This model was then inverted using a lookup table approach to estimate plant density from leaf tip density dynamics. The results obtained on three test datasets indicated that two observations performed before the appearance of the second and third leaves could be sufficient to attain a relative plant density estimation error of about 10%. We also discussed that this method should be able to work on other datasets without recalibration, and estimate other variables such as phyllochron at early growth stages. The code will be available at: https://github.com/wdwzytc/WheatPlantDensity.
Why it matches plant phenotyping methodsRGB画像から葉先密度を抽出し、植物密度を推定する画像ベースの表現型計測法を開発・評価しており、方法が研究の中心である。
abstractwe propose a novel small-grain cereal plant density estimation method that uses leaf tip density dynamics derived from submillimeter-scale images
The synergistic development of Internet of Things(IoT), robotics, and Artificial Intelligence (AI) is reshaping the technological paradigms of interdisciplinary laboratories and industrial ecosystems. IoT-enabled vertical farming systems demonstrate significant advantages, achieving yield enhancement while reducing carbon emissions compared to traditional agriculture, thereby providing innovative solutions for sustainable food production. The advancement of robotic technologies further expands the application dimensions of mobile intelligent sensors in vertical farm IoT networks. Based on an autonomous farming system that integrates Unmanned Aerial Vehicle (UAV), sensors, and modular vertical farming units, this study proposes a three dimensional Scene Graph (3DSG)-based hierarchical mapping method for the dynamic monitoring of plant and fruit growth. Through feedback mechanisms, the system optimizes growth conditions by adjusting lighting and nutrient delivery, while the hierarchical mapping architecture reduces detection errors and enables comprehensive 3D visualization. The main contributions of this research include: 1) Pioneering application of 3DSG technology to establish a multi-dimensional spatiotemporal representation model for plant growth processes, supporting interpretable analysis and traceable monitoring; 2) Establishing an uncertainty model through error propagation by systematically analyzing sensor models (covering various common sensor combinations) and integrating these models into 3D object pose estimation algorithms. This highlights the necessity of hierarchical abstraction levels. The system is validated through simulations and real-world experiments, providing a quantitative evaluation of object pose estimation; and 3) An IoT-driven intelligent vertical farming architecture that integrating mobile robotic perception networks and environmental regulation devices, enabling dynamic acquisition and closed-loop control of plant growth parameters. Open-source code is available at https://github.com/allenthreee/scene_graph, video link: https://youtu.be/dhc8RLmX7hc.
Why it matches plant phenotyping methods植物・果実の成長監視と計数を目的に、3Dシーングラフ、SLAM、センサー融合、誤差伝播モデルを開発・検証しており、表現型取得手法が研究の中心である。
titleHierarchical 3-D Scene-Graph-Based Semantic-Metric SLAM for Plant Inspection and Fruit Counting in Intelligent Hydroponics System
This paper presented Bit-STED, a novel and simplified transformer encoder architecture for efficient agave plant detection and accurate counting using unmanned aerial vehicle (UAV) imagery. Addressing the critical need for accessible and cost-efficient solutions in agricultural monitoring, this approach automates a process that is typically time-consuming, labor-intensive, and prone to human error in manual practices. The Bit-STED model features a lightweight transformer design that incorporates innovative techniques for efficient feature extraction, model compression through quantization, and shape-aware object localization using circular bounding boxes for the roughly circular shape of the agave rosettes. To complement the detection model, a novel counting algorithm was developed to manage plants spanning multiple image tiles accurately. The experimental results demonstrated that the Bit-STED model outperformed the baseline models in terms of detection and agave plant count performance. Specifically, the Bit-STED nano model achieved F1 scores of 96.66% on a map with younger plants and 96.43% on a map with larger, highly overlapping plants. These scores surpassed state-of-the-art baselines, such as YOLOv8 Nano (F1 scores of 96.42% and 96.38%, respectively) and DETR (F1 scores of 93.03% and 85.61%, respectively). Furthermore, the Bit-STED nano model was significantly smaller, being less than one-eighth the size of the YOLOv8 nano model (1.4 MB compared to 12.0 MB), had fewer trainable parameters (0.35M compared to 3.01M), and was faster in average inference times (14.62 ms compared to 18.28 ms).
Why it matches plant phenotyping methodsUAV画像からアガベ個体数を推定する軽量検出・カウント手法の開発と性能比較が中心であり、植物個体数という観測可能な形質を抽出している。
abstracta novel and simplified transformer encoder architecture for efficient agave plant detection and accurate counting using unmanned aerial vehicle (UAV) imagery
The spatial distribution and abundance of plant species are of critical importance for the identification of plant communities, the assessment of biodiversity, and the fulfilment of environmental policy requirements, such as those outlined in the Habitat Directive 92/43/EEC. Recent advancement in high-resolution drone imaging provides new opportunities for the identification of plant species, offering significant advantages over traditional expert-based methods, which, while accurate, are often time-consuming. This study utilizes deep learning models, namely Vision Transformer (VIT-B16 and VIT-H14) and Convolutional Neural Networks (VGG19 and Resnet101), to quantify the abundance of tree species from RGB images captured by drones in multiple areas of central Italy. The images were segmented into 256 × 256-pixel tiles to enable efficient computational analysis. Following a rigorous training and evaluation process, the ViT-H14 model was identified as the most effective approach, demonstrating an accuracy of over 0.93. The model’s efficacy was substantiated through a comparison with manual analyses conducted by botanical experts, utilising the Mantel Test. This analysis revealed a strong correlation (r =0.87), substantiating the model’s capacity to interpret forest images with a high degree of accuracy. These findings demonstrate the potential of deep learning models, particularly ViT-B16 and VIT-H14, for efficient and scalable ecological monitoring and biodiversity assessments.
Why it matches plant phenotyping methodsUAV画像から樹種の abundance を推定する深層学習手法を開発・比較し、専門家による手動解析で検証しており、植物群落の構成・量の推定が中心的な方法論的貢献です。
abstractThis study utilizes deep learning models, namely Vision Transformer (VIT-B16 and VIT-H14) and Convolutional Neural Networks (VGG19 and Resnet101), to quantify the abundance of tree species from RGB images captured by drones in multiple areas of central Italy.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 6 Sept 2026
Growth chamberLeafClassificationCountingObject detectionGrowth / development / phenologyLeaf traits
Climate change, shrinking arable land, urbanization, and labor shortages increasingly threaten stable crop production, attracting growing attention toward AI-based indoor farming technologies. Accurate growth stage classification is essential for nutrient management, harvest scheduling, and quality improvement; however, conventional studies rely on time-based criteria, which do not adequately capture physiological changes and lack reproducibility. This study proposes a phenotyping-based and physiologically grounded growth stage classification pipeline for basil. Among various morphological traits, the number of leaf pairs emerging from the shoot apex was identified as a robust indicator, as it can be consistently observed regardless of environmental variations or leaf overlap. This trait enables non-destructive, real-time monitoring using only low-cost fixed cameras. The research employed top-view images captured under various artificial lighting conditions across seven growth chambers. YOLO automatically detected multiple plants, followed by K-means clustering to align positions and generate an individual dataset of crop images–leaf pairs. A regression model was then trained to predict leaf pair counts, which were subsequently converted into growth stages. Experimental results demonstrated that the YOLO model achieved high detection accuracy with mAP@0.5 = 0.995, while the A convolutional neural network regression model reached MAE of 0.13 and R² of 0.96 for leaf pair prediction. Final growth stage classification accuracy exceeded 98%, maintaining consistent performance in cross-validation. In conclusion, the proposed pipeline enables automated and precise growth monitoring in multi-plant environments such as plant factories. By relying on low-cost equipment, the pipeline provides a technological foundation for precision environmental control, labor reduction, and sustainable smart agriculture.
Why it matches plant phenotyping methodsバジルの葉対数という植物形質を低コストカメラ画像から自動推定し、生育段階へ分類する画像解析パイプラインを開発・評価しており、フェノタイピング手法が中心である。
abstractThis study proposes a phenotyping-based and physiologically grounded growth stage classification pipeline for basil.
Rapeseed / canolaNeRF / 3D Gaussian SplattingLiDAR / point cloudFruitCountingSegmentationFruit / seed / panicle traits
Accurate analysis of plant phenotypic traits is crucial for crop breeding and precision agriculture. This study proposes a lightweight semantic segmentation model named KAN-GLNet (Kolmogorov-Arnold Network with Global-Local Feature Modulation), based on an enhanced PointNet++ architecture and integrated with an optimized Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm, to achieve high-precision segmentation and automatic counting of canola siliques. A multi-view point cloud acquisition platform was built, and high-fidelity canola point clouds were reconstructed using Neural Radiance Fields (NeRF) technology. The proposed model includes three key modules: Reverse Bottleneck Kolmogorov-Arnold Network Convolution, a Global-Local Feature Modulation (GLFN) block, and a contrastive learning-based normalization module called ContraNorm. KAN-GLNet contains only 5.72M parameters and achieves 94.50% mIoU, 96.72% mAcc, and 97.77% OAcc in semantic segmentation tasks, outperforming all baseline models. In addition, the DBSCAN workflow was optimized, achieving a counting accuracy of 97.45% in the instance segmentation task. This method achieves an excellent balance between segmentation accuracy and model complexity, providing an efficient solution for high-throughput plant phenotyping. The code and dataset have been made publicly available at: https://anonymous.4open.science/r/KAN-GLNet-6432/.
Why it matches plant phenotyping methodsカノーラ莢のセグメンテーションと自動計数という植物形質抽出手法を、3D点群取得基盤・NeRF再構成・新規モデル・DBSCANワークフローとして開発・評価しており、植物フェノタイピング手法が中心である。
abstractThis study proposes a lightweight semantic segmentation model named KAN-GLNet (Kolmogorov-Arnold Network with Global-Local Feature Modulation), based on an enhanced PointNet++ architecture and integrated with an optimized Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm, to achieve high-precision segmentation and automatic counting of canola siliques.
Reproduction assets foundThe authors explicitly state that their curated code and dataset (canola silique point cloud phenotyping data and KAN-GLNet analysis code) are publicly available at an anonymous.4open.science repository, which is an allowed URL.Code · publicDBSCAN workflow was optimized, achieving a counting accuracy of 97.45% in the instance segmentation task. This method achieves an excellent balance between segmentation accuracy and model complexity, providing an efficient solution for high-throughput plant phenotyping. The code and dataset have been made publicly available at: https://anonymous.4open.science/r/KAN-GLNet-6432/ .
http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China
32301762
Liu Jie
This project is supported by National Natural Science Foundation of China, grant number 32301762.
pmc-status-qastatus 0 pmc-status-live yes pmc-status-embargo no pmc-status-released yes pmc-prop-open-access yes pmc-pOpen asset ↗anonymous.4open.science/r/KAN-GLNet-6432lines:1-65Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Tea flowers play a crucial role in taxonomic research and hybrid breeding of tea plants. As traditional methods of observing tea flower traits are labor-intensive and inaccurate, TflosYOLO and Tea Flowering Stage Classification (TFSC) models were proposed for tea flowering quantification, which enable the estimation of flower count and flowering period. In this study, a highly representative and diverse dataset was constructed by collecting flower images from 29 tea accessions in 2 years. Based on this dataset, the TflosYOLO model was built on the YOLOv5 architecture and enhanced with the Squeeze-and-Excitation (SE) network, Adaptive Rectangular Convolution, and Attention Free Transformer, which is the first model to offer a viable solution for detecting and counting tea flowers. The TflosYOLO model achieved a mean Average Precision at 50% IoU (mAP50) of 0.844, outperforming YOLOv5, YOLOv7, and YOLOv8. Furthermore, the TflosYOLO model was tested on 31 datasets encompassing 26 tea accessions and five flowering stages, demonstrating high generalization and robustness. The correlation coefficient (R 2 ) between the predicted and actual flower counts was 0.964. Additionally, the TFSC model-a seven-layer neural network-was designed for the automatic classification of the flowering period. The TFSC model was evaluated for 2 years and achieved an accuracy of 0.738 and 0.899. Using the TflosYOLO+TFSC model, the tea flowering dynamics were monitored, and the changes in flowering stages were tracked across various tea accessions. The framework provides crucial support for tea plant breeding programs and the phenotypic analysis of germplasm resources.
Why it matches plant phenotyping methods茶花画像から花数と開花期を推定するモデルを開発・検証しており、植物表現型の取得・抽出手法が研究の中心である。
abstractTflosYOLO and Tea Flowering Stage Classification (TFSC) models were proposed for tea flowering quantification, which enable the estimation of flower count and flowering period.
Reproduction assets foundThe paper's data availability statement explicitly deposits the tea flower datasets and models in a public GitHub repository (sufie-mi/tea-flower-model), which directly supports this paper's tea flower phenotyping measurements and models. The labelImg repository is a generic third-party annotation tool, not a paper-ownDataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://github.com/sufie-mi/tea-flower-model .Open asset ↗https://github.com/sufie-mi/tea-flower-model · tea-flower-modellines:764-781Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Published12 Nov 20252025 IEEE International Conference on Data Mining Workshops (ICDMW)Cited by 0 · OpenAlex ↗
Leaf counting has numerous applications in plant phenotyping, such as plant growth analysis, yield prediction, and disease detection. However, manual counting is labor-intensive and time-consuming, posing a significant limitation. To address this issue, deep learning-based object detection models are implemented for automated leaf counting. Existing works that explore automated leaf counting are limited due to the inherent structure of plants, e.g., plants are small and occur in overlapping clusters, causing the models to perform sub-optimally. Furthermore, the numerous amount of object detection models introduces a problem of choosing the proper model. Currently, there is limited understanding in determining which object detection architectures perform well for automated leaf counting, particularly when dealing with complex plant anatomy. To address the gap in the literature, three foundational object detection models with different characteristics are compared: YOLOv8, YOLOv12, and Faster R-CNN. The three models chosen display key distinctions such as one-stage vs. two-stage detections and convolution-based vs. attention-based. To provide comprehensive results, multiple optimization techniques were applied to each model. Our experimental results showed that YOLOv8 had the best performance on all performance metrics (mAP50, mAP5095, IOU, MAE, and MAPE). Specifically, YOLOv8 achieved an IOU of 0.6412, while YOLOv12 and Faster R-CNN only achieved an IOU of 0.567 and 0.606, respectively.
Why it matches plant phenotyping methods植物の葉数という形態形質を対象に、重なり葉条件下での自動計数手法として複数の物体検出モデルを比較・最適化し、性能評価しているため、方法比較・検証が中心です。
abstractdeep learning-based object detection models are implemented for automated leaf counting
Why it matches plant phenotyping methods深層学習と画像処理により、感染植物細胞内のハウストリアを自動検出・セグメンテーションする公開パイプラインを開発し、手動計数および種間移植性を検証しているため、植物病害表現型の取得手法が中心である。
abstractWe report an openly available pipeline that automates the detection of β-glucuronidase (GUS)-stained epidermal cells and the intracellular haustoria formed by powdery mildew on barley and wheat leaves.
In modern agricultural production, accurate monitoring of maize growth and leaf counting is crucial for precision management and crop breeding optimization. Current UAV-based methods for detecting maize seedlings and leaves often face challenges in achieving high accuracy due to issues such as low spatial-resolution, complex field environments, variations in plant scale and orientation. To address these challenges, this study develops an integrated detection and visualization software, DP-MaizeTrack, which incorporates the DP-YOLOv8 model based on YOLOv8. The DP-YOLOv8 model integrates three key improvements. The Multi-Scale Feature Enhancement (MSFE) module improves detection accuracy across different scales. The Optimized Spatial Pyramid Pooling-Fast (OSPPF) module enhances feature extraction in diverse field conditions. Experimental results in single-plant detection show that the DP-YOLOv8 model outperforms the baseline YOLOv8 with improvements of 3.9% in Precision (95.1%), 4.1% in Recall (91.5%), and 4.0% in mAP50 (94.9%). The software also demonstrates good accuracy in the visualization results for single-plant and leaf detection tasks. Furthermore, DP-MaizeTrack not only automates the detection process but also integrates agricultural analysis tools, including region segmentation and data statistics, to support precision agricultural management and leaf-age analysis. The source code and models are available at https://github.com/clhclhc/project.
Why it matches plant phenotyping methodsUAV画像からトウモロコシ個体数と葉数を抽出するソフトウェアを開発しており、植物形質取得が研究の中心です。
abstractthis study develops an integrated detection and visualization software, DP-MaizeTrack
Reproduction assets foundThe paper explicitly states that the authors' source code and trained models for DP-MaizeTrack/DP-YOLOv8 are publicly available on GitHub. No public dataset deposit is stated; the UAV image dataset is described but not declared publicly available.Code · publicThe source code and models are available at https://github.com/clhclhc/project .Open asset ↗https://github.com/clhclhc/projectlines:224-300Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 6 Sept 2026
Abstract Mungbean ( Vigna radiata (L.) R. Wilczek) is a vital source of digestible proteins and is well-suited for the plant-based protein industry. In this study, we analyzed pod morphological traits in the Iowa Mungbean Diversity (IMD) panel with 372 genotypes (2022-23) with AI-assisted image phenotyping using 2,418 pod images. Pod morphological traits were extracted using deep learning image analysis, achieving excellent agreement with manual measurements (r>0.96 for pod length and seed per pod). Four complementary GWAS models identified 45 significant SNPs associated with pod curvature, length, width, and seed per pod traits. Notably, a significant SNP (5_35265704) on chromosome 1 was linked to pod dimensional traits, length, width, and curvature. A candidate gene, Vradi01g00001116 , was located within the linkage disequilibrium (LD) region of this SNP, is part of the GH3 gene family, and has an Arabidopsis ortholog ( AT4G27260 ) known for influencing organ elongation, pod, and seed development. Another SNP, 5_210437 on chromosome 2, has been found to be significantly associated with both pod length and seed per pod. A candidate gene, Vradi02g00003971 , located in the LD region of this SNP, belongs to the potassium transporter family and shares homology with the HAK5 gene family ( AT4G13420 ) in Arabidopsis , which influences pod and seed growth. Image-based measurements achieved genomic prediction accuracies ranging from 0.61 to 0.85 across various traits, exhibiting an improvement of 12-22% over manual methods. These results demonstrate the potential of AI-assisted phenomics integrated with genomic tools to accelerate selection for improved pod architecture in mungbean breeding programs across the Midwestern United States and globally.
Why it matches plant phenotyping methodsAI画像解析による莢形態形質の抽出と手測定との技術検証が研究の中心であり、GWAS応用も行っているため含める。
abstractwith AI-assisted image phenotyping using 2,418 pod images
Abstract. The seedling emergence rate is a crucial indicator for evaluating the growth status of crops in agricultural production and can provide valuable recommendations for subsequent crop planting and field management strategies. Currently, the determination of the emergence rate relies on manual seedling counting, which is not only labour-intensive and time-consuming, but also prone to human errors. Therefore, we utilize drone-captured images of peanut seedlings and employs deep learning networks to estimate seedling numbers. Specifically, we incorporate the BIFPN (Bidirectional Feature Pyramid Network) feature fusion module into the original Centernet model, which would combine multi-scale feature information. This modification not only enhances the accuracy of identification but also improves the localization of seedlings. To address the issue of false positives caused by complex field backgrounds in seedling recognition, we integrate the Contrastive Loss module to increase the discrepancy between positive and negative samples. The results demonstrate that the proposed method significantly enhances both precision and recall rates for peanut seedling recognition under three different scenes, compared to the original model. Furthermore, the proposed method is also applied in real peanut breading field, fulfilling the practical requirements for emergence rate calculation.
Why it matches plant phenotyping methodsUAV画像と深層学習により、ピーナッツ幼苗の識別・計数から出芽率を推定する手法を開発・評価しており、植物形質取得が中心である。
abstractwe utilize drone-captured images of peanut seedlings and employs deep learning networks to estimate seedling numbers.
As global agriculture shifts to intelligence and precision, crop attribute detection has become foundational for intelligent systems (harvesters, UAVs, sorters). It enables real-time monitoring of key indicators (maturity, moisture, disease) to optimize operations—reducing crop losses by 10–15% via precise cutting height adjustment—and boosts resource-use efficiency. This review targets harvesting-stage and in-field monitoring for grains, fruits, and vegetables, highlighting practical technologies: near-infrared/Raman spectroscopy (non-destructive internal attribute detection), 3D vision/LiDAR (high-precision plant height/density/fruit location measurement), and deep learning (YOLO for counting, U-Net for disease segmentation). It addresses universal field challenges (lighting variation, target occlusion, real-time demands) and actionable fixes (illumination compensation, sensor fusion, lightweight AI) to enhance stability across scenarios. Future trends prioritize real-world deployment: multi-sensor fusion (e.g., RGB + thermal imaging) for comprehensive perception, edge computing (inference delay
Why it matches plant phenotyping methods作物属性の検出・監視技術を主題とするレビューで、分光、3Dビジョン、LiDAR、深層学習による植物形質・病害状態の取得方法を中心に整理している。
titleA Review of Crop Attribute Monitoring Technologies for General Agricultural Scenarios
Leaf detection and counting are essential in plant phenotyping, but traditional manual methods are slow and error prone. To improve the efficiency and accuracy of leaf counting, this study introduces a lightweight, high-precision model for leaf detection and counting based on the optimized YOLOv8 computer vision model, called MobileViT-Asymptotic Feature Pyramid Network-YOLOv8 (MAF-YOLOv8). This model integrates the MobileViT architecture and an Adaptive Feature Pyramid Network (AFPN) structure, achieving a lightweight model with enhanced feature representation capabilities, thereby improving leaf counting accuracy. In this work, we constructed a dataset consisting of 711 RGB images with a 640 × 640 resolution and expanded it to 2136 images using data augmentation methods to enhance model robustness. The MAF-YOLOv8 model was able to achieve a mean average precision (mAP) of 91.7 %, a recall of 95.0 %, and a precision of 86.5 % in leaf counting tasks. Compared to YOLOv8, mAP improved by 2.6 %, while the number of parameters was reduced by 34.1 %. Ablation experiments evaluating the contributions of each model component further confirmed that MobileViT and AFPN critically improve model performance; together, they led to a 2.3 % improvement in precision and a 3 % increase in recall. This study also validates the performance of MAF-YOLOv8 on resource-constrained mobile devices, achieving an inference time of only 5.110 s. The findings indicate that the model has superior accuracy, inference speed, and resource efficiency, which renders it suitable for extensive applications within agriculture and environmental monitoring. This study provides an efficient technological approach for plant phenotyping and precision agriculture.
Why it matches plant phenotyping methods葉の検出・計数という植物表現型を抽出する画像解析モデルを開発し、データセット、アブレーション、精度・速度評価、モバイル実装まで行っており、方法が研究の中心である。
abstractLeaf detection and counting are essential in plant phenotyping, but traditional manual methods are slow and error prone.
High precision 3D data is becoming crucial for accurate feature extraction. Acquiring 3D data from plants with different growing patterns and thier growth under different environmental conditions is still a challenging task. The utilization of deep learning techniques can overcome some of these challenges, but these techniques often demand good quality training data for 3D point cloud analysis. One of the main challenges in plant phenotyping is the general lack of annotated 3D datasets available to the research community. Constructing such datasets is particularly difficult due to the complexity of capturing high-quality data that accurately represent the intricate structures and diverse morphologies of plants. The development of robust data sets is critical to advance plant phenotyping, allowing precise quantification of plant traits, and addressing challenges in modern agriculture. However, the lack of high-quality, annotated datasets for complex plant structures, such as wheat, hinders the development of effective methodologies. To address this, we introduce Wheat3D PartNet, a comprehensive repository of 1303 3D point cloud models of wheat (Triticum L.), comprising three cultivars: Paragon, Gladius, and Apogee. The 3D point clouds are reconstructed from RGB images of real plants that were acquired from multiple viewpoints and represent different plant structures at different growth rates. Wheat3D PartNet samples are manually labeled into two parts i.e., ears (wheat spikes) and non-ears (leaves and stems) and that captured in drought and watered conditions. Wheat3D PartNet is designed to support segmentation-based trait quantification tasks such as spike counting, spike length estimation, and stress detection-facilitating more precise yield prediction and enabling early agronomic intervention. Extensive experiments using several state-of-the-art 3D deep learning models validate the dataset's utility and challenge level. The methodology behind Wheat3D PartNet is extensible to other crops, including rice and potato, and is expected to significantly boost the research, understanding, and measurements of plants of interest.
Why it matches plant phenotyping methods植物の3D点群を用いた部位セグメンテーション用データセットの構築・検証が中心で、植物形質の定量化を直接支援するため。
abstractTo address this, we introduce Wheat3D PartNet, a comprehensive repository of 1303 3D point cloud models of wheat (Triticum L.)
Les avancées récentes des drones et du traitement des données permettent aujourd’hui de produire des images hautes résolution et des modèles 3D utiles pour évaluer les attributs des arbres. Cette étude a été menée à Widou thiengoly dans la localité du Ferlo, Nord du Sénégal, avec comme objectif général d’appliquer une approche photogrammétrique pour la mesure de la densité des tiges (tiges/ha) avec des images drones. Une méthode de l’approche arbre basée sur un modèle numérique de hauteur d'une zone d'étude de 10 hectares a été mise en œuvre, ce modèle a été construit à partir d'images obtenues par des drones. Au total, 92 arbres de référence ont été comptés dans le cadre de cette étude et l'algorithme a détecté 75 arbres, ce qui donne une précision supérieure à 90 % (score F de 0,93). Dans l'ensemble, l'algorithme a manqué 10 arbres (erreurs d'omission) et a faussement détecté 3 arbres (erreurs de commission), ce qui donne un compte total de 88 arbres. Cette étude suggère que l'algorithme de filtrage des maxima locaux combiné avec des tailles de fenêtre optimale, appliqués sur un Modèle Numérique de Hauteur construit par photogrammétrie est capable d’effectuer des comptages d'arbres avec une précision acceptable (F > 0,90) dans la zone sahélienne. Recent advances in drone technology and data processing now make it possible to generate high-resolution images and 3D models that are useful for assessing tree attributes. This study was conducted in Widou Thiengoly, in the Ferlo area of northern Senegal, with the overall objective of applying a photogrammetric approach to measure stem density (stems/ha) using drone imagery. A tree-based method was implemented on a Digital Height Model covering a 10-hectare study area, constructed from drone images. In total, 92 reference trees were counted during the study, and the algorithm detected 75 trees, resulting in an accuracy above 90% (F-score of 0.93). Overall, the algorithm missed 10 trees (omission errors) and falsely detected 3 trees (commission errors), giving a total count of 88 trees. This study suggests that the local maxima filtering algorithm, combined with optimal window sizes and applied to a photogrammetrically derived Digital Height Model, can perform tree counts with acceptable accuracy (F > 0.90) in the Sahelian zone.
Why it matches plant phenotyping methodsドローン画像とフォトグラメトリによる樹木密度の推定手法を開発・評価し、Fスコアで精度検証しているため、植物形質取得が研究の中心である。
abstractavec comme objectif général d’appliquer une approche photogrammétrique pour la mesure de la densité des tiges (tiges/ha) avec des images drones.
Oil palm is a globally significant economic crop, especially in Malaysia and Indonesia, where it contributes substantially to national economies. Accurate monitoring of plantations, particularly tree count and health, is essential for effective management and yield estimation. However, traditional field-based methods are costly and labor-intensive, and while UAVs and very high-resolution satellite imagery offer precision, they are often limited by high cost, limited coverage, and technical constraints such as altitude variation and image mosaicking. This study proposes a scalable and cost-effective pipeline that utilizes UAV imagery implementing two tree-counting approaches Template Matching, serving as a classical baseline, and YOLOv8, a deep learning object detection model chosen for its high accuracy and inference speed. The oil palm tree density was classified into High, Medium, and Low category for easy result observation. Preliminary results demonstrate promising True Positive Rate (TPR), False Negative Rate (FNR), and Estimation Error (EE) values, with YOLOv8 achieving its best performance in medium-density regions (TPR: 109.54%, FNR: 9.50%, EE: 12.91 %) and Template Matching showing the weakest performance in low-density areas (TPR: 305.45%, FNR: 205.45 %, EE: 205.45 %). These results indicate that the proposed approach offers a practical, real-time, and resourceefficient solution for large-scale oil palm monitoring, although improvements are still required in low-density plantations.
Why it matches plant phenotyping methodsUAV画像から油ヤシの樹木数・密度を推定する画像解析パイプラインを提案し、Template MatchingとYOLOv8を比較評価しており、植物個体数という観測可能な形態・群落特性の取得手法が中心である。
abstractThis study proposes a scalable and cost-effective pipeline that utilizes UAV imagery implementing two tree-counting approaches Template Matching, serving as a classical baseline, and YOLOv8, a deep learning object detection model chosen for its high accuracy and inference speed.
Unmanned aerial vehicles (UAVs) are transforming agriculture through enhanced data acquisition, improved monitoring efficiency, and support for data-driven decision-making. Complementing this, AI-driven platforms provide intuitive and reliable tools for advanced UAV analytics. However, their integration remains underexplored, particularly in specialty crops. Therefore, in this study, we evaluated the performance of an AI-driven web platform (Solvi) for automated plant counting and biometric trait estimation in two contrasting systems: pecan, a perennial nut crop, and onion, an annual vegetable. Ground-truth measurements included pecan tree number, tree height, and canopy area, as well as onion bulb number and diameter, the latter used for market class classification. Counting performance was assessed using precision, recall, and F1 score, while trait estimation was evaluated with linear regression analysis. UAV-based counts showed strong agreement with ground-truth data, achieving precision, recall, and F1 scores above 97% for both crops. For pecans, UAV-derived estimates of tree height (R2 = 0.98, error = 11.48%) and canopy area (R2 = 0.99, error = 23.16%) demonstrated high accuracy, while errors were larger in young trees compared with mature trees. For onions, UAV-derived bulb diameters achieved an R2 of 0.78 with a 6.29% error, and market class classification (medium, jumbo, colossal) was predicted with
Why it matches plant phenotyping methodsUAV画像とAIプラットフォームによる植物個体数・樹高・樹冠面積・球根径の自動推定を評価しており、表現型取得手法の性能評価が研究の中心です。
abstractwe evaluated the performance of an AI-driven web platform (Solvi) for automated plant counting and biometric trait estimation
Wheat is an important food crop, wheat seedling count is very important to estimate the emergence rate and yield prediction. Timely and accurate detection of wheat seedling count is of great significance for field management and variety breeding. In actual production, the method of artificial field investigation and statistics of wheat seedlings is time-consuming and laborious. Aiming at the problems of small targets, dense distribution and easy occlusion of wheat seedling in the field, a wheat seedling number detection model (DM_IOC_fpn) combining local and global features was proposed in this study. Firstly, the wheat seedling image is preprocessed, and the wheat seedling dataset is built by using the point annotation method. Secondly, the density enhanced encoder module is introduced to improve the network structure and extract local and global contextual feature information of wheat seedling. Finally, the total loss function is constructed by introducing counting loss, classification loss, and regression loss to optimize the model, so as to enable accurate judgment of wheat seedling position and category information. Experiment on self-built dataset have shown that the root mean square error (RMSE) and mean absolute error (MAE) of DM_IOC_fpn were 2.91 and 2.23, respectively, which were 1.78 and 1.04 lower than the original IOCFormer. Compared with the current mainstream object detection models, DM_IOC_fpn has better counting performance. DM_IOC_fpn can accurately detect the number of small target wheat seedling, and better solve the problem of occlusion and overlapping of wheat seedling, so as to achieve the accurate detection of wheat seedling, which provides important theoretical and technical support for automatic counting of wheat seedlings and yield prediction in complex field environment.
Why it matches plant phenotyping methods小麦幼苗数という植物形質をUAV画像から自動抽出・計数するモデルを開発し、自作データセット上で性能評価しており、表現型取得手法が中心である。
abstractExperiment on self-built dataset have shown that the root mean square error (RMSE) and mean absolute error (MAE) of DM_IOC_fpn were 2.91 and 2.23, respectively
Corn is a globally important economic crop. Certain trait parameters of corn ears kernels per ear are essential indicators for corn breeding. However, acquiring these parameters faces two challenges: i) manual measurement is labor-intensive and error-prone, and ii) vision-based corn phenotyping machines require fixed image capturing environment and are cost-prohibitive. To address these limitations, we introduce CornPheno, a user-friendly, low-end, smartphone-based approach capable of executing corn ear phenotyping in the wild. CornPheno highlights three corn ear parameters: kernels per ear, rows per ear, and kernels per row. Technically, inspired by crowd localization in computer vision, we first extract kernels per ear based on a Corn data-trained Point quEry Transformer (CornPET). CornPET generates interpretable per-kernel point predictions and supports subsequent row detection. To detect rows, we introduce a novel point-based corn row detection approach, termed unicorn, featured by sqUeezed clusteriNg and bI-direCtional pOint seaRchiNg, to phenotype rows per ear and kernels per row. With adaptive geometric modeling, our approach is robust to partial rows, curved rows, and missing kernels. To promote the use of CornPheno, we have integrated it into OpenPheno, a WeChat-based mini-program, and made it open-access for corn breeders. We hope our approach can provide the community with a user-friendly and cost-effective way to facilitate corn breeding.
Why it matches plant phenotyping methodsトウモロコシ穂の粒数・列数などの形質を、スマートフォン画像から抽出する手法とソフトウェアを開発しており、フェノタイピング手法が研究の中心である。
abstractwe introduce CornPheno, a user-friendly, low-end, smartphone-based approach capable of executing corn ear phenotyping in the wild.
Background To address the challenge of real-time plant monitoring in greenhouse environments, this industry-driven research focuses on developing an autonomous quadrotor UAV system specifically designed for monitoring strawberry plants. Traditional methods for greenhouse monitoring are labor-intensive and lack scalability, particularly in precision agriculture applications. Method This research introduces a mature strawberry detection model specifically designed for greenhouse environments. The proposed YOLOv9-GLEAN approach enables the identification of small mature strawberries through an onboard camera mounted on the quadrotor. Additionally, a hybrid trajectory tracking controller for the quadrotor is developed and tested in both simulated and real-world conditions. The UAV navigates through the greenhouse using predetermined waypoints, operating as a semi-autonomous system for navigation while maintaining full autonomy in mature strawberry detection tasks. The system incorporates an integrated onboard vision platform that utilizes an innovative YOLOv9-GLEAN-based algorithm to perform real-time and offline detection and counting of mature strawberries. Results The YOLOv9-GLEAN model achieves high detection accuracy, as confirmed by evaluation metrics such as precision, recall, and F1-score. The proposed hybrid (PID+LQR) controller demonstrates superior tracking performance compared to other conventional controllers. The integrated control and perception system proves effective in both simulated and real-world greenhouse environments. Discussion The research validates the efficacy of deep learning models, with YOLOv9-GLEAN showing exceptional performance in enabling rapid, precise, and automated detection of ripe strawberries through quadrotor deployment in greenhouse environments. Such agricultural monitoring technologies represent a substantial advancement beyond conventional manual inspection approaches, empowering farmers and greenhouse operators to execute well-informed, time-sensitive management decisions that minimize crop losses and optimize production yields. This investigation underscores the revolutionary impact that deep learning technologies can have within greenhouse agriculture.
Why it matches plant phenotyping methodsイチゴ果実の成熟状態を画像から検出・計数する深層学習モデルとUAV搭載視覚プラットフォームの開発・評価が研究の中心であり、植物器官の状態を直接推定している。
abstractThis research introduces a mature strawberry detection model specifically designed for greenhouse environments.
Accurately detecting and counting potatoes during early harvest is essential for estimating yield, automating sorting, and supporting data-driven agricultural decisions. However, field environments often present practical challenges-such as soil occlusion, overlapping tubers, and inconsistent lighting-that hinder robust visual recognition. In response, we introduce SCG-YOLOv8n, a compact and field-adapted detection framework built upon the YOLOv8n architecture and specifically tailored for small-object detection in real-world farming conditions. The model incorporates three practical enhancements: a C-SPD module that preserves spatial detail to improve recognition of partially buried tubers; an S-CARAFE operator that reconstructs fine-scale features during upsampling; and GhostShuffleConv layers that reduce computational overhead without sacrificing accuracy. Through extensive field-based experiments, SCG-YOLOv8n consistently outperforms YOLOv5n and its base version across all key metrics. Float16 quantization compresses the model to 3.2 MB, enabling real-time inference on Android devices. We also developed PotatoDetector, a mobile application that demonstrates stable performance in field trials, achieving an RMSE of 1.38 and [Formula: see text] of 0.96 in counting tasks. These results suggest that SCG-YOLOv8n offers a practical and scalable tool for precision agriculture, with potential applicability to other root and tuber crop monitoring scenarios.
Why it matches plant phenotyping methodsジャガイモ塊茎の画像検出・計数を行うモデルとモバイル実装を開発し、圃場で性能検証している。塊茎数という植物器官形質の取得が中心であり、単なる収量測定ではない。
abstractwe introduce SCG-YOLOv8n, a compact and field-adapted detection framework built upon the YOLOv8n architecture and specifically tailored for small-object detection in real-world farming conditions.
Reproduction assets foundThe paper's custom potato image dataset is not publicly available (available only from the corresponding author on request), but the authors provide a public GitHub repository for the SCG-YOLOv8n analysis code with an explicit availability statement and URL.Code · publicCode availability
Code can be found at https://github.com/AiXia520/SCG-YOLOv8n.git.Open asset ↗https://github.com/AiXia520/SCG-YOLOv8n.githtml-lines:359-392Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Accurate detection and counting of wheat spikes are crucial for yield estimation and variety selection in precision agriculture. However, challenges such as complex field environments, morphological variations, and small target sizes hinder the performance of existing models in real-world applications. This study proposes FEWheat-YOLO, a lightweight and efficient detection framework optimized for deployment on agricultural edge devices. The architecture integrates four key modules: (1) FEMANet, a mixed aggregation feature enhancement network with Efficient Multi-scale Attention (EMA) for improved small-target representation; (2) BiAFA-FPN, a bidirectional asymmetric feature pyramid network for efficient multi-scale feature fusion; (3) ADown, an adaptive downsampling module that preserves structural details during resolution reduction; and (4) GSCDHead, a grouped shared convolution detection head for reduced parameters and computational cost. Evaluated on a hybrid dataset combining GWHD2021 and a self-collected field dataset, FEWheat-YOLO achieved a COCO-style AP of 51.11%, AP@50 of 89.8%, and AP scores of 18.1%, 50.5%, and 61.2% for small, medium, and large targets, respectively, with an average recall (AR) of 58.1%. In wheat spike counting tasks, the model achieved an R 2 of 0.941, MAE of 3.46, and RMSE of 6.25, demonstrating high counting accuracy and robustness. The proposed model requires only 0.67 M parameters, 5.3 GFLOPs, and 1.6 MB of storage, while achieving an inference speed of 54 FPS. Compared to YOLOv11n, FEWheat-YOLO improved AP@50, AP_s, AP_m, AP_l, and AR by 0.53%, 0.7%, 0.7%, 0.4%, and 0.3%, respectively, while reducing parameters by 74%, computation by 15.9%, and model size by 69.2%. These results indicate that FEWheat-YOLO provides an effective balance between detection accuracy, counting performance, and model efficiency, offering strong potential for real-time agricultural applications on resource-limited platforms.
Why it matches plant phenotyping methodsコムギ穂の検出・計数という植物器官形質の画像ベース推定手法を開発し、データセット上で精度・計算効率・堅牢性を評価しているため、植物フェノタイピング手法が中心である。
abstractThis study proposes FEWheat-YOLO, a lightweight and efficient detection framework optimized for deployment on agricultural edge devices.
Reproduction assets foundThe paper uses the public GWHD2021 wheat spike detection dataset (available on Kaggle) as part of its hybrid dataset, with an explicit availability statement and URL. The self-collected Xinjiang field dataset is private and available only on request. No author analysis code, trained model checkpoints, or other paper-特定Dataset · publiciting, W.W., S.L. and Y.L.; supervision, J.C.; project administration, X.Z.; funding acquisition, X.Z. All authors have read and agreed to the published version of the manuscript.
Data Availability Statement
The public part of the dataset used in this study is available from the Global Wheat Head Detection (GWHD2021) dataset at https://www.kaggle.com/competitions/global-wheat-detection , accessed on 30 September 2025. The remaining part of the dataset is private and cannot be shared due to institutional or privacy restrictions. Requests for access to the private dataset may be directed to the corresponding author.
Conflicts of Interest
The authors declare no conflicts of interest. The funderOpen asset ↗GWHD2021lines:854-929Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Oct 2025Computers and Electronics in Agriculture.
Accurate counting of rice grains plays a critical role in rice breeding and thousand-grain weight measurement. However, the accuracy of existing algorithms is insufficient under conditions of high-density and densely bonded rice grain distribution. To quickly and accurately detect high-density and densely bonded rice grains with as many grains as possible, this study developed a lightweight model based on YOLOv5s. First, we built an efficient lightweight model architecture to obtain small-target location and semantic information of rice grains. Second, we use an omni-dimensional dynamic convolution (ODConv) module to replace some of the convolutions of the backbone network to fully extract feature information. We then introduce the mixed local channel attention (MLCA) mechanism to weigh local features through spatial information, allowing the model to locate and identify dense rice grains accurately. Finally, we use the SIoU loss function to improve the convergence speed and accuracy of model training. The model’s detection accuracy was verified via ablation experiments. The results indicated that compared with the original YOLOv5s network, the model size, parameters and floating-point operations per second (FLOPS) of the improved model decreased by 64.16 %, 70.8 % and 28.3 %, respectively, while mAP₀.₅:₀.₉₅ increased by 7.21 %. The mean error rate and mean detection time of the improved model were 0.234 % and 25.9 ms, respectively. Its superior capacity against other detection algorithm models at rapidly detecting densely bonded rice grains. Furthermore, an android application was further developed. After comparative testing on three types of mobile phones, the application was able to effectively and accurately detect and count rice grains, providing an effective solution for rice grain detection and counting.
Why it matches plant phenotyping methods米粒の検出・計数という植物形質取得を目的に、YOLOv5s改良モデルを開発し、アブレーション実験と比較評価、モバイルアプリ実装まで行っており、フェノタイピング手法が研究の中心である。
abstractAccurate counting of rice grains plays a critical role in rice breeding and thousand-grain weight measurement.
Accurate plant counting is essential for tobacco yield estimation and planting density regulation. However, manual quadrat surveys are inefficient and error-prone, and conventional detectors often suffer from feature loss and background confusion under complex field conditions. This study proposes an improved YOLOv8n framework, YOLOv8-AKConv-MLCA, tailored for UAV imagery of plain-field and mountain-field tobacco. First, an Alterable Kernel Convolution (AKConv) module is embedded inside the original C2f blocks to replace all conventional convolutions, enabling adaptive sampling and richer multi-scale representation of small and densely distributed targets. Second, a mixed local channel attention (MLCA) module is inserted between the last C2f-AKConv block and SPPF to fuse local spatial cues with global channel dependencies, suppressing clutter and occlusion effects. Extensive experiments on UAV datasets show that the proposed model achieves counting accuracies of 97.20% (plain-field) and 96.13% (mountain-field), improving over baseline YOLOv8 by 3.98% and 3.25%, respectively. Detection metrics likewise improve: mean average precision (mAP) reaches 0.936 and 0.914 in the two scenarios, surpassing SSD (0.844, 0.827), Faster R-convolutional neural network (0.865, 0.842), and a Transformer-based variant (YOLOv8-Trans, 0.923, 0.907). Relative to YOLOv8, maximum gains of 12.1% in precision, 1.9% in recall, and 7.3% in mAP are observed. Crucially, real-time throughput is preserved, with inference speeds of 219-227 frames per second across datasets. Grad-CAM visualizations further confirm that YOLOv8-AKConv-MLCA concentrates attention on canopy regions and suppresses background interference, offering intuitive evidence of enhanced feature learning. Overall, the proposed framework delivers a strong accuracy-efficiency trade-off and robust generalization under complex terrain, providing an effective solution for automated tobacco plant counting and supporting precision cultivation and smart agricultural management. Code and trained weights are available upon reasonable request for replication and evaluation.
Why it matches plant phenotyping methodsUAV画像からタバコ個体数を推定する検出・計数手法を開発し、複数条件で比較検証しており、植物表現型取得が中心である。
abstractThis study proposes an improved YOLOv8n framework, YOLOv8-AKConv-MLCA, tailored for UAV imagery of plain-field and mountain-field tobacco.
Accurately obtaining the crop quantity and density is not only crucial for the demand-based input of water and fertilizer in the field but also vital for ensuring the yield and quality of crops. Aerial photography by unmanned aerial vehicles (UAVs) can quickly acquire the distribution image information of crops over a large area. However, the accurate recognition of a single type of dense targets is a huge challenge for most recognition algorithms. Taking banana seedlings as an example in this study, we captured the images of banana plantations by UAVs from high altitudes to explore an efficient recognition method for dense targets. We proposed a strategy of "cut-recognition-stitch" and constructed a counting method based on the improved Faster R-CNN algorithm. First, the images containing highly dense targets were cropped into a large number of image tiles according to different sizes (simulating different flight altitudes), and the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm was adopted to improve the image quality. A banana seedling dataset containing 36 000 image tiles was constructed. Then, the Faster R-CNN network with optimized parameters was used to train the banana seedling recognition model. Finally, the recognition results were reversely stitched together, and a boundary deduplication algorithm was designed to correct the final counting results to reduce the repeated recognition caused by image cropping. The results show that the recognition accuracy of the Faster R-CNN with optimized parameters for banana image datasets of different sizes can reach up to 0.99 at most. The deduplication algorithm can reduce the average counting error for the original aerial images from 1.60% to 0.60%, and the average counting accuracy of banana seedlings reaches 99.4%. The proposed method effectively addresses the challenge of recognizing dense small objects in high-resolution aerial images, providing an efficient and reliable technical solution for intelligent crop density monitoring in precision agriculture.
Why it matches plant phenotyping methodsバナナ苗の密度・個体数という植物状態を、UAV画像と改良Faster R-CNN、切り出し・再結合・重複除去により定量化する手法を開発・評価しており、フェノタイピング手法が中心である。
abstractconstructed a counting method based on the improved Faster R-CNN algorithm
Yield forecasting is crucial for growers, enabling efficient resource management and informed decision-making. Such decisions impact storage, product processing, and logistics, leading to increased productivity and cost savings. However, this heavily relies on accurate yield forecasts. This work addresses such a need by presenting the development and testing of a reliable method for yield forecasting. The proposed methodology combines high-resolution object detection with a multi-variate input forecasting model that accurately computes the yield for incoming harvests. The forecasting approach incorporates a physically-constrained model based on a Long Short-Term Memory (LSTM) network. This model dynamically applies weights to the time-series data composed of counts for the phenological stages: flower, green, small white, large white, pink, and red (ripe fruit). These counts are obtained from detections made by a YOLOv10s, achieving an mAP@50 of 0.74 for all classes. As a result, the forecasting model's capacity to interpret input data is enhanced, translating it into a valid ripe count forecast. To validate the proposed approach, the forecasting model was trained and evaluated using (a) untreated count sequences and (b) weighted count sequences. The results indicate that phenologically-weighted input sequences outperform untreated sequences, with the following evaluation metrics: R² = 0.74, Root Mean Square Error (RMSE) = 12.67, Mean Absolute Error (MAE) = 10.95, and Mean Absolute Percentage Error (MAPE) = 39.4, improving 15%, 19.26%, 17.13%, and 11.3%, respectively.
Why it matches plant phenotyping methods画像ベースのYOLO検出でイチゴの生育段階・果実数を抽出し、収量予測へ利用する方法を開発・検証しており、植物表現型の取得と解析が中心的な貢献である。
abstractThis work addresses such a need by presenting the development and testing of a reliable method for yield forecasting.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 6 Sept 2026
Abstract Stand count, the number of plants per unit ground area, and leaf area index (LAI), the ratio of leaf area to ground area, are critical traits for crop research but are traditionally measured using labor-intensive methods. While new sensing technologies are being developed, quantifying improvement in measurement efficiency and data quality, relative to traditional techniques, is lacking. In this study, we use LiDAR to generate 3D scans of corn and soybean plots and evaluate two computational methods: a gap fraction approach to estimate LAI and a persistent homology algorithm to estimate stand count by detecting structural peaks in the canopy. Validation experiments and statistical comparisons of bias and variance demonstrate that LiDAR-derived LAI estimates in corn are comparable in quality to those from established instruments. However, in soybean, the LiDAR method performs poorly, likely due to dense canopies limiting light penetration and structural differentiation. Stand count estimations in corn closely match manual counts, with the added benefit of full-plot coverage and significantly faster data collection. In soybean, stand count estimates are unreliable under dense canopy conditions. These results offer practical guidance for the use of LiDAR in field phenotyping and highlight both its current capabilities and limitations. While a trade-off between speed and precision remains, particularly in high-density canopies, LiDAR’s scalability and multi-trait potential make it a promising tool for high-throughput breeding programs. Continued improvements in LiDAR hardware and algorithm design may further enhance measurement accuracy and extend applicability across crops and growth stages.
Why it matches plant phenotyping methodsLiDARによるLAI・立ち株数推定を開発・比較検証し、バイアス、分散、精度、適用限界を評価しており、植物表現型取得法が研究の中心である。
abstractwe use LiDAR to generate 3D scans of corn and soybean plots and evaluate two computational methods: a gap fraction approach to estimate LAI and a persistent homology algorithm to estimate stand count by detecting structural peaks in the canopy.
Mango is a fruit of great economic importance in India. India is the top mango-producing nation in the world, accounting for over half of global mango output. In order to determine the production capability of the insured orchards, a complete inventory is carried out in situ every three years. The inventory includes counting number of trees, grouping them into yield categories, and assessing damaged ones. Satellite Remote Sensing proves to be a vital tool for estimating ecological parameters such as population density, tree health, volume, biomass, and carbon sequestration rates. The significance of tree counting extends beyond orchard evaluations, playing a vital role in environmental protection, agricultural planning, and crop yield forecast. unfortunately, conventional tree counting methods often require very expensive feature engineering, which leads to more errors as well as lower overall optimization. In order to overcome these obstacles, deep learning-based methods have been used to count trees, exhibiting cutting-edge results in this crucial activity. This paper introduces a novel approach employing deep learning for Image-Based Mango Tree counting in high-resolution satellite imagery data. The proposed model, named Bi-directional Feature Pyramid Network (BiFPN)-YOLOv8m an improved version of YOLOv8, employs object detection to effectively separate, locate, and count mango trees with in orchards. A dataset of 1700 training and 300 testing images of mango orchards with trees of various ages is used to evaluate the various YOLOv8 variants, YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, YOLOv8x, including YOLOv9, YOLOv10, and BiFPN-YOLOv8m, with a focus on computational efficiency, accuracy, and speed. Experimental findings show that, even under difficult circumstances, the proposed method continuously outperforms state-of-the-art techniques.
Why it matches plant phenotyping methods衛星画像からマンゴー樹木を分離・位置推定・計数する深層学習手法を開発・評価しており、植物個体数という観測可能な形態・構造形質の抽出が中心である。
abstractThis paper introduces a novel approach employing deep learning for Image-Based Mango Tree counting in high-resolution satellite imagery data.
Reproduction assets foundThe paper's satellite remote sensing image dataset used for mango tree counting is publicly deposited on GitHub per the Data Availability Statement. No separate analysis code or trained model checkpoint is explicitly deposited.Dataset · publicRemote Sensing Image Data that support the findings of this study have been deposited in the GitHub. The url to the data uploaded is https://github.com/lbirla/Mango_tree_satellite_data.Open asset ↗https://github.com/lbirla/Mango_tree_satellite_datahtml-lines:497-525Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Insect pests in stor ed products cause qualitative and quantitative losses in seed lots, reducing their commercial value by directly compromising the physiological and sanitary quality of the seeds. The objective of this study was to evaluate the physiological quality and perform a proximate analysis of rice seeds infested with Sitophilus zeamais Motschulsky (Coleoptera: Curculionidae), using radiographic images. The X-ray analysis was used to detect and identify the weevil development stages and quantify the percentage of infestation in rice seeds. The physiological quality and the proximate analysis were evaluated after the seeds were subjected to four levels of infestation by S. zeamais: 0%, 2%, 3%, and 5%. The radiographic images enabled efficient detection of infestation levels, identification of the weevil's developmental stages, and assessment of damaged and empty seeds. The following physiological tests were performed: germination test, first germination count test, emergency test, retention capacity of the substrate, emergency speed index, and electrical conductivity test. For the physiological and proximate analysis, the experimental design was completely randomized, with four treatments and four replications. Statistical differences were observed in physiological assessments and proximate analysis across infestation levels, confirming that infestation intensity directly affects seed viability and nutritional value. This emphasizes the importance of effective monitoring methods to mitigate pest damage to stored seeds.
Why it matches plant phenotyping methodsX線画像を用いた種子の害虫侵入・発育段階・損傷状態の検出と侵入率の定量が研究の中心であり、種子の状態・品質という植物形質を画像から評価している。
Introduction Wheat is a vital global staple crop, where accurate ear detection and counting are essential for yield prediction and field management. However, the complexity of field environments poses significant challenges to achieving lightweight yet high-precision detection. Methods This study proposes YOLOv8-FDA, a lightweight detection and counting method based on YOLOv8. The approach integrates RFAConv for enhanced feature extraction, DySample for efficient multi-scale upsampling, HWD for compressed and accelerated model training, and the SDL loss for improved bounding box regression. Results Experimental results on the GWHD dataset show that YOLOv8-FDA achieves a precision of 86.3%, recall of 77.5%, and mAP@0.5 of 84.9%, outperforming the original YOLOv8n by significant margins. The model size is 2.96MB with a computational cost of 8.3 GFLOPs, and it operates at 19.2 FPS, enabling real-time counting with over 97.5% accuracy using cross-row segmentation. Discussion The proposed YOLOv8-FDA model demonstrates strong detection performance, lightweight characteristics, and efficient real-time capability, indicating its high practicality and suitability for deployment in real-world agricultural applications.
Why it matches plant phenotyping methodsコムギ穂の検出・計数という植物器官形質の画像ベース抽出法をYOLOv8改良モデルとして開発し、データセット上で性能検証しているため、方法が研究の中心である。
abstractThis study proposes YOLOv8-FDA, a lightweight detection and counting method based on YOLOv8.
Reproduction assets foundThe paper's wheat ear detection/counting experiments were run on the public 2021 GWHD dataset, which the authors explicitly link via a Zenodo DOI in the data availability statement. No author-specific code or trained model checkpoints are disclosed.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: Zenodo at https://doi.org/10.5281/zenodo.5092309 .Open asset ↗Zenodo · 10.5281/zenodo.5092309lines:698-734Plant phenotyping relevance match · UnverifiedEurope PMC · bioRxiv · checked 13 Sept 2026
Automated seed phenotyping has wide applications in research and agriculture and relies on easy-to-use platforms and pipelines. Seed phenotyping in the model species Arabidopsis thaliana poses a significant challenge due to the large number of tiny seeds produced by individual plants, which are difficult to manually separate and count. Manual counting methods are time-consuming and prone to user bias, particularly for subtle phenotypic changes. To address these limitations, we developed Samplify , a scalable, automated pipeline for seed segmentation and classification. By integrating classical image processing techniques with Meta’s Segment Anything Model (SAM), Samplify effectively segments Arabidopsis seeds, even in dense clusters where conventional methods fail. To demonstrate its versatility, we quantified the seed abortion occurring in interploidy crossings in Arabidopsis, often referred to as ‘triploid block’. Samplify includes a Random Forest classifier trained on a set of computed seed shape features that enables the categorization of seeds into normal, partially aborted, and fully aborted seeds, automating the manual classification process. The tool, designed as a command-line application, significantly reduces manual annotation workload. Our validation across multiple datasets demonstrates high segmentation and classification reliability, making Samplify a valuable resource for the plant research community.
Why it matches plant phenotyping methods種子の画像セグメンテーション・分類による表現型抽出パイプラインを開発し、複数データセットで検証しており、方法論が研究の中心である。
abstractwe developed Samplify , a scalable, automated pipeline for seed segmentation and classification.
Leaves are the key organs in photosynthesis and nutrient production, and leaf counting is an important indicator of banana plant health and growth rate. However, in complex orchard environments, leaves often overlap, the background is cluttered, and illumination varies, making accurate segmentation and detection challenging. To address these issues, we propose a lightweight banana leaf detection and counting method deployable on embedded devices, which integrates a space–depth-collaborative reasoning strategy with multi-scale feature enhancement to achieve efficient and precise leaf identification and counting. For complex background interference and occlusion, we design a multi-scale attention guided feature enhancement mechanism that employs a Mixed Local Channel Attention (MLCA) module and a Self-Ensembling Attention Mechanism (SEAM) to strengthen local salient feature representation, suppress background noise, and improve discriminability under occlusion. To mitigate feature drift caused by environmental changes, we introduce a task-aware dynamic scale adaptive detection head (DyHead) combined with multi-rate depthwise separable dilated convolutions (DWR_Conv) to enhance multi-scale contextual awareness and adaptive feature recognition. Furthermore, to tackle instance differentiation and counting under occlusion and overlap, we develop a detection-guided space–depth position modeling method that, based on object detection, effectively models the distribution of occluded instances through space–depth feature description, outlier removal, and adaptive clustering analysis. Experimental results demonstrate that our YOLOv8n MDSD model outperforms the baseline by 2.08% in mAP50-95, and achieves a mean absolute error (MAE) of 0.67 and a root mean square error (RMSE) of 1.01 in leaf counting, exhibiting excellent accuracy and robustness for automated banana leaf statistics.
Why it matches plant phenotyping methodsバナナ葉の検出・計数という植物形態・生育指標を対象に、複雑環境での画像解析手法を開発し、精度評価しているため、植物フェノタイピング手法が中心である。
abstractwe propose a lightweight banana leaf detection and counting method deployable on embedded devices
Introduction As a major food crop, accurate detection and counting of wheat ears in the field are of great significance for yield estimation. Aiming at the problems of low detection accuracy and large computational load of existing detection and counting methods in complex farmland environments, this study proposes a lightweight wheat ear detection model, YOLOv11-EDS. Methods First, the Dysample dynamic upsampling operator is introduced to optimize the upsampling process of feature maps and enhance feature information transmission. Second, the Direction-aware Oriented Efficient Channel Attention mechanism is introduced to make the model focus more on key features and improve the ability to capture wheat ear features. Finally, the Slim-Neck module is introduced to optimize the feature fusion structure and enhance the model's processing capability for features of different scales. Results Experimental results show that the performance of the improved YOLOv11-EDS model is significantly improved on the global wheat ear dataset. The precision is increased by 2.0 percentage points, the recall by 3.5 percentage points, mAP@0.5 by 1.5 percentage points, and mAP@0.5:0.95 by 2.5percentage points compared with the baseline model YOLOv11. Meanwhile, the model parameters are reduced to 2.5 M, and the floating-point operations are reduced to 5.8 G, which are 0. 1 M and 0.5 G lower than the baseline model, respectively, achieving dual optimization of accuracy and efficiency. The model still demonstrates excellent detection performance on a self-built iPhone-view wheat ear datasets, fully verifying its robustness and environmental adaptability. Discussion This study provides an efficient solution for the automated analysis of wheat phenotypic parameters in complex farmland environments, which is of great value for promoting the development of smart agriculture.
Why it matches plant phenotyping methods小麦穂の検出・計数という植物器官の表現型取得を目的に、YOLOモデルを開発し、公開・自作データセットで精度と頑健性を検証しているため、方法が中心である。
abstractthis study proposes a lightweight wheat ear detection model, YOLOv11-EDS.
Plant phenotyping involves analyzing observable characteristics of plants to better understand their growth, health, and development. In the context of deep learning, this analysis is often approached through single-view classification or regression models. However, these methods often fail to capture all information required for accurate estimation of target phenotypic traits, which can adversely affect plant health assessment and harvest readiness prediction. To address this, the Growth Modelling (GroMo) Grand Challenge at ACM Multimedia 2025 provides a multi-view dataset featuring multiple plants and two tasks: Plant Age Prediction and Leaf Count Estimation. Each plant is photographed from multiple heights and angles, leading to significant overlap and redundancy in the captured information. To learn view-invariant embeddings, we incorporate 24 views, referred to as the selection vector, in a random selection. Our ViewSparsifier approach won both tasks. For further improvement and as a direction for future research, we also experimented with randomized view selection across all five height levels (120 views total), referred to as selection matrices.
Why it matches plant phenotyping methods多視点画像から植物年齢と葉数を推定する手法を開発・評価し、ベンチマーク課題で性能を検証しているため、植物表現型取得が中心である。
abstractTo learn view-invariant embeddings, we incorporate 24 views, referred to as the selection vector, in a random selection.
The aim of this study is to propose a lightweight YOLOv8n maize seedling detection algorithm that incorporates multi-scale features to address the problems of large number of model parameters and computation, low real-time performance, and small detection range of the existing maize seedling detection models during plant detection. By fusing RepConv with HGNetV2 using the idea of reparameterisation, a Rep_HGBlock structure is designed to form a new lightweight backbone network, Rep_HGNetV2,; BiFPN is introduced into the neck network portion of the model to enhance the interactive fusion of bidirectional information flow between multiple scales and hierarchies; and a fusion task decomposition, dynamic convolutional alignment is designed, DFL (Distribution Focal Loss) ideas, TDADH, a task dynamically aligned detection head, which uses shared convolution and dynamically aligns the tasks of classification and localization to extract features; and Grad-CAM++ technique is used to generate a heat map for model detection, visualize effective features of the target and understand the model focus region. The experimental results show that the improved model achieves a detection accuracy of 96.5%, which is basically the same as the original model. The weight size, number of parameters, and computational FLOPs are reduced to 3.5 MB, 1.58 M, and 7.4 G, respectively, which are reduced by about 43%, 47%, and 8.6%. The frame rate FPS is only reduced from 149.98 to 146.3, a reduction of about 2.4%. The results show that the lightweight model has high recognition accuracy, speed and low complexity, which is more suitable for practical deployment in resource-constrained edge devices, UAVs, and embedded systems, and is able to provide technical support for the precise management of maize during the seedling stage of drip irrigation water-fertilizer integration.
Why it matches plant phenotyping methodsトウモロコシ苗の検出・計数という植物個体数の取得を目的に、軽量YOLOv8nアルゴリズムを開発・評価しており、フェノタイピング手法が中心である。
abstractThe aim of this study is to propose a lightweight YOLOv8n maize seedling detection algorithm
The article is devoted to the development and testing of technology for recognizing pea sprouts and estimating its biomass based on images from UAVs using neural networks. Rocket peas were sown by the “Kuzbass” sowing complex in the Topkinsky district of the Kemerovo Region on an area of 21.55 hectares. The soil type is slightly leached chernozem. The predecessor is spring wheat. The seed depth is 6 cm, the seeding rate is 1.1 million seeds per 1 hectare. Aerial photography was performed three weeks later with a quadcopter with a 20MP camera resolution from a flight altitude of 3 m. The shooting was carried out in two stages — in the early morning in cloudy conditions to obtain images of pea shoots without shadows and in the daytime with shadows from sprouts and weeds. As a result, two sets of 120 source photos were generated to train the neural network. Based on the obtained datasets, the Ultralytics YOLOv8 neural network model was trained. Testing of the obtained models was performed in a Python program for batch image processing and counting the number of plants in each image. The accuracy of recognizing sprouts on the first dataset was 97.3%, on the second — 67.3%. This is due to the different shooting conditions. Combining the two datasets allowed for a recognition accuracy of 94.7%. This is slightly lower than the first option, but much closer to the actual conditions of aerial photography. The result of the work is a program that allows batch image processing for automatic counting of pea sprouts and calculating their area in the images.
Why it matches plant phenotyping methodsUAV画像とニューラルネットワークを用いて、エンドウ苗の認識、個体数計数、面積およびバイオマス推定技術を開発・検証しており、表現型取得手法が研究の中心である。
abstractthe development and testing of technology for recognizing pea sprouts and estimating its biomass based on images from UAVs using neural networks
Rice serves as the staple food for over 50% of the world's population, making its yield prediction crucial for food security. The number of panicles per unit area is a core parameter for estimating rice yield. However, traditional manual counting methods suffer from low efficiency and significant subjective bias, while unmanned aerial vehicle (UAV) images used for panicle detection face challenges such as densely distributed panicles, large scale variations, and severe occlusion. To address the above challenges, this paper proposes a rice panicle detection model based on an improved You Only Look Once version 11x (YOLOv11x) architecture. The main improvements include: 1) Introducing a Bi-level Routing Attention (BRA) mechanism into the backbone network to improve the feature representation capability for small objects; 2) Adopting a Transformer-based detection head (TransHead) to capture long-term spatial dependencies; 3) Integrating a Selective Kernel (SK) Attention module to achieve dynamic multi-scale feature fusion; 4) Designing a multi-level feature fusion architecture to enhance multi-scale adaptability. Experimental results demonstrate that the improved model achieves an mAP@0.5 of 89.4% on our self-built dataset, representing a 3% improvement over the baseline YOLOv11x model. It also achieves a Precision of 87.3% and an F1-score of 84.1%, significantly outperforming mainstream algorithms such as YOLOv8 and Faster R-CNN. Additionally, panicle counting tests conducted on 300 rice panicle images show that the improved model achieves R 2 = 0.85, RMSE = 2.33, and rRMSE = 0.13, indicating a good fitting effect. The proposed model provides a reliable solution for intelligent in-field rice panicle detection using UAV images and holds significant importance for precise rice yield estimation.
Why it matches plant phenotyping methodsUAV画像からイネ穂数を検出・計数する深層学習手法を開発し、データセット上で性能検証しており、植物形質取得が研究の中心である。
abstractthis paper proposes a rice panicle detection model based on an improved You Only Look Once version 11x (YOLOv11x) architecture.
Accurate quantification of open bolls and their distribution is crucial for understanding cotton growth, development, and yield in optimized crop management and enhanced plant breeding. Manual boll counting methods are time-consuming, labor-intensive, and subjective. Leveraging the potential of high-resolution images for high-throughput phenotyping offers a promising avenue for efficient trait quantification. The objectives of this study were to develop methods to detect and count open cotton bolls using LiDAR point cloud and RGB images and to compare the effectiveness of these two data sources. A DJI Phantom 4 RTK Unmanned Aerial System (UAS) equipped with a 4 K RGB camera was used to acquire high-resolution RGB images, and a DJI Matrice 300 RTK with a Zenmuse L1 sensor was used to acquire LiDAR point cloud data. The RGB images were converted to point cloud using photogrammetry by measuring multiple points of overlapping images. The boll detection workflow involved data filtering and clustering using the density-based spatial clustering of applications with noise (DBSCAN) method. Evaluation of the methods involved 48 plots representing small, medium, and large plant sizes using metrics including mean absolute percentage error (MAPE), root mean square error (RMSE), and coefficient of determination (r²). The methods using both data sources performed well in estimating open bolls, with LiDAR point cloud data slightly outperforming those derived from RGB images. Generally, the performance of the DBSCAN method in boll detection improved with decreasing plant sizes. Specifically, LiDAR data yielded MAPE values of 5.03 %, 8.05 %, and 13.46 %, RMSE values of 7.26, 14.33, and 23.40 bolls per m², and r² values of 0.93, 0.84, and 0.84 for small, medium, and large plant sizes, respectively. RGB image-based data exhibited MAPE values of 7.21 %, 6.49 %, and 16.41 %, RMSE values of 11.05, 13.66, and 26.49 bolls per m², and r² values of 0.82, 0.74, and 0.83 for small, medium, and large plant sizes, respectively. The method demonstrates the potential of RGB imagery and LiDAR data for estimating boll counts, offering valuable tools for enhanced plant phenotyping in plant breeding and site-specific crop management. Both data sources underestimated boll counts, with smaller plants showing less undercounting, likely due to improved light penetration and separation of bolls. These findings highlight the influence of plant structure on boll detection accuracy and the need to address challenges posed by dense canopies to enhance detection reliability.
Why it matches plant phenotyping methodsLiDARとRGB画像を用いた綿花の開花ボール数の検出・計数手法を開発し、比較評価した研究であり、植物表現型取得が中心です。
abstractThe objectives of this study were to develop methods to detect and count open cotton bolls using LiDAR point cloud and RGB images and to compare the effectiveness of these two data sources.
Fruit shape and stomatal distribution influence fruit water status and postharvest quality in southern highbush blueberry ( Vaccinium corymbosum hybrids). This study developed scalable phenotyping methods to quantify fruit morphology and stomatal traits in cultivars Colossus and Optimus across four developmental stages. Fruit volume was estimated using geometric models validated against three-dimensional (3D) scans, with the spheroid model offering the best compromise between accuracy and efficiency ( R 2 = 0.96). StoManager1 software was validated for automated stomatal phenotyping, showing strong concordance with manual counts ( R 2 = 0.96). Results revealed cultivar-specific differences in shape development and stomatal distribution. As fruits matured, both cultivars exhibited increases in volume and surface area with decreasing sphericity. Stomata were localized primarily to distal regions of the fruit, particularly the calyx, suggesting heterogeneous water loss pathways across the fruit surface. These findings establish a framework for integrating morphometric and anatomical traits into high-throughput phenotyping pipelines and future studies on fruit water relations.
Why it matches plant phenotyping methods果実形態と気孔形質を定量化するスケーラブルな表現型解析法を開発し、3Dスキャンおよび手動計数で検証しており、方法開発・検証が研究の中心である。
abstractThis study developed scalable phenotyping methods to quantify fruit morphology and stomatal traits
Downy mildew is one of the most destructive diseases in wine-growing regions, severely reducing yield and fruit quality. Traditional detection methods rely on expert scouting, which is labour-intensive, subjective and often imprecise. In this context, Artificial intelligence (AI) offers a promising alternative, enabling the use of conventional RGB images for field level disease detection. Therefore, this study proposes a deep learning-based approach to identify leaves with downy mildew symptoms using canopy-level RGB imagery. A comprehensive dataset was generated from fourteen commercial blocks with varying cultivars and disease intensities in northern Spain. RGB images of the canopy were collected under different lighting conditions throughout the day to increase the dataset variability. The YOLOv4 (You Only Look Once) algorithm was trained using this heterogeneous dataset to enhance robustness of the model. The model achieved a mAP of 67 %, an F1-score of 0.69 and an IoU of 62 % in the testing process applied on full canopy images. The number of infected leaves predicted by the model closely matched expert annotation reaching a determination coefficient R² of 0.93. Detection performance remained consistent across different infection levels, suggesting the model’s adaptability to a wide range of conditions. Furthermore, the model was able to accurately localise symptomatic leaves within the full canopy. The results of this study demonstrate that RGB images of the whole canopies can be effectively used to detect downy mildew symptoms, offering a practical approach for in-field disease monitoring. The proposed methodology facilitates the development of automated, on-the-go disease detection systems using mobile platforms such as agricultural robots or vehicles, thereby enabling real-time crop health assessment.
Why it matches plant phenotyping methodsブドウ樹冠のRGB画像からうどんこ病症状葉を検出・計数する深層学習手法を開発し、専門家アノテーションと性能検証を行っており、植物病害状態の取得が中心である。
abstractthis study proposes a deep learning-based approach to identify leaves with downy mildew symptoms using canopy-level RGB imagery
Root hair counting is a specialized aspect of plant biology and agronomy research that offers valuable insights into plant health, nutrient uptake, and overall growth potential. Root hairs are tiny extensions from the root epidermis that significantly increase the surface area and constitute roughly 70% of the total root area of a plant root system, enhancing the ability of plants to absorb water and nutrients from the soil. Understanding the importance of root hair counting involves looking at various aspects of plant physiology and soil–plant interactions. Despite these benefits, counting root hairs, especially manually, can be tedious, time-consuming, and, more often, inaccurate due to differences in the perception of individuals. Therefore, we have proposed a novel method for root hair counting and further observed an improvement in root hair count measurements when utilizing image super-resolution as a preprocessing step. Our approach of counting root hairs can tackle real-world challenges and be able to count overlapping hairs as well. By visualizing the rhizosphere in binary space, we can see a considerable increase in root hair count from 37 to 68 when counting manually to our approach for Bell pepper, and from 44 to 88, when counting manually to our method for Arabidopsis root images. To the best of the authors’ knowledge, this research study is specifically designed for root hair counting and measurement improvement using super-resolution, is the first of its kind, and has yet to be acknowledged.
Why it matches plant phenotyping methods植物の根毛数という形態形質を画像から自動計測する新規手法を開発し、超解像前処理による測定改善も検証しており、フェノタイピング手法が中心である。
titleA Novel Approach for Plant Root Hair Counting and its Improvement via Image Super-Resolution
Thinning is a critical practice in apple orchard management, directly influencing crop load and fruit quality. To assist automated crop load management, a machine vision system for apple bud detection was developed to be integrated with robotic platforms. The system employed a Kinect Azure sensor for real-time bud detection and branch diameter measurement, utilizing a YOLOv8-based object detection model trained and evaluated across multiple datasets. The evaluation identified the best-performing model by balancing precision, recall, and robustness in the complex and unstructured environments of apple orchards. Several training configurations were assessed, with the selected setup demonstrating a strong balance between precision (68 %), recall (55 %), F1-score (61 %), and mean average precision (mAP: 59 %) across diverse and unstructured orchard environments. This configuration, trained on a combination of FLIR and Kinect Azure data, was chosen for deployment due to its robustness and compatibility with the Kinect Azure sensor in real-world applications. Two proposed imaging methods for branch diameter measurement were validated against manual caliper-based measurements, with statistical analysis revealing no significant differences (p = 0.98). These findings confirm the semi-automated methods as reliable and labor-efficient alternatives for field applications. Additionally, the bud counting algorithm demonstrated accurate tracking and counting of apple buds, effectively avoiding omissions and duplications in real orchard settings. This study underscores the potential of vision systems to revolutionize apple bud thinning, providing a strong foundation for the development of fully automated solutions in precision orchard management.
Why it matches plant phenotyping methodsリンゴ芽の画像検出に加え、枝径という植物形質の画像計測法を開発・手動測定と検証しており、フェノタイピング手法が中心である。
abstracta machine vision system for apple bud detection was developed to be integrated with robotic platforms
Accurate identification and counting of wheat ear-heads are critical for reliable crop yield estimation. This study presents a novel pipeline utilizing the YOLOv8 model specifically designed for detecting wheat ear-heads in challenging agricultural environments. Leveraging orthomosaic imagery captured by drones, our approach integrates several advanced techniques to enhance detection accuracy. The pipeline begins with the identification of plots from orthomosaic images, followed by extraction and tiling of these plots for detailed analysis. The detection model is applied to the tiled images, and the results are stitched together to reconstruct the original image annotated with detected wheat ear-heads. To improve image quality—addressing issues like brightness, contrast, exposure, and blur—we employed the Real ESR GAN technique alongside a random cut out method for effective occlusion handling. Evaluations on Mahyco's dataset demonstrated a mean average precision (mAP) of 99.2% for plot detection and an accuracy of 86.7% for wheat ear-head detection against ground truth. Our model exhibited robust performance across varying growth stages and adverse conditions, underscoring its potential for practical agricultural applications. This research introduces an end-to-end pipeline that automates wheat ear-head detection, enabling scalable in-season yield prediction and providing a valuable tool for farmers and agronomists to enhance crop management decisions. Furthermore, our pipeline can be adapted into a desktop app or web portal, allowing farmers to upload orthomosaic images and receive an output Excel file with plot numbers and corresponding wheat ear-head counts, thus enabling comprehensive wheat yield estimation for their farmland.
Why it matches plant phenotyping methodsドローン画像から小麦穂数を自動検出・計数し、収量推定に用いる画像解析パイプラインを開発・評価しており、植物形質の取得方法が中心である。
abstractThis study presents a novel pipeline utilizing the YOLOv8 model specifically designed for detecting wheat ear-heads in challenging agricultural environments.
Background and scope Trunks of saguaro cacti (Carnegiea gigantea) grow for many years, and during this time the shoot apical meristem (SAM) of each trunk not only grows in diameter, it also initiates new orthostichies (ribs). Several questions were examined. Is a saguaro SAM's diameter correlated with the number of orthostichies/ribs it is producing? Is SAM diameter tightly controlled, or does it vary among individuals of the same age? When saguaro trunks are ~3 m tall, their SAMs stop adding new orthostichies/ribs: do SAMs stop growing only after reaching a critical diameter, or do the SAMs vary in diameter when each stops growing? Methods Ribs were counted at various heights (corresponding to various ages) on saguaro plants in habitat. Shoot apical meristem diameter was measured by light microscopy in sectioned material. Shoot apical meristems of Echinocactus grusonii were also studied. Key results Shoot apical meristem diameter is strongly correlated with the number of ribs being maintained: the circumferential distance between newly initiated leaf primordia remains constant (145 ± 10.6 µm in C. gigantea; 193 ± 10.7 µm in E. grusonii) even as an SAM grows in diameter. An SAM's diameter and circumference can be estimated by counting the number of ribs it is maintaining. The diameter of each SAM of C. gigantea increases for many years but it eventually stabilizes; the final, stable diameter of each C. gigantea SAM varies from shoot to shoot. Conclusions Shoot apical meristem diameter in both species can be estimated non-destructively by simply counting the number of orthostichies/ribs the SAM is producing (or produced in the past). The growth rate of C. gigantea SAMs varies from plant to plant and can change with age. All C. gigantea SAMs stop increasing in diameter at some point, but that diameter varies from plant to plant.
Why it matches plant phenotyping methodsサボテンのシュート頂端分裂組織径という植物形態形質を、肋数から非破壊推定する測定法が研究の中心であり、手法の成立性と適用結果を示している。
abstractAn SAM's diameter and circumference can be estimated by counting the number of ribs it is maintaining.
Introduction Accurate seed counting is an essential task in agricultural research and farming, supporting activities such as crop breeding, yield prediction, and weed management. Traditional manual seed counting, while accurate, is time-consuming, labor-intensive, and prone to human error, particularly for large quantities of micro-sized seeds. Methods This study developed two automated computer vision approaches integrated into a mobile application (app) for seed counting: one utilizing image processing (IP) and the other based on deep learning (DL). These methods aim to address the limitations of traditional manual counting by providing automated, efficient alternatives. Results The IP-based method demonstrated high accuracy comparable to manual counting and offered substantial time savings. However, its reliance on controlled environmental conditions, such as uniform lighting, limits its versatility for field apps. The DL-based method excelled in speed and scalability, processing counts in as little as 0.33 seconds per image, but its accuracy was inconsistent for visually complex or densely clustered seeds. Discussion Both automated methods significantly enhance the efficiency of seed counting, providing a practical and accessible solution for various agricultural contexts. The integration of these methods into a mobile app streamlines seed counting for laboratory research, field studies, seed production, and breeding trials, offering a transformative approach to modernizing seed counting practices while reducing time and labor requirements.
Why it matches plant phenotyping methods画像処理と深層学習による種子計数手法およびモバイルアプリを開発・評価しており、植物種子数の取得が研究の中心です。
abstractThis study developed two automated computer vision approaches integrated into a mobile application (app) for seed counting: one utilizing image processing (IP) and the other based on deep learning (DL).
Background Accurate sorghum spike detection is critical for monitoring growth conditions, accurately predicting yield, and ensuring food security. Deep learning models have improved the accuracy of spike detection thanks to advances in artificial intelligence. However, the dense distribution of sorghum spikes, variable sizes and complex background information in UAV images make detection and counting difficult. Methods We propose a multiscale and oriented sorghum spike detection and counting model in UAV images (MOSSNet). The model creates a Deformable Convolution Spatial Attention (DCSA) module to improve the network's ability to capture small sorghum spike features. It also integrated Circular Smooth Labels (CSL) to effectively represent morphological features. The model also employs a Wise IoU-based localization loss function to improve network loss. Results Results show that MOSSNet accurately counts sorghum spike under field conditions, achieving mAP of 90.3%. MOSSNet shows excellent performance in predicting spike orientation, with RMSEa and MAEa of 14.6 and 12.5 respectively, outperforming other directional detection algorithms. Compared to general object detection algorithms which output horizonal detection boxes, MOSSNet also demonstrates high efficiency in counting sorghum spikes, with RMSE and MAE values of 9.3 and 8.1, respectively. Discussion Sorghum spikes have a slender morphology and their orientation angles tend to be highly variable in natural environments. MOSSNet 's ability has been proved to handle complex scenes with dense distribution, strong occlusion, and complicated background information. This highlights its robustness and generalizability, making it an effective tool for sorghum spike detection and counting. In the future, we plan to further explore the detection capabilities of MOSSNet at different stages of sorghum growth. This will involve implementing object model improvements tailored to each stage and developing a real-time workflow for accurate sorghum spike detection and counting.
Why it matches plant phenotyping methodsUAV画像からソルガム穂の検出・計数・向き推定を行う深層学習手法を開発し、精度評価も実施しており、植物形態形質の取得手法が中心である。
abstractWe propose a multiscale and oriented sorghum spike detection and counting model in UAV images (MOSSNet).
Leaf detection and counting are essential in plant phenotyping, but traditional manual methods are slow and error prone. To improve the efficiency and accuracy of leaf counting, this study introduces a lightweight, high-precision model for leaf detection and counting based on the optimized YOLOv8 computer vision model, called MobileViT-Asymptotic Feature Pyramid Network-YOLOv8 (MAF-YOLOv8). This model integrates the MobileViT architecture and an Adaptive Feature Pyramid Network (AFPN) structure, achieving a lightweight model with enhanced feature representation capabilities, thereby improving leaf counting accuracy. In this work, we constructed a dataset consisting of 711 RGB images with a 640 × 640 resolution and expanded it to 2136 images using data augmentation methods to enhance model robustness. The MAF-YOLOv8 model was able to achieve a mean average precision (mAP) of 91.7 %, a recall of 95.0 %, and a precision of 86.5 % in leaf counting tasks. Compared to YOLOv8, mAP improved by 2.6 %, while the number of parameters was reduced by 34.1 %. Ablation experiments evaluating the contributions of each model component further confirmed that MobileViT and AFPN critically improve model performance; together, they led to a 2.3 % improvement in precision and a 3 % increase in recall. This study also validates the performance of MAF-YOLOv8 on resource-constrained mobile devices, achieving an inference time of only 5.110 s. The findings indicate that the model has superior accuracy, inference speed, and resource efficiency, which renders it suitable for extensive applications within agriculture and environmental monitoring. This study provides an efficient technological approach for plant phenotyping and precision agriculture. • Replaced YOLOv8's backbone with MobileViT to reduce model size. • Integrated the Asymptotic Feature Pyramid Network to improve model accuracy. • The robustness of MAF-YOLOv8 is validated using a multi-environment dataset. • MAF-YOLOv8 demonstrates efficient performance when deployed on Raspberry Pi.
Why it matches plant phenotyping methods植物葉の検出・カウントという表現型取得を目的に、YOLOv8を改良した画像解析モデルを開発し、精度・頑健性・モバイル実装を検証しているため、植物フェノタイピング手法が中心である。
abstractTo improve the efficiency and accuracy of leaf counting, this study introduces a lightweight, high-precision model for leaf detection and counting based on the optimized YOLOv8 computer vision model
Yield mapping in agricultural crops remains a significant challenge, particularly in uncontrolled environments. This study evaluates four instance segmentation algorithms: YOLOv8n, YOLOv8s, YOLOv8m, and YOLOv8l, along with a low-cost GNSS RTK system to detect and count strawberries in a hydroponic environment. A depth camera is used to remove background information from nonrelevant furrows, improving fruit detection accuracy. The low-cost RTK receiver, configured in Base Rover mode, provides centimeter-level precision and enables the generation of detailed yield maps that can be seamlessly integrated into commercial systems to increase growers' yield and profit. Data were collected in the municipality of Arcabuco, Boyacá (Colombia), resulting in 8848 images processed after augmentation. Among the models evaluated, YOLOv8l achieved the highest performance with a maximum F1-Score of 0.9295 and a mAP50 of 0.9689 during validation. Furthermore, in the fruit counting process - evaluated against manual counts - the same model achieved a R 2 of 0.9997 and a mean relative error (MRE) of 1.5511%. In general, this work presents a systematic methodology for the extraction and visualization of information in fruit crops using computer vision and deep learning, showcasing a robust yield mapping system. The approach integrates pre-processing and post-processing steps, as well as 2D–3D image acquisition, georeferencing, and processing technologies, offering thus a novel solution for accurate and efficient hydroponic strawberry yield mapping. • Evaluation of instance segmentation to detect and count hydroponic strawberries. • 3D camera integration to remove background noise and improve fruit detection. • Use of GNSS RTK with centimeter accuracy for fruit localization in yield maps. • Real-world yield mapping system showing exact strawberry count and location. • Pre/postprocessing and georeferencing for a yield mapping approach in strawberries.
Why it matches plant phenotyping methodsイチゴ果実の検出・計数という植物器官の収量形質を、RGB-D画像、深層学習、GNSS RTKで取得・抽出・地図化する方法が研究の中心である。
abstractThis study evaluates four instance segmentation algorithms: YOLOv8n, YOLOv8s, YOLOv8m, and YOLOv8l, along with a low-cost GNSS RTK system to detect and count strawberries in a hydroponic environment.
This study presents an innovative unmanned aerial vehicle (UAV)-based intelligent detection method utilizing an improved Faster Region-based Convolutional Neural Network (Faster R-CNN) architecture to address the inefficiency and inaccuracy inherent in manual wheat spike counting. We systematically collected a high-resolution image dataset (2000 images, 4096 × 3072 pixels) covering key growth stages (heading, grain filling, and maturity) of winter wheat ( Triticum aestivum L.) during 2022-2023 using a DJI M300 RTK equipped with multispectral sensors. The dataset encompasses diverse field scenarios under five fertilization treatments (organic-only, organic-inorganic 7:3 and 3:7 ratios, inorganic-only, and no fertilizer) and two irrigation regimes (full and deficit irrigation), ensuring representativeness and generalizability. For model development, we replaced conventional VGG16 with ResNet-50 as the backbone network, incorporating residual connections and channel attention mechanisms to achieve 92.1% mean average precision (mAP) while reducing parameters from 135 M to 77 M (43% decrease). The GFLOPS of the improved model has been reduced from 1.9 to 1.7, an decrease of 10.53%, and the computational efficiency of the model has been improved. Performance tests demonstrated a 15% reduction in missed detection rate compared to YOLOv8 in dense canopies, with spike count regression analysis yielding R 2 = 0.88 ( p 2 ) and varying illumination conditions, maintaining >85% accuracy even under cloudy weather. Furthermore, by integrating spike recognition with agronomic parameters (e.g., grain weight), we developed a comprehensive yield estimation model achieving 93.5% accuracy under optimal water-fertilizer management (70% ETc irrigation with 3:7 organic-inorganic ratio). This work systematically addresses key technical challenges in automated spike detection through standardized data acquisition, lightweight model design, and field validation, offering significant practical value for smart agriculture development.
Why it matches plant phenotyping methodsUAV画像からコムギ穂を認識・計数する画像解析手法を開発し、データセット、モデル改良、比較検証、圃場性能評価まで行っており、植物形質取得が研究の中心です。
abstractThis study presents an innovative unmanned aerial vehicle (UAV)-based intelligent detection method utilizing an improved Faster Region-based Convolutional Neural Network (Faster R-CNN) architecture to address the inefficiency and inaccuracy inherent in manual wheat spike counting.
The number of spike grains is an important parameter for wheat yield estimation. However, it is challenging to automatically and intelligently count wheat spike grains in the open field environment. In this study, a deep learning framework, called Wheat Spike Grain Point-to-Point Network (WSG-P2PNet), is proposed to count and locate the wheat spike grains in the open field environment. This framework incorporates Efficient Channel Attention (ECA) and Coordinate Attention (CA) after feature extraction and feature concatenation, respectively. These mechanisms effectively highlight the channel features and positional information of the wheat spike grains while suppressing background interference from factors such as stems, leaves and wheat ears. Additionally, standard convolutions in the regression and classification branches are replaced with Spatial and Channel reconstruction Convolutions (SCConv), further enhancing representational capabilities and improving model performance. The results demonstrate that WSG-P2PNet, using VGG19_bn as the backbone network, outperforms five other state-of-the-art methods in terms of accuracy and stability, with an MAE of 1.72 (95% CI 1.67, 1.77), an Acc of 94.93% (95% CI 94.92, 94.93), an RMSE of 2.35 (95% CI 2.26, 2.44), and an R² of 0.8311 (95% CI 0.8218, 0.8404). Ablation experiments illustrate the impact of SCConv, ECA, and CA on the performance of WSG-P2PNet. Notably, WSG-P2PNet still maintains high accuracy in different varieties and growth periods, demonstrating its robustness and generalizability in real-world scenarios. Preliminary experiments also evaluated the correlation between predicted spike grain numbers and wheat yield, with an average Pearson Correlation Coefficient r of 0.7944, indicating a strong positive statistical relationship. The proposed deep learning framework enables rapid and accurate counting and localization of wheat spike grains in the open field environment, which is of great significant for integrated wheat yield estimation.
Why it matches plant phenotyping methodsコムギ穂粒数という植物形態・収量関連形質を、圃場画像から深層学習で計数・位置推定する手法を開発し、比較・アブレーション・異品種および生育期で検証しているため、方法が中心である。
abstracta deep learning framework, called Wheat Spike Grain Point-to-Point Network (WSG-P2PNet), is proposed to count and locate the wheat spike grains in the open field environment.
The number of stems in wheat populations is a fundamental parameter to achieve high yields and a critical agronomic trait in wheat production and variety selection. Although smart agricultural technology can estimate various agronomic parameters, the wheat stem is often obscured by multiple canopy leaves, making estimation challenging. Consequently, the current method to determine the stem number predominantly relies on labor-intensive manual techniques, which are inefficient and significantly influenced by subjective factors. This study proposes the use of augmented reality (AR) glasses as an imaging data acquisition tool to detect the number of wheat stems with high precision based on features from the top canopy and lateral images of wheat clusters. Following a correlation analysis, four color features, Coverage, the texture feature Contrast, and two lateral peak features SI (Peaks1 and Peaks2) of the top canopy image were identified. The study comparatively analyzed the image features from three perspectives for their accuracy in detecting the number of wheat stems. The results indicated a strong correlation between the peak feature (SI) and the number of wheat stems with an R² value above 0.75. The estimation using only canopy image features (CC) resulted in significant errors, where the RMSE was 20 under high-density planting conditions. Using only Peaks1 and Peaks2 yielded higher accuracy in the stem estimation, but uncertainties persisted in some high-density scenarios. Furthermore, the study combined CC and SI for the estimation and used a random forest algorithm to construct a stem estimation model. This model maintained an RMSE below 10, even under high planting densities and below 5 under low densities, which demonstrated high accuracy. This study could provide insights into stem detection for crops similar to wheat and offer a reference for other studies that require hands-free and first-person perspective image acquisition.
Why it matches plant phenotyping methodsARスマートグラスによる多視点画像取得と画像特徴・ランダムフォレストを用いて小麦の茎数を推定する手法が研究の中心であり、植物形質の取得・抽出方法を実質的に開発・評価している。
abstractThis study proposes the use of augmented reality (AR) glasses as an imaging data acquisition tool to detect the number of wheat stems with high precision based on features from the top canopy and lateral images of wheat clusters.
Flowering is one of the most important and sensitive processes throughout a plant's life and marks the start of the reproductive phase. Flowering traits largely define yield potential and are therefore crucial for crop breeding. To observe flowering dynamics under field conditions, visual ratings have been a standard method for decades. Today, high-throughput field phenotyping (HTFP) methods provide opportunities for objective and efficient data collection. We developed an object detection approach (based on YOLOv8) that allows to collect detailed data about flower and pod density. RGB-images from 12 pea breeding lines were automatically acquired by the field phenotyping platform (FIP) of ETH Zurich in two years. The trained model reached high accuracy for open flower detection, which allowed to monitor flowering dynamics and flower density over time. Maximal flower density (Max.Fl.Dens) was highly correlated (R 2 = 0.967) to ground truth data taken in the field. Clear differences in timing of flowering and flower density were detected between breeding lines and years. Furthermore, a high correlation was observed between the maximal flower density and yield components. This automated, data-driven method of flower and pod detection proved itself as a reliable tool. Therefore, the results are promising for the use of RGB imaging methods to objectively assess not only flowering dynamics but also flower density and fruiting efficiency. Maximal flower density allows to predict seed amount and therefore has potential as selection trait in breeding programs. Fruiting efficiency could be used to identify stress-tolerant breeding lines.
Why it matches plant phenotyping methods花と莢の密度をRGB画像から自動推定する物体検出法を開発し、精度を地上真値と比較検証しており、植物表現型取得が研究の中心です。
abstractWe developed an object detection approach (based on YOLOv8) that allows to collect detailed data about flower and pod density.
PURPOSE: Efficient orchard management requires high-throughput phenotyping technologies to assist growers in crop monitoring and decision-making. This study presents Agrosense, an advanced artificial intelligence (AI) powered sensing system designed for real-time phenotypic data collection in orchards, addressing the limitations of traditional manual methods. METHODS: Agrosense integrates four RGB-D cameras with a Jetson Xavier microprocessor to collect high-resolution data and perform tree crop counting, canopy density classification, and tree height estimation. A citrus orchard served as a case study, where 337 trees were imaged to train and validate AI models. YOLOv8 was employed for object detection and classification tasks, while five methods were tested for estimating tree height. RESULTS: The YOLOv8 model achieved a mean average precision (mAP) of 0.977 for tree trunkdetection and 0.974 for canopy density classification. In field testing on 157 citrus trees, the system achieved 95% accuracy for tree trunk detection and 94% accuracy for canopy density classification, with only 11 misclassifications. The best-performing method for height estimation achieved a mean absolute percentage error (MAPE) of 8.53%. Agrosense completed phenotyping tasks in 398 s, a 515% speed improvement over manual methods (2,446 s). CONCLUSION: Agrosense effectively supports precision orchard management by automating key phenotyping tasks with high accuracy and efficiency. The system significantly reduces data collection time and improves consistency. Future work will focus on algorithm refinement and adaptation to other tree crops to broaden the system’s utility in precision agriculture.
Why it matches plant phenotyping methodsRGB-DカメラとAIを統合した果樹フェノタイピングシステムを開発・検証し、樹冠密度や樹高などの形質を定量化しているため、方法が研究の中心である。
abstractThis study presents Agrosense, an advanced artificial intelligence (AI) powered sensing system designed for real-time phenotypic data collection in orchards
This study proposes a method for maize seedling reconstruction and spatial distribution analysis based on ground-based laser three-dimensional point cloud scanning technology. Using high-precision terrestrial laser scanning (TLS), 3D point cloud data was collected from multiple maize seedling plots, followed by detailed preprocessing and analysis using Trimble Realworks. During the data processing, a regression-based empirical formula, grounded in maize seedling growth characteristics, was proposed. This formula effectively mitigates the challenges of leaf occlusion in densely planted conditions, providing a solution for further point cloud segmentation and analysis. In terms of algorithm design, this study combines DBSCAN and K-means clustering algorithms to effectively overcome the challenges posed by the dense distribution of plants, leaf occlusion, and noise in the point cloud data. Through this multi-clustering approach, plant positions and distributions were accurately identified, row and column spacing calculations were optimized, and a missing plant detection function was implemented. Furthermore, a dynamic plant height calculation method based on ground undulation was proposed, significantly improving the accuracy of plant height measurement and addressing errors caused by terrain variations. Experimental results show that the proposed algorithm achieves high accuracy and robustness across multiple experimental plots, with a plant counting accuracy rate of 98.33%, a row and column spacing deviation rate controlled within 5%, and a plant height calculation accuracy exceeding 97%. These results demonstrate the effectiveness of this method in precise measurement and spatial distribution analysis during the maize seedling stage, providing strong support for precision agriculture. In the future, with further optimization of the technology, this method could be widely applied in agricultural automation and intelligent management.
Why it matches plant phenotyping methodsTLS点云、クラスタリング、遮蔽補正、動的草丈推定を組み合わせたトウモロコシ個体の再構成・形質計測法が研究の中心であり、精度評価も実施している。
abstractThis study proposes a method for maize seedling reconstruction and spatial distribution analysis based on ground-based laser three-dimensional point cloud scanning technology.
Dwarf tomatoes, with high edible and ornamental value, require monitoring multiple growth parameters to balance yield and aesthetics. While deep learning has been widely applied in phenotype monitoring, most studies focus on individual growth parameters, overlooking intrinsic relationships. To simultaneously monitor multiple growth parameters across the entire growth stage and different cultivars, this study develops a multi-modal multi-task phenotype monitoring network for dwarf tomatoes (TomPhenoNet). The network model utilizes top-view RGB-D images to evaluate four key growth parameters: height, leaf area, fresh weight, and the number of red fruits. TomPhenoNet generates mask images, fruit detection features, and the number of detected fruits based on RGB images. By fusing RGB-D images, mask images, and fruit detection features, and introducing the cross-stitch network, the network predicts plant height, leaf area, and fresh weight. The predicted values are further used to generate the dynamic occlusion coefficient, adjusting the number of detected fruits to accurately predict the number of red fruits. Results reveal that TomPhenoNet achieves high prediction performances, with R² values of 0.828, 0.930, 0.945, and 0.881 for plant height, leaf area, fresh weight, and the number of red fruits, respectively. Ablation experiments show that the cross-stitch network and fruit detection features improve the prediction performances of growth parameters, with TomPhenoNet combining both modules performing best. Feature importance analysis indicates the network model captures plant growth characteristics and corrects the impact of leaf occlusion from the top view. This study promotes accurate tomato monitoring and provides data support for optimizing cultivation strategies.
Why it matches plant phenotyping methodsトマトの複数形質をRGB-D画像から推定するマルチモーダル・マルチタスク手法を開発し、性能評価とアブレーション実験まで行っており、表現型取得・推定法が研究の中心である。
abstractthis study develops a multi-modal multi-task phenotype monitoring network for dwarf tomatoes (TomPhenoNet).
Quantifying the number of Amorphophallus konjac (Konjac) plants can provide valuable insights for yield prediction. Early monitoring of the plant population facilitates timely adjustments in cultivation practices, ultimately leading to improved productivity of Konjac. The majority of research employed deep learning (DL) for plant counting using original images derived from unmanned aerial vehicle (UAV) or ground-based platforms, but this method may lack adaptability to different scenarios and face challenges in achieving plant counting over large areas. This study systematically evaluated the performance of UAV-based original images, the generated orthomosaic, and the combination of both for the detection and counting of the Konjac plant. We proposed an innovative approach by integrating three Convolutional Block Attention Modules (CBAM) into YOLOv5 and utilizing the combined dataset of original images and orthomosaic, which exhibited the highest accuracy performance in Konjac plants recognition (Precision = 94.3 %, Recall = 96.0 %, F1-Score = 95.1 %). Our findings illustrate that the orthomosaic generated from original images acquired via UAV outperformed individual original images in terms of accuracy for counting Konjac plants across expansive areas. This study provides new insight into the recognition and counting of various crop plants across large-scale regions, presenting a practical and efficient approach.
Why it matches plant phenotyping methodsUAV画像とオルソモザイクを用いた植物個体数の検出・計数手法を開発・比較し、YOLOv5改良モデルの性能を評価しているため、植物表現型取得が中心です。
abstractThis study systematically evaluated the performance of UAV-based original images, the generated orthomosaic, and the combination of both for the detection and counting of the Konjac plant.
Early crop yield prediction is a major challenge in precision agriculture, and efficient and rapid yield prediction is highly important for sustainable fruit production. The accurate detection of major fruit characteristics, including flowering, green fruiting, and ripening stages, is crucial for early yield estimation. Currently, most crop yield estimation studies based on the YOLO model are only conducted during a single stage of maturity. Combining multi-growth period data for crop analysis is of great significance for crop growth detection and early yield estimation. In this study, a new network model, YOLOv8-RL, was proposed using citrus multigrowth period characteristics as a data source. A citrus yield estimation model was constructed and validated by combining network identification counts with manual field counts. Compared with YOLOv8, the number of parameters of the improved network is reduced by 50.7%, the number of floating-point operations is decreased by 49.4%, and the size of the model is only 3.2 MB. In the test set, the average recognition rate of citrus flowers, green fruits, and orange fruits was 95.6%, the mAP@.5 was 94.6%, the FPS value was 123.1, and the inference time was only 2.3 milliseconds. This provides a reference for the design of lightweight networks and offers the possibility of deployment on embedded devices with limited computational resources. The two estimation models constructed on the basis of the new network had coefficients of determination R 2 values of 0.91992 and 0.95639, respectively, with a prediction error rate of 6.96% for citrus green fruits and an average error rate of 3.71% for orange fruits. Compared with network counting, the yield estimation model had a low error rate and high accuracy, which provided a theoretical basis and technical support for the early prediction of fruit yield in complex environments.
Why it matches plant phenotyping methods柑橘の花・果実を画像認識して収量を推定するYOLOv8改良モデルとワークフローを開発・検証しており、植物形質取得法が中心的です。
abstractIn this study, a new network model, YOLOv8-RL, was proposed using citrus multigrowth period characteristics as a data source.
Legume root nodules are important for biological nitrogen fixation, a process critical for plants to gain additional nitrogen from the environment. Nodule quantification is valuable for evaluating nitrogen fixation efficiency, assessing symbiotic relationships, monitoring responses to nitrogen, and supporting genetic studies on legume adaptation and productivity. However, accurate quantification of root nodules is difficult and time-consuming due to the complexity of the root system and soil interference. Here, we explore the utility of hyperspectral imaging as a non-destructive tool to detect active fixing root nodules with minimal preparation and show that we can differentiate nodules and root tissues through unique spectral signatures while also distinguishing between fixing and non-fixing nodules. We applied deep learning techniques to develop an automated nodule counting pipeline adaptable across different legume species and under diverse growth conditions. This approach eliminates the need for labor-intensive counting and enables the detection of nodules embedded within dense root tangles with high accuracy. This automated hyperspectral approach offers a promising alternative to support assessments of nodule abundance and their activity across legume species grown under various environments.
Why it matches plant phenotyping methods根粒の検出・計数と固定活性の識別という植物形質の取得を、ハイパースペクトル画像と深層学習による自動化手法として開発しており、方法が研究の中心です。
abstractHere, we explore the utility of hyperspectral imaging as a non-destructive tool to detect active fixing root nodules
Trait-based breeding has been shown to enhance and sustain yield potential of various crops under current and future changing climate. To be successful, trait-based breeding requires extensive phenotyping of plants in large-scale field trials that may include hundreds of genotypes. Computer vision approaches have been used extensively for image-based high-throughput phenotyping of diverse traits. However, studies focused on estimating grain count, a trait that directly influences the overall yield, are limited due to a lack of benchmark grain image datasets. In this work, we focus on grain count estimation in sorghum, a crop that holds immense significance for both food and energy production. We introduce a large Sorghum-Grain-Count (SGC) dataset consisting of (front and back, with and without flash) images of 1264 panicles from 316 genotypes (i.e., 4 panicles per genotype), for a total of approximately 5000 images, as well as approximately 12,500 images containing the corresponding threshed grains, together with machine counts per panicle. To develop baseline models, we manually annotated grains in images from 100 genotypes using bounding boxes. We used the manually annotated images to train baseline models for small object detection and counting. We also trained regression-based models for grain count estimation. The best overall model for count estimation from panicle images was a regression model, which achieved a mean absolute percent error of 29.97 and an R 2 value of 0.75. We make our dataset and baselines publicly available to facilitate further research on grain count estimation in sorghum and other crops. • We curated the Sorghum-Grain-Count dataset, which includes ∼17,500 panicle and threshed grain images covering 316 genotypes. • This is the largest dataset for grain count estimation and can help advance research on small object count estimation. • We trained strong baseline object detection models, as well as regression-based models for grain count estimation. • Our best model for grain count estimation from panicle images was a regression model that achieved an R 2 value of 0.75. • Our models provide a foundation for non-destructive yield estimation tools, which are greatly needed by breeding programs.
Why it matches plant phenotyping methodsソルガムの穂・粒画像から粒数という植物収量関連形質を推定する大規模データセット、ベンチマーク、検出・回帰モデルを開発・公開しており、表現型取得・推定手法が研究の中心です。
abstractWe introduce a large Sorghum-Grain-Count (SGC) dataset consisting of (front and back, with and without flash) images of 1264 panicles from 316 genotypes
The aim of this study was to use remote sensing with a drone equipped with a multispectral camera to take a stand survey of maize after the phenological stage of emergence, and to count the number of emerged plants and determine its accuracy. Our investigations were carried out at the University of Debrecen, Látókép Production Experimental Station in a sowing date long-term experiment. In the 2024 growing season, Sowing Date I was on 4 April and Sowing Date II on 12 April. The same maize hybrids with 8-8 different genotypes were used for each sowing date. There is a strong correlation between number of plants/plot and number of plants/rowx2 for the two plant density measurements presented in this paper, with an r value of 0.977*** (p
Why it matches plant phenotyping methodsドローン搭載マルチスペクトルカメラによる出芽個体数・作物密度の自動測定と精度評価が研究の中心であり、植物表現型の取得手法を扱っている。
abstractto count the number of emerged plants and determine its accuracy
Pod numbers are important for assessing soybean yield. How to simplify the traditional manual process and determine the pod number phenotype of soybean maturity more quickly and accurately is an urgent challenge for breeders. With the development of smart agriculture, numerous scientists have explored the phenotypic information related to soybean pod number and proposed corresponding methods. However, these methods mainly focus on the total number of pods, ignoring the differences between different pod types and do not consider the time-consuming and labor-intensive problem of picking pods from the whole plant. In this study, a deep learning approach was used to directly detect the number of different types of pods on non-disassembled plants at the maturity stage of soybean. Subsequently, the number of pods wascorrected by means of a metric learning method, thereby improving the accuracy of counting different types of pods. After 200 epochs, the recognition results of various object detection algorithms were compared to obtain the optimal model. Among the algorithms, YOLOX exhibited the highest mean average precision (mAP) of 83.43% in accurately determining the counts of diverse pod categories within soybean plants. By improving the Siamese Network in metric learning, the optimal Siamese Network model was obtained. SE-ResNet50 was used as the feature extraction network, and its accuracy on the test set reached 93.7%. Through the Siamese Network model, the results of object detection were further corrected and counted. The correlation coefficients between the number of one-seed pods, the number of two-seed pods, the number of three-seed pods, the number of four-seed pods and the total number of pods extracted by the algorithm and the manual measurement results were 92.62%, 95.17%, 96.90%, 94.93%, 96.64%,respectively. Compared with the object detection algorithm, the recognition of soybean mature pods was greatly improved, evolving into a high-throughput and universally applicable method. The described results show that the proposed method is a robust measurement and counting algorithm, which can reduce labor intensity, improve efficiency and accelerate the process of soybean breeding.
Why it matches plant phenotyping methods大豆莢数という植物形質を対象に、非解体植物画像から莢の分類・計数を行う深層学習およびメトリックラーニング手法を開発・検証しており、フェノタイピング手法が中心である。
abstracta deep learning approach was used to directly detect the number of different types of pods on non-disassembled plants at the maturity stage of soybean
Accurate quantification of open bolls and their distribution is crucial for understanding cotton growth, development, and yield in optimized crop management and enhanced plant breeding. Manual boll counting methods are time-consuming, labor-intensive, and subjective. Leveraging the potential of high-resolution images for high-throughput phenotyping offers a promising avenue for efficient trait quantification. The objectives of this study were to develop methods to detect and count open cotton bolls using LiDAR point cloud and RGB images and to compare the effectiveness of these two data sources. A DJI Phantom 4 RTK Unmanned Aerial System (UAS) equipped with a 4 K RGB camera was used to acquire high-resolution RGB images, and a DJI Matrice 300 RTK with a Zenmuse L1 sensor was used to acquire LiDAR point cloud data. The RGB images were converted to point cloud using photogrammetry by measuring multiple points of overlapping images. The boll detection workflow involved data filtering and clustering using the density-based spatial clustering of applications with noise (DBSCAN) method. Evaluation of the methods involved 48 plots representing small, medium, and large plant sizes using metrics including mean absolute percentage error (MAPE), root mean square error (RMSE), and coefficient of determination (r²). The methods using both data sources performed well in estimating open bolls, with LiDAR point cloud data slightly outperforming those derived from RGB images. Generally, the performance of the DBSCAN method in boll detection improved with decreasing plant sizes. Specifically, LiDAR data yielded MAPE values of 5.03 %, 8.05 %, and 13.46 %, RMSE values of 7.26, 14.33, and 23.40 bolls per m², and r 2 values of 0.93, 0.84, and 0.84 for small, medium, and large plant sizes, respectively. RGB image-based data exhibited MAPE values of 7.21 %, 6.49 %, and 16.41 %, RMSE values of 11.05, 13.66, and 26.49 bolls per m², and r 2 values of 0.82, 0.74, and 0.83 for small, medium, and large plant sizes, respectively. The method demonstrates the potential of RGB imagery and LiDAR data for estimating boll counts, offering valuable tools for enhanced plant phenotyping in plant breeding and site-specific crop management. Both data sources underestimated boll counts, with smaller plants showing less undercounting, likely due to improved light penetration and separation of bolls. These findings highlight the influence of plant structure on boll detection accuracy and the need to address challenges posed by dense canopies to enhance detection reliability. • The study developed methods to count open cotton bolls using LiDAR point clouds and RGB images. • LiDAR slightly outperformed the RGB image-derived point cloud, with better accuracy for smaller plants due to less canopy density. • Dense canopies reduced detection accuracy, highlighting the influence of plant structure.
Why it matches plant phenotyping methodsLiDARとRGB画像を用いて綿花の開花ボール数を検出・計数する手法を開発し、精度比較・検証しており、植物表現型取得が研究の中心です。
abstractThe objectives of this study were to develop methods to detect and count open cotton bolls using LiDAR point cloud and RGB images and to compare the effectiveness of these two data sources.
Due to the limitations of the sowing machine performance and rice seed germination rates, missing seedlings inevitably occur after rice is sown in large fields. This phenomenon has a direct impact on the rice yield. In the field environment, the existing methods for detecting missing seedlings based on unmanned aerial vehicle (UAV) remote sensing images often have unsatisfactory effects. Therefore, to enable the fast and accurate detection of missing rice seedlings and facilitate subsequent reseeding, this study proposes a UAV remote-sensing-based method for detecting missing rice seedlings in large fields. The proposed method uses an improved PCERT-DETR model to detect rice seedlings and missing seedlings in UAV remote sensing images of large fields. The experimental results show that PCERT-DETR achieves an optimal performance on the self-constructed dataset, with an mean average precision (mAP) of 81.2%, precision (P) of 82.8%, recall (R) of 78.3%, and F 1 -score (F 1 ) of 80.5%. The model's parameter count is only 21.4 M and its FLOPs reach 66.6 G, meeting real-time detection requirements. Compared to the baseline network models, PCERT-DETR improves the P, R, F 1 , and mAP by 15.0, 1.2, 8.5, and 6.8 percentage points, respectively. Furthermore, the performance evaluation experiments were carried out through ablation experiments, comparative detection model experiments and heat map visualization analysis, indicating that the model has a strong detection performance on the test set. The results confirm that the proposed model can accurately detect the number of missing rice seedlings. This study provides accurate information on the number of missing seedlings for subsequent reseeding operations, thus contributing to the improvement of precision farming practices.
Why it matches plant phenotyping methodsUAV画像からイネの欠株数を推定するPCERT-DETR手法の開発と検証が研究の中心であり、植物状態の表現型を直接評価している。
abstractThe proposed method uses an improved PCERT-DETR model to detect rice seedlings and missing seedlings in UAV remote sensing images of large fields.
This study presents a novel and efficient approach to accurately assess post-sowing rice plant density by leveraging unmanned aerial vehicles (UAVs) equipped with high-resolution RGB cameras. In contrast to labor-intensive and spatially limited traditional methods that rely on manual sampling and extrapolation, our proposed methodology uses UAVs to rapidly and comprehensively survey entire paddy fields at optimized altitudes (4, 6, 8, and 10 m). Aerial imagery was autonomously acquired 17 days post-sowing, following a pre-defined flight path. The robust rice plant density estimation process incorporates two key innovations: first, a dynamic system of 12 adaptive segmentation thresholding blocks that effectively detects rice seed presence across diverse and variable background conditions. Second, a tailored three-layer convolutional neural network (CNN) accurately classifies vegetative situations. To maximize the training efficiency and performance, we implemented both a pretrained model and a deep learning model, conducting a rigorous comparative analysis against the state-of-the-art YOLOv10. Notably, under favorable imaging conditions, our findings indicate that a 6-m flight altitude yields optimal results, achieving a high degree of accuracy with rice plant density estimates that closely align with those obtained through traditional ground-based methods. This investigation unequivocally highlights the significant advantages of UAV-based monitoring as an economically viable, spatially comprehensive, and demonstrably accurate tool for precise rice field management, ultimately contributing to enhanced crop yields, improved food security, and the promotion of sustainable agricultural practices.
Why it matches plant phenotyping methodsUAV RGB画像と適応的セグメンテーション、CNNを用いてイネ個体密度を推定する手法を開発・比較・検証しており、植物形質の取得が研究の中心です。
abstractThis study presents a novel and efficient approach to accurately assess post-sowing rice plant density by leveraging unmanned aerial vehicles (UAVs) equipped with high-resolution RGB cameras.
Reproduction assets foundThe paper's data availability statement points to a public Zenodo deposit containing the study's datasets (UAV RGB imagery/labels used for rice plant density estimation). No separate author analysis code repository is stated.Dataset · publicvaluate the accuracy of the proposed labels, subsequently enhancing the training model's speed, convergence, accuracy, and efficiency.
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://zenodo.org/records/10960906 .
Author contributions
TH: Writing – original draft. TN: Data curation, Resources, Validation, Writing – original draft. QN: Data curation, Writing – review & editing. HN: Funding acquisition, Investigation, Methodology, Writing – review & editing. PP: Methodology, Software, Supervision, Writing – review & editing.
Funding
TheOpen asset ↗zenodo · 10960906lines:500-523Plant phenotyping relevance match · UnverifiedOpenAlex · checked 15 Sept 2026
This review paper focuses on the digital transformation of yield prediction for the sustainable development of the Jeju citrus industry, emphasizing current applications and future challenges of image analysis technologies. Accurate yield prediction is essential for stabilizing farm income, improving distribution efficiency, balancing supply and demand, and optimizing cultivation strategies. However, traditional statistics-based approaches are limited by climate change, cultivation area fluctuations, and labor shortages. In this context, Al-driven digital technologies —especially non-invasive image analysis —have emerged as promising alternatives. The paper provides an in-depth overview of image analysis applications in two key areas: fruit detection and counting, and fruit size and growth prediction. Notably, deep learning-based object detection models (e.g., YOLO, Faster R-CNN) and 3D reconstruction technologies have improved prediction accuracy. Integrating auxiliary data, such as maturity and quality indicators, is also discussed. Despite these advancements, challenges remain for real-world implementation. These include data collection under varied environments, model robustness (especially against occlusion), and the need for real-time processing and user-friendly system design. Future research should prioritize integrating heterogeneous data — including weather and soil — long-term time-series learning, and developing cost-effective, high-efficiency solutions. These efforts are expected to enhance the accuracy and reliability of citrus yield predictions, driving the digital transformation and sustainable future of the Jeju citrus industry.
Why it matches plant phenotyping methods画像解析による柑橘果実の検出・計数、サイズ・成長推定を中心にレビューしており、植物形質取得手法が主要内容である。
abstractThe paper provides an in-depth overview of image analysis applications in two key areas: fruit detection and counting, and fruit size and growth prediction.
To enhance the operational performance of green manure sowing, seed harvesting, and processing equipment, as well as to obtain the physical characteristic parameters required for equipment simulation analysis, it is necessary to carry out the seed physical properties such as dimensions, bulk density, and frictional characteristics. This study focuses on the main cultivated varieties of green manure in China—milk vetch, hairy vetch, and sesbania—and innovatively proposes three low-cost, convenient, and high-precision experimental methods of geometric size for small-sized seeds. The measured results show that geometric dimensions of the seeds for milk vetch were 2.60mm-3.40mm in length, 1.90mm-2.40mm in width and 0.71mm-0.97mm in height, the pod dimensions of hairy vetch were 23.03mm-32.83mm in length, 7.39mm-9.74mm in width, 4.06mm-6.15mm in height, and seed diameters were 3.04mm-4.10mm, the seed dimensions of sesbania were 3.81mm-4.29mm in length, 1.98mm-2.37mm in width, 1.85mm-2.08mm in height. Using Python programming language and open-source AI algorithms, we innovatively developed an image-based seed counting experimental method. This cost-effective approach requires no specialized instruments while significantly saving time and labor. Based on the counting results, the thousand-seed weights of milk vetch, hairy vetch, and sesbania seeds were determined to be 3.40g, 26.50g and 15.58g, with respective seed moisture contents of 7.18%, 9.81% and 8.73% at the time of measurement. The instruments and methods for measuring bulk density and angle of repose of green manure seeds were defined. The bulk densities of milk vetch, hairy vetch, and sesbania seeds were determined to be 732 g/L, 761 g/L and 845 g/L, by using the electronic grain densitometer. Meanwhile, the angles of repose of milk vetch, hairy vetch, and sesbania seeds were determined to be 31.66°, 28.15°and 29.82°, by using the angle of repose tester. Based on the principle that the indexing head can accurately determine the angle, the sliding friction angles of milk vetch, hairy vetch, and sesbania seeds were determined to be 25.85°, 23.55°and 24.03°, by using the di-viding head. The testing methods developed in this study, along with the results of measuring the physical properties of green manure seeds, provide valuable reference data for the measurement of the physical properties of small-seeded crop (e.g., green manures) as well as other small or irregularly-shaped particles.
Why it matches plant phenotyping methods種子の幾何形状・個数という植物器官形質の取得法を中心に、低コストな測定法と画像ベース計数法を開発しているため。
abstractinnovatively proposes three low-cost, convenient, and high-precision experimental methods of geometric size for small-sized seeds
Restoration and conservation of native plant populations will benefit from identifying individual plants with high reproductive success. While high-fecundity plants are ideal for seed sourcing, locating these plants across heterogeneous landscapes presents a logistical challenge. This challenge is especially significant for big sagebrush (Artemisia tridentata), a foundational species that is the focus of large-scale seed collection for restoration efforts in western rangelands. We evaluated whether cost-effective RGB imagery from unoccupied aerial vehicles (UAVs) could map flower stalk production in big sagebrush plants. Models were trained using three years of data from four sites spanning an elevational gradient that included all three big sagebrush subspecies: A. t. wyomingensis, A. t. vaseyana, and A. t. tridentata. Our model predicted flower stalk production from UAV imagery with a Mean Absolute Error (MAE) of [~]100 stalks, which is relatively low given that some plants produced more than 700 stalks. A hurdle model that explicitly accounted for excess zeroes outperformed simpler negative binomial models, suggesting that reproductive failure is distinct from flower stalk production in reproductive plants. Structural metrics, including height differences between June and September, canopy height, and edge-to-area ratio of plant crowns, had stronger effects in our model for counts of flower stalk production than spectral data. Model performance was consistent across environmentally heterogeneous sites but declined when applied to years excluded from training, indicating that year-specific training data may be necessary for interannual predictions. These results demonstrate that UAVs can monitor reproductive potential in wild plants and help identify high-fecundity individuals for seed collection. Our work underscores the need for future research that can improve predictions of flower production, including integrating multispectral data and increasing model reliability across years to support climate-resilient restoration strategies.
Why it matches plant phenotyping methodsUAV RGB画像から個体の花茎生産数を推定するモデルを開発・評価しており、植物の繁殖形質取得が研究の中心です。
abstractWe evaluated whether cost-effective RGB imagery from unoccupied aerial vehicles (UAVs) could map flower stalk production in big sagebrush plants.
Currently, rice transplanters are extensively employed for the mechanized cultivation of rice seedlings. However, few technologies or systems are available to monitor the operational quality parameters, i.e., the number of missing seedlings, row spacing, plant distance, etc., of rice transplanters. The performance of rice transplanters is directly linked to the growth quality of the seedlings and has a crucial effect on the final yield. Therefore, monitoring the various issues that arise during the operation of rice transplanters in a timely and accurate manner to ensure the quality of the transplanting process is particularly important. To address the above issues, this paper develops a real-time monitoring system for rice transplanters. The system architecture includes embedded devices, an image capture module, and a data upload module. A rice seedling detection model based on an enhanced YOLOv5-Lite neural network is developed, and comparative experimental results demonstrate that the proposed model achieves an mAP@0.5 of 81.9 % for rice seedling detection, which is higher than that of the original YOLOv5-Lite model. We additionally propose a RANSAC-based algorithm to detect rice seeding paths in real time, and the rice seeding path detection results are used to determine the row spacing and plant distance. Specifically, a distance mapping algorithm based on triangular transformations is developed to calculate the row spacing and plant distance in a field. We subsequently calculate the number of missing seedlings between adjacent plants on the basis of the spacing between plants in the same row. Furthermore, a rice seedling tracking and counting algorithm based on an improved ByteTrack algorithm is developed to determine the missing seedling rate, as well as the seeding quantity. We integrate the developed algorithms into a real-time monitoring system and test them at Qixing Farm. The experimental results indicate that the monitoring system achieves an accuracy of 99.2 % for seedling quantity counting and an accuracy of 90.3 % for missing rate counting, with a processing speed of 3.95 frames per second.
Why it matches plant phenotyping methodsイネ苗の検出・追跡・計数、欠株率、条間および株間距離を画像から推定するリアルタイムシステムを開発・評価しており、植物形質取得が研究の中心である。
abstractthis paper develops a real-time monitoring system for rice transplanters
The number of grains of a cereal plant characterizes its yield, while grain size and shape are closely related to its weight. To estimate the number of grains, their shape and size, digital image analysis is now generally used. The grains in such images may be completely separated, touching or densely packed. In the first case, the simplest binarization/segmentation algorithms, such as the watershed algorithm, can achieve high accuracy in segmentation and counting grains in an image. However, in the case of touching grains, simple machine vision algorithms may lead to inaccuracies in determining the contours of individual grains. Therefore, methods for accurately determining the contours of individual grains when they are in contact are relevant. One approach is based on the search for pixels of the grain contact area, in particular, by identification of concave points on the grain contour boundary. However, some grains may have chips, depressions and bulges, which leads to the identification of the corner points that do not correspond to the grain contact region. Additional data processing is required to avoid these errors. In this paper, we propose an algorithm for the identification of wheat grains in an image and determine their boundaries in the case when they are touching. The algorithm is based on using a modification of the concave point search algorithm and utilizes a method of assigning contour boundary pixels to a single grain based on approximation of grain contours by ellipses. We have shown that the proposed algorithm can identify grains in the image more accurately compared to the algorithm without such approximation and the watershed algorithm. However, the time cost for such an algorithm is significant and grows rapidly with increasing number of grains and contours including multiple grains.
Why it matches plant phenotyping methods小麦粒画像から接触粒を分離・計数し、粒形状とサイズを推定する画像解析アルゴリズムの開発・比較が中心であり、植物形質取得手法に該当する。
abstractIn this paper, we propose an algorithm for the identification of wheat grains in an image and determine their boundaries in the case when they are touching.
Abstract Urban green spaces (UGS) provide various ecological, cultural, aesthetic, and psychological functions contributing to public health and well-being. To ensure proper management and the optimal performance of all these functions, it is essential to closely monitor their structure: number of trees, species composition, tree size, health status, location and surrounding objects. Traditionally, parameters are measured using conventional methods: calipers, altimeters, and diameter tapes. However, while modern 3D data collection technologies such as terrestrial and mobile laser scanners or photogrammetry have been employed for monitoring UGS, their use is often challenging. Therefore, we aimed to test and create a methodology using a novel device (Pix4D & Emlid Scanning Kit), which uniquely combines photogrammetry and light detection and ranging with real-time kinematics in a smartphone. Since the solution is smartphone-based, it provides relatively low-cost employability. The research was conducted in Adolf Priesol Park in Zvolen (Slovakia) over an area of approximately 6,000 m 2 . We focused on the device’s positional accuracy in measuring footpaths, park amenities, and trees, as well as its tree detection rate and accuracy in determining tree diameter. The results demonstrated that the device exhibited high positional accuracy (horizontal RMSE = 0.08 m, vertical RMSE = 0.07 m), a 100% tree detection rate, and exceptional diameter at the breast height accuracy, with an RMSE of 1.1 cm in the area-based approach and 0.42 cm in the individual approach. Notably, the 6,000 m 2 area was covered in 80 minutes, including collecting almost 85 trees, all footpaths and all park amenities.
Why it matches plant phenotyping methodsスマートフォン搭載の写真測量・LiDAR・RTKを組み合わせた樹木計測手法を開発・検証し、樹木検出率と胸高直径の精度を評価しているため、植物形質取得法が研究の中心である。
abstractwe aimed to test and create a methodology using a novel device (Pix4D & Emlid Scanning Kit), which uniquely combines photogrammetry and light detection and ranging with real-time kinematics in a smartphone.
Efficient and accurate acquisition of tree distribution and three-dimensional geometric information in forest scenes, along with three-dimensional reconstructions of entire forest environments, hold significant application value in precision forestry and forestry digital twins. However, due to complex vegetation structures, fine geometric details, and severe occlusions in forest environments, existing methods—whether vision-based or LiDAR-based—still face challenges such as high data acquisition costs, feature extraction difficulties, and limited reconstruction accuracy. This study focuses on reconstructing tree distribution and extracting key individual tree parameters, and it proposes a forest 3D reconstruction framework based on high-resolution remote sensing images. Firstly, an optimized Mask R-CNN model was employed to segment individual tree crowns and extract distribution information. Then, a Tree Parameter and Reconstruction Network (TPRN) was constructed to directly estimate key structural parameters (height, DBH etc.) from crown images and generate tree 3D models. Subsequently, the 3D forest scene could be reconstructed by combining the distribution information and tree 3D models. In addition, to address the data scarcity, a hybrid training strategy integrating virtual and real data was proposed for crown segmentation and individual tree parameter estimation. Experimental results demonstrated that the proposed method could reconstruct an entire forest scene within seconds while accurately preserving tree distribution and individual tree attributes. In two real-world plots, the tree counting accuracy exceeded 90%, with an average tree localization error under 0.2 m. The TPRN achieved parameter extraction accuracies of 92.7% and 96% for tree height, and 95.4% and 94.1% for DBH. Furthermore, the generated individual tree models achieved average Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) scores of 11.24 and 0.53, respectively, validating the quality of the reconstruction. This approach enables fast and effective large-scale forest scene reconstruction using only a single remote sensing image as input, demonstrating significant potential for applications in both dynamic forest resource monitoring and forestry-oriented digital twin systems.
Why it matches plant phenotyping methods樹冠画像から個体樹の樹高・DBHなどの形態形質を抽出し、3D再構成する手法を開発・検証しており、植物形質取得が研究の中心である。
abstracta Tree Parameter and Reconstruction Network (TPRN) was constructed to directly estimate key structural parameters (height, DBH etc.) from crown images and generate tree 3D models.
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:1-52Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Abstract Background To address the challenge of real-time plant monitoring in greenhouse environments, this industry-driven research focuses on developing an autonomous quadrotor UAV system specifically designed for monitoring strawberry plants. Traditional methods for greenhouse monitoring are labor-intensive and lack scalability, particularly in precision agriculture applications. Method The study begins by proposing the mature strawberry detection model for greenhouse environment. The YOLOv9 with GLEAN advantage is proposed to detect small mature strawberries via on board camera on the quadrotor. Also the hybrid trajectory tracking controller for quadrotor is proposed and validated in both simulation and real time environment. The UAV follows predefined way points for navigation in the greenhouse environment. An onboard vision system is integrated, employing a novel YOLOv9-GLEAN-based algorithm for online and offline mature strawberry detection and counting. Results The YOLOv9-GLEAN model achieves high detection accuracy, as confirmed by evaluation metrics such as precision, recall, and F1-score. The proposed hybrid (PID+LQR) controller demonstrates superior tracking performance compared to other conventional controllers. The integrated control and perception system proves effective in both simulated and real-world greenhouse environments. Discussion The research validates the efficacy of deep learning models, with YOLOv9-GLEAN showing exceptional performance in enabling rapid, precise, and automated detection of ripe strawberries through quadrotor deployment in greenhouse environments. Such agricultural monitoring technologies represent a substantial advancement beyond conventional manual inspection approaches, empowering farmers and greenhouse operators to execute well-informed, time-sensitive management decisions that minimize crop losses and optimize production yields. This investigation underscores the revolutionary impact that deep learning technologies can have within greenhouse agriculture.
Why it matches plant phenotyping methods温室イチゴの成熟果実を画像認識で検出・計数する手法とUAV搭載システムが研究の中心であり、果実の成熟状態・数量という植物器官の形質を抽出しているため含める。
abstractThe YOLOv9 with GLEAN advantage is proposed to detect small mature strawberries via on board camera on the quadrotor.
Background Transgenic plants are essential for both basic and applied plant biology. Recently, fluorescent and colorimetric markers were developed to enable nondestructive identification of transformed seeds and accelerate the generation of transgenic plant lines. Yet, transformation often results in the integration of multiple copies of transgenes in the plant genome. Multiple transgene copies can lead to transgene silencing and complicate the analysis of transgenic plants by requiring researcher to track multiple T-DNA loci in future generations. Thus, to simplify analysis of transgenic lines, plant researchers typically screen transformed plants for lines where the T-DNA inserted in a single locus - an analysis that involves laborious manual counting of fluorescent and non-fluorescent seeds for screenable markers. Results To expedite T-DNA segregation analysis, we developed SeedSeg, an image analysis tool that uses a segmentation algorithm to count the number of transformed and wild-type seeds in an image. SeedSeg runs a chi-squared test to determine the number of T-DNA loci. Parameters can be adjusted to optimize for different brightness intensities and seed sizes. Conclusions By automating the seed counting process, SeedSeg reduces the manual labor associated with identifying transgenic lines containing a single T-DNA locus. SeedSeg is adaptable to different seed sizes and visual transgene markers, making it a versatile tool for accelerating plant research.
Why it matches plant phenotyping methods種子画像から形質転換種子数と野生型種子数を自動抽出する画像解析ツールの開発であり、植物表現型（種子状態）の取得・定量が中心です。
abstractwe developed SeedSeg, an image analysis tool that uses a segmentation algorithm to count the number of transformed and wild-type seeds in an image.
The morphological structure of wheat spikes plays a central role in wheat yield. Wheat spike morphology, closely associated with crop yield, has attracted considerable attention in the fields of genetics and breeding. However, traditional measurement methods can only measure simple traits, and precise phenotypes remain difficult to obtain, constraining the study and improvement of complex spike-related traits. This study utilized deep learning technologies to develop a pipeline, called SpikePheno, for the acquisition of precise wheat spike phenotypes. Our pipeline demonstrated high accuracy in spike segmentation, achieving a mean Intersection over Union (mIoU) of 0.948. Additionally, our method accurately identified spikelet counts, achieving an R 2 of 0.9923. Using experimental data of 221 wheat cultivars from various regions of China grown in Zhao County, Hebei Province, our pipeline extracted 45 different phenotypes and studied their correlations with thousand grain weight (TGW) and spike yield. Our findings indicate that precise measurement of spike area, spikelet area, and other phenotypic traits enables a clearer understanding of the correlation between spike morphology and wheat yield. Through hierarchical clustering based on spike morphology, we categorized wheat spikes into six classes and identified phenotypic differences between these classes and their impact on TGW and yield. Furthermore, this study revealed phenotypic differences between wheat cultivars from different geographical regions and over different decades, with an increase in large-spike cultivars over time, especially in southern China. This research may help breeders understand the relationship between wheat spike morphology and yield, providing an important basis for future wheat breeding efforts.
Why it matches plant phenotyping methodsSpikePhenoという深層学習画像解析パイプラインを開発し、コムギ穂の分割・小穂数同定・複数形質抽出を中心に評価しているため、植物フェノタイピング手法論文に該当する。
abstractThis study utilized deep learning technologies to develop a pipeline, called SpikePheno, for the acquisition of precise wheat spike phenotypes.
This study explores the development and evaluation of high-throughput 3D plant reconstruction and 2D feature detection and segmentation methods for plant phenotyping. A robotic system was employed to collect datasets of individual cucumber plants, utilizing automated mechanisms for efficient, high-throughput data acquisition. Three types of 3D reconstruction methods called Instant-NGP, Nerfacto, and 3D Gaussian Splatting were adopted and compared in terms of rendering quality and speed.Among them, 3D Gaussian Splatting performed the best, achieving PSNR: 25, SSIM: 0.84, LPIPS: 0.20, and also an impressive rendering speed of 6.39 FPS. Novel viewpoint renderings and depth maps further demonstrated its ability to generate accurate and photo-realistic representations of plants. Additionally, rendered images were utilized for training YOLO models to segment plant features into two classes: leaf and fruit. The YOLOv11s model achieved the highest F1 Score (0.932), balancing speed and accuracy. Ultra-view renderings and segmentation provided valuable insights into plant morphology, including leaf and fruit counts, paving the way for scalable, automated phenotyping applications. This study highlights the potential of integrating 3D Gaussian Splatting with advanced segmentation models for precise and efficient plant phenotyping.
Why it matches plant phenotyping methods植物の高スループット3D再構成、画像セグメンテーション、ロボット計測を開発・比較評価し、葉・果実数や形態の抽出に用いており、植物フェノタイピング手法が中心である。
abstractThis study explores the development and evaluation of high-throughput 3D plant reconstruction and 2D feature detection and segmentation methods for plant phenotyping.
Rice panicle detection is a key technology for improving rice yield and agricultural management levels. Traditional manual counting methods are labor-intensive and inefficient, making them unsuitable for large-scale farmlands. This paper proposes FRPNet, a novel lightweight convolutional neural network optimized for multi-altitude rice panicle detection in UAV images. The architecture integrates three core innovations: a CSP-ScConv backbone with self-calibrating convolutions for efficient multi-scale feature extraction; a Feature Pyramid Shared Convolution (FPSC) module that replaces pooling with multi-branch dilated convolutions to preserve fine-grained spatial information; and a Dynamic Bidirectional Feature Pyramid Network (DynamicBiFPN) employing input-adaptive kernels to optimize cross-scale feature fusion. The model was trained and evaluated on the open-access Dense Rice Panicle Detection (DRPD) dataset, which comprises UAV images captured at 7 m, 12 m, and 20 m altitudes. Experimental results demonstrate that our method significantly outperforms existing advanced models, achieving an AP50 of 0.8931 and an F2 score of 0.8377 on the test set. While ensuring model accuracy, the parameters of the proposed model decreased by 42.87% and the GFLOPs by 48.95% compared to Panicle-AI. Grad-CAM visualizations reveal that FRPNet exhibits superior background noise suppression in 20 m altitude images compared to mainstream models. This work establishes an accuracy-efficiency balanced solution for UAV-based field phenotyping.
Why it matches plant phenotyping methodsUAV画像からイネ穂の検出・計数という植物形質を抽出する軽量深層学習手法を開発し、公開データセットで評価しており、表現型取得手法が中心である。
abstractThis paper proposes FRPNet, a novel lightweight convolutional neural network optimized for multi-altitude rice panicle detection in UAV images.
Reproduction assets foundThe paper states its data (the DRPD rice panicle UAV dataset used for training/evaluation) is publicly available via a GitHub release URL, which matches an allowed URL.Dataset · publicData Availability Statement: Data is available at https://github.com/changcaiyang/Panicle-AI/
releases/tag/v1.0.Open asset ↗Panicle-AI · v1.0pdf-page:22 lines:1-59Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 6 Sept 2026
BACKGROUND: Plant phenotyping has become increasingly important for advancing plant science, agriculture, and biotechnology. Classic manual methods are labor-intensive and time-consuming, while existing computational tools often require advanced coding skills, high-performance hardware, or PC-based environments, making them inaccessible to non-experts, to resource-constrained users, and to field technicians. RESULTS: To respond to these challenges, we introduce OpenPheno, an open-access, user-friendly, and smartphone-based platform encapsulated within a WeChat Mini-Program for instant plant phenotyping. The platform is designed for ease of use, enabling users to phenotype plant traits quickly and efficiently with only a smartphone at hand. We currently instantiate the use of the platform with tools such as SeedPheno, WheatHeadPheno, LeafAnglePheno, SpikeletPheno, CanopyPheno, TomatoPheno, and CornPheno; each offering specific functionalities such as seed size and count analysis, wheat head detection, leaf angle measurement, spikelet counting, canopy structure analysis, and tomato fruit measurement. In particular, OpenPheno allows developers to contribute new algorithmic tools, further expanding its capabilities to continuously facilitate the plant phenotyping community. CONCLUSIONS: By leveraging cloud computing and a widely accessible interface, OpenPheno democratizes plant phenotyping, making advanced tools available to a broader audience, including plant scientists, breeders, and even amateurs. It can function as a role in AI-driven breeding by providing the necessary data for genotype-phenotype analysis, thereby accelerating breeding programs. Its integration with smartphones also positions OpenPheno as a powerful tool in the growing field of mobile-based agricultural technologies, paving the way for more efficient, scalable, and accessible agricultural research and breeding.
Why it matches plant phenotyping methodsスマートフォンで植物形質を取得・解析するソフトウェアプラットフォームの開発が中心であり、複数の具体的な表現型解析ツールを提供している。
abstractwe introduce OpenPheno, an open-access, user-friendly, and smartphone-based platform encapsulated within a WeChat Mini-Program for instant plant phenotyping.
Reproduction assets foundThe paper's authors publicly release the OpenPheno platform code (GitHub repository) and the evaluation sample data used for algorithm validation and demonstration (dataset subdirectory). Both are paper-specific, public, and actionable.Dataset · publicEvaluation sample data used for algorithm validation and demonstration has been made publicly available at out GitHub repository: https://github.com/openpheno/OpenPheno/tree/main/dataset .Open asset ↗openpheno/OpenPheno · tree/main/datasetlines:171-191Plant phenotyping relevance match · UnverifiedarXiv · checked 13 Sept 2026
Field / plotNeRF / 3D Gaussian SplattingFruitCounting2D/3D reconstruction
Accurate 3D fruit counting in orchards is challenging due to heavy occlusion, semantic ambiguity between fruits and surrounding structures, and the high computational cost of volumetric reconstruction. Existing pipelines often rely on multi-view 2D segmentation and dense volumetric sampling, which lead to accumulated fusion errors and slow inference. We introduce FruitLangGS, a language-guided 3D fruit counting framework that reconstructs orchard-scale scenes using an adaptive-density Gaussian Splatting pipeline with radius-aware pruning and tile-based rasterization, enabling scalable 3D representation. During inference, compressed CLIP-aligned semantic vectors embedded in each Gaussian are filtered via a dual-threshold cosine similarity mechanism, retrieving Gaussians relevant to target prompts while suppressing common distractors (e.g., foliage), without requiring retraining or image-space masks. The selected Gaussians are then sampled into dense point clouds and clustered geometrically to estimate fruit instances, remaining robust under severe occlusion and viewpoint variation. Experiments on nine different orchard-scale datasets demonstrate that FruitLangGS consistently outperforms existing pipelines in instance counting recall, avoiding multi-view segmentation fusion errors and achieving up to 99.7% recall on Pfuji-Size_Orch2018 orchard dataset. Ablation studies further confirm that language-conditioned semantic embedding and dual-threshold prompt filtering are essential for suppressing distractors and improving counting accuracy under heavy occlusion. Beyond fruit counting, the same framework enables prompt-driven 3D semantic retrieval without retraining, highlighting the potential of language-guided 3D perception for scalable agricultural scene understanding.
Why it matches plant phenotyping methods果実個数という植物器官形質を推定する3D画像解析手法を開発し、複数データセットで性能検証しているため、植物フェノタイピング手法が研究の中心である。
abstractWe introduce FruitLangGS, a language-guided 3D fruit counting framework
The spike number (SN) is an important trait that significantly impacts grain yield in wheat. Manual counting of SN is time-consuming, hindering large-scale breeding efforts. Hence, there is an urgent need to develop efficient and accurate methodologies for SN counting. A YOLOX algorithm was used to determine the optimal growth stage for developing wheat spike detection models among recombinant inbred lines (RILs) across Zhongmai 175 × Lunxuan 987 and a diverse panel of 166 cultivars. We subsequently increased the precision of spike identification by developing a new YOLOX-P algorithm that incorporates the convolutional block attention module and increasing the resolution of the input images. We also used these SN data to identify underlying loci in the Zhongmai 578 × Jimai 22 RIL population. The results revealed that the late grain-filling stage presented the highest precision among the SN detection models, with accuracies ranging from 91.8 to 95.02 %. The improved YOLOX-P algorithm demonstrated higher mean average precision scores (5.30−5.99 %) and F1 scores (0.06) than did the YOLOX algorithm when it was applied to the same subsets. Three new SN loci, namely, QSN.caas-4A2, QSN.caas-4D and QSN.caas-5B2, were identified using the 50k SNP arrays. Two kompetitive allele-specific PCR markers linked with QSN.caas-4A2 and QSN.caas-5B2 were developed, and their genetic effects were validated in a diverse panel of 166 cultivars. These findings provide useful tools for high-throughput identification of SNs and novel loci in wheat.
Why it matches plant phenotyping methodsRGB画像とYOLOX系コンピュータビジョンによりコムギの穂数を自動推定する手法の開発・比較が中心であり、遺伝子座同定への応用も行っているため、植物表現型手法論文として採用。
abstractManual counting of SN is time-consuming, hindering large-scale breeding efforts. Hence, there is an urgent need to develop efficient and accurate methodologies for SN counting.
Precisely identifying missing virus-free strawberry mother plants in nutrient pots post-transplantation is crucial for optimizing seedling management and maximizing yields in glass greenhouses. Thus, we present an automated method for detecting and counting missing seedlings based on SSP-MambaNet. Challenges in this process include the variable growth morphology of seedlings and complex environmental conditions in the greenhouse. Our approach starts with SPDFFA (Spatial-to-Depth Feature Fusion Attention) to enhance feature representation while retaining critical information, ensuring the preservation of key details. Additionally, the multi-scale CVSSB(Complex Visual State Space) and CVSSB-E(Expanded CVSSB) modules combine multi-scale and multi-directional spatial features, augmenting the model's capacity to recognize inter-image dependencies. Secondly, the MPDIoU is a novel loss function to tackle the optimization challenge of bounding boxes with similar shapes but different sizes, which enhances the accuracy of localizing strawberry seedlings and nutrient pots. Finally, Distance Intersection over Union is utilized for establishing a belongingness relationship between strawberry seedlings and pots, accurately identifying missing seedlings and counting the corresponding pots. Experimental results demonstrate that SSP-MambaNet achieves 94.9 %in average precision, 92.8 % in recall rate,88.1 % in precision, and 90.4 % F1 score for strawberry seedlings and pots. It outperforms the YOLOv7 by 4.7 % in average precision, and 2.6 % in recall rate while reducing 66.7 f/s in FPS. Furthermore, the proposed method shows 94.29 % accuracy in detecting missing seedlings and 97.14 % accuracy in counting nutrient pots with missing seedlings. These results showcase its effectiveness in improving overall seedling quality and providing timely replanting guidance in glass greenhouses.
Why it matches plant phenotyping methods温室内のイチゴ苗の欠損状態を画像から検出・計数する自動手法を開発し、精度比較・検証しており、植物状態の取得方法が研究の中心です。
abstractwe present an automated method for detecting and counting missing seedlings based on SSP-MambaNet.
AppleMangoPeachPearPlumField / plotNeRF / 3D Gaussian SplattingRGB / grayscaleFruitWhole plant / canopy / plot / field
FruitNeRF++: A Generalized Multi-Fruit Counting Method Utilizing Contrastive Learning and Neural Radiance Fields We introduce FruitNeRF++, a novel fruit-counting approach that combines contrastive learning with neural radiance fields to count fruits from unstructured input photographs of orchards. Our work is based on FruitNeRF [6], which employs a neural semantic field combined with a fruit-specific clusteringapproach. The requirement for adaptation for each fruit type limits the applicability of the method, and makes it difficult to use in practice. To lift this limitation, we design a shape-agnostic multi-fruit counting framework, that complements the RGB and semantic data with instance masks predicted by a vision foundation model. The masks are used to encode the identity of each fruit as instance embeddings into a neural instance field. By volumetrically sampling the neural fields, we extract apoint cloud embedded with the instance features, which can be clustered in a fruit-agnostic manner to obtain the fruit count. We evaluate our approach using a synthetic dataset containing apples, plums, lemons, pears, peaches, and mangoes, as well as a real-world benchmark apple dataset. Our results demonstrate that FruitNeRF++ is easier to control and compares favorably to other state-of-the-art methods.
Why it matches plant phenotyping methods果実を対象とした画像ベースの汎用カウント手法を開発し、合成および実データで評価しているため、植物形質（果実数）の取得・推定が研究の中心です。
abstractWe introduce FruitNeRF++, a novel fruit-counting approach that combines contrastive learning with neural radiance fields to count fruits from unstructured input photographs of orchards.
Introduction Missing seedlings is a common issue in field maize planting, arising from limitations in sowing machinery and seed germination rates. This phenomenon directly impacts maize yields owing to the poor effect of unmanned aerial vehicle (UAV) remote sensing images based on seedling leakage detection in fields. Therefore, this study proposed a method for detecting missing seedling in fields based on UAV remote sensing to quickly and accurately detect missing seedling and facilitate subsequent crop management decisions. Methods The method calculates the rated inter-seedling distance in UAV-captured images of maize fields using a combination of image processing techniques, including background segmentation, stalk center region detection, linear fitting of plant rows, and average plant distance calculation. Based on these calculations, an improved Maize-YOLOv8n model was employed to detect actual seedling emergence. Results The experimental results demonstrate that the new model achieved superior performance on a self-constructed dataset, with a mean average precision (mAP) of 97.4%, precision (P) of 94.3%, recall (R) of 93.1%, and an F1 score of 93.7%. The model was lightweight, comprising only 1.19 million parameters and requiring 20.2 floating-point operations per second (FLOPs). The inference time was 12.8 ms, satisfying real-time detection requirements. Performance evaluations across various conditions, including different leaf stages, light intensities, and weed interference levels, further indicated the robustness of the model. In addition, a linear regression equation was developed to predict the total number of missing seedlings, with model performance evaluated using the root mean squared error (RMSE) and mean absolute error (MAE) metrics. Discussion The results confirm the ability of the model to accurately detect maize seedling gaps. This study can evaluate the quality of seeding operations and provide accurate information on the number of missing seedlings for timely replacement work in areas with high rates of missing seedlings. This study advances precision agriculture by enhancing the efficiency and accuracy of maize planting management.
Why it matches plant phenotyping methodsUAV画像と画像処理・YOLOモデルを用いて、トウモロコシの欠株という植物状態を検出・定量化する手法が研究の中心であり、性能評価も実施している。
abstractthis study proposed a method for detecting missing seedling in fields based on UAV remote sensing to quickly and accurately detect missing seedling
Chloroplast density in cells varies among different types of cells and plants. In current single-cell spatiotemporal analysis, the automatic detection and quantification of chloroplasts at the single-cell level is crucial. We developed DeepD&Cchl (Deep-learning-based Detecting-and-Counting-chloroplasts), an AI tool for single-cell chloroplast detection and cell-type clustering. It utilizes You-Only-Look-Once (YOLO), a real-time detection algorithm, for accurate and efficient performance. DeepD&Cchl has been proved to identify chloroplasts in plant cells across various imaging types, including light microscopy, electron microscopy, and fluorescence microscopy. Integrated with an Intersection Over Union (IOU) module, DeepD&Cchl precisely counts chloroplasts in single- or multi-layered images, while eliminating double-counting errors. Furthermore, when combined with Cellpose, a single-cell segmentation tool, DeepD&Cchl enhances its effectiveness at the single-cell level. By counting chloroplasts within individual cells, it supports cell-type-specific clustering based on chloroplast number versus cell size, offering valuable morphological insights for single-cell studies. In summary, DeepD&Cchl is a significant advancement in plant cell analysis. It offers accuracy and efficiency in chloroplast identification, counting and cell-type classification, providing a useful tool for plant research.
Why it matches plant phenotyping methods植物細胞画像から葉緑体を検出・計数し、細胞型をクラスタリングするAIツールの開発が中心であり、植物の形態的状態を定量化するフェノタイピング手法に該当する。
abstractWe developed DeepD&Cchl (Deep-learning-based Detecting-and-Counting-chloroplasts), an AI tool for single-cell chloroplast detection and cell-type clustering.
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 raw dataset, as well as the scripts for the DeepD&Cchl model training and 17application macro, were shared on GitHub https://github.com/Shaokai9/AI4LifeScience_ECNU/tree/main/Deep%20subcellular%20detection .Open asset ↗Shaokai9/AI4LifeScience_ECNUlines:400-410Plant phenotyping relevance match · UnverifiedCrossref · checked 6 Sept 2026
Published21 May 2025Journal of The Institution of Engineers (India): Series ACited by 1 · OpenAlex ↗
High resolution three-dimensional (3D) point clouds enable the mapping of cotton boll spatial distribution, aiding breeders in better understanding the correlation between boll positions on branches and overall yield and fiber quality. This study developed a segmentation workflow for point clouds of 18 cotton genotypes to map the spatial distribution of bolls on the plants. The data processing workflow includes two independent approaches to map the vertical and horizontal distribution of cotton bolls. The vertical distribution was mapped by segmenting bolls using PointNet++ and identifying individual instances through Euclidean clustering. For horizontal distribution, TreeQSM segmented the plant into the main stem and individual branches. PointNet++ and Euclidean clustering were then used to achieve cotton boll instance segmentation. The horizontal distribution was determined by calculating the Euclidean distance of each cotton boll relative to the main stem. Additionally, branch types were classified using point cloud meshing completion and the Dijkstra shortest path algorithm. The results highlight that the accuracy and mean intersection over union (mIoU) of the 2-class segmentation based on PointNet++ reached 0.954 and 0.896 on the whole plant dataset, and 0.968 and 0.897 on the branch dataset, respectively. The coefficient of determination (R 2 ) for the boll counting was 0.99 with a root mean squared error (RMSE) of 5.4. For the first time, this study accomplished high-granularity spatial mapping of cotton bolls and branches, but directly predicting fiber quality from 3D point clouds remains a challenge. This method provides a promising tool for 3D cotton plant mapping of different genotypes, which potentially could accelerate plant physiological studies and breeding programs.
Why it matches plant phenotyping methods3D点群の分割・個体抽出ワークフローを開発し、綿花の果実数と枝・果実の空間分布という植物形質を定量化・検証しており、フェノタイピング手法が研究の中心である。
abstractThis study developed a segmentation workflow for point clouds of 18 cotton genotypes to map the spatial distribution of bolls on the plants.
Reproduction assets foundThe authors explicitly state that the code, data, and trained PointNet++ weights for cotton boll and branch mapping are publicly available in their GitHub repository, which directly reproduces this paper's phenotyping analysis.Code · publicThe code, data, and training weights for cotton boll and branch mapping are available at https://github.com/UGA-BSAIL/cotton_organ_mapping.git .Open asset ↗UGA-BSAIL/cotton_organ_mappinglines:101-109Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Accurate estimation of hazelnut yield is crucial for optimizing resource management and harvest planning. Although the number of female flowers on a flowering plant is a reliable indicator of annual production, counting them remains difficult because of their extremely small size and inconspicuous shape and color. Currently, manual flower counting is the only available method, but it is time-consuming and prone to errors. In this study, a novel vision-based method for automatic flower counting specifically designed for hazelnut plants ( Corylus avellana ) exploiting a commercial high-resolution imaging system and an image-tiling strategy to enhance small-object detection is proposed. The method is designed to be fast and scalable, requiring less than 8 s per plant for processing, in contrast to 30-60 min typically required for manual counting by human operators. A dataset of 2000 labeled frames was used to train and evaluate multiple female hazelnut flower detection models. To improve the detection of small, low-contrast flowers, a modified YOLO11x architecture was introduced by adding a P2 layer, improving the preservation of fine-grained spatial information and resulting in a precision of 0.98 and a Mean Average Precision (mAP@50-95) of 0.89. The proposed method has been validated on images collected from hazelnut groves and compared with manual counting by four experienced operators in the field, demonstrating its ability to detect small, low-contrast flowers despite occlusions and varying lighting conditions. A regression-based bias correction was applied to compensate for systematic counting deviations, further improving accuracy and reducing the mean absolute percentage error to 27.44%, a value comparable to the variability observed in manual counting. The results indicate that the system can provide a scalable and efficient alternative to traditional female flower manual counting methods, offering an automated solution tailored to the unique challenges of hazelnut yield estimation.
Why it matches plant phenotyping methodsハシバミ植物の雌花数という収量関連形質を画像から自動抽出する手法を開発し、データセット、モデル改良、現地検証、手動計数との比較まで行っており、フェノタイピング手法が中心である。
abstractIn this study, a novel vision-based method for automatic flower counting specifically designed for hazelnut plants ( Corylus avellana ) exploiting a commercial high-resolution imaging system and an image-tiling strategy to enhance small-object detection is proposed.
We present an AI pipeline that involves using smart drones equipped with computer vision to obtain a more accurate fruit count and yield estimation of the number of blueberries in a field. The core components are two object-detection models based on the YOLO deep learning architecture: a Bush Model that is able to detect blueberry bushes from images captured at low altitudes and at different angles, and a Berry Model that can detect individual berries that are visible on a bush. Together, both models allow for more accurate crop yield estimation by allowing intelligent control of the drone's position and camera to safely capture side-view images of bushes up close. In addition to providing experimental results for our models, which show good accuracy in terms of precision and recall when captured images are cropped around the foreground center bush, we also describe how to deploy our models to map out blueberry fields using different sampling strategies, and discuss the challenges of annotating very small objects (blueberries) and difficulties in evaluating the effectiveness of our models.
Why it matches plant phenotyping methodsドローン画像とYOLOモデルによりブルーベリー果実数および収量を推定する手法を開発・評価しており、植物形質の取得・推定が研究の中心である。
abstractWe present an AI pipeline that involves using smart drones equipped with computer vision to obtain a more accurate fruit count and yield estimation of the number of blueberries in a field.
The real-time monitoring and counting of maize seed germination at seedling stage is of great significance for seed quality detection, field management and yield estimation. Traditional manual monitoring and counting is very time-consuming, cumbersome and error-prone. In order to quickly and accurately identify and count maize seedlings in a complex field environment, this study proposes an end-to-end maize seedling plant detection model H-RT-DETR (Hierarchical-Real-Time DEtection TRansformer) based on hierarchical feature extraction and RT-DETR (Real-Time DEtection TRansformer). H-RT-DETR uses Hierarchical Feature Representation and Efficient Self-Attention as the backbone network for feature extraction, thereby improving the network's ability to extract features of maize seedling stage in UAV remote sensing images. Through experiments on the UAV remote sensing data set of maize seedling stage, the mean Average Precision mAP0.5-0.95, mAP0.5 and mAP0.75 of the improved H-RT-DETR model reached 51.2%, 94.7% and 48.1%, respectively, and the Average Recall (AR) reached 68.5%. In order to verify the efficiency of the proposed method, H-RT-DETR is compared with the widely used and advanced target recognition methods. The results show that the detection accuracy of H-RT-DETR is better than that of the comparison methods. In terms of detection speed, the H-RT-DETR model does not require Non-Maximum Suppression (NMS) post-processing operations, the Frames Per Second (FPS) on the test dataset reaches 84f/s, which is 19,12,11 and 21 higher than that of YOLOv5, YOLOv7, YOLOv8 and YOLOX, respectively, under the same hardware environment. This model can provide technical support for real-time detection of maize seedlings under UAV remote sensing images in terms of both detection accuracy and speed (see https://github.com/wylSUGAR/H-RT-DETR for model implementation and results).
Why it matches plant phenotyping methodsUAV画像からトウモロコシ幼苗を検出・計数する深層学習手法を開発し、既存手法と精度・速度を比較検証しているため、植物フェノタイピング手法が中心である。
abstractIn order to quickly and accurately identify and count maize seedlings in a complex field environment, this study proposes an end-to-end maize seedling plant detection model H-RT-DETR
Reproduction assets foundThe authors provide a public GitHub repository containing the H-RT-DETR model implementation and results for maize seedling detection in UAV images. The UAV image dataset itself is not stated as publicly available, and other linked repositories (labelme, YOLOv5, ultralytics) are generic third-party tools, not paper-Code · publicthe test dataset reaches 84f/s, which is 19,12,11 and 21 higher than that of YOLOv5, YOLOv7, YOLOv8 and YOLOX, respectively, under the same hardware environment. This model can provide technical support for real-time detection of maize seedlings under UAV remote sensing images in terms of both detection accuracy and speed (see https://github.com/wylSUGAR/H-RT-DETR for model implementation and results).
Keywords
Maize seedling stage UAV remote sensing RT-DETR Target recognition Real-time detection Plant counting the National Key Research and Development Program of China 2023YFD1900704 2023YFD1900704 pmc-status-qastatus 0 pmc-status-live yes pmc-status-embargo no pmc-status-released yes pmOpen asset ↗wylSUGAR/H-RT-DETRlines:1-26Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
The spike number (SN) is an important trait that significantly impacts grain yield in wheat. Manual counting of SN is time-consuming, hindering large-scale breeding efforts. Hence, there is an urgent need to develop efficient and accurate methodologies for SN counting. A YOLOX algorithm was used to determine the optimal growth stage for developing wheat spike detection models among recombinant inbred lines (RILs) across Zhongmai 175 × Lunxuan 987 and a diverse panel of 166 cultivars. We subsequently increased the precision of spike identification by developing a new YOLOX-P algorithm that incorporates the convolutional block attention module and increasing the resolution of the input images. We also used these SN data to identify underlying loci in the Zhongmai 578 × Jimai 22 RIL population. The results revealed that the late grain-filling stage presented the highest precision among the SN detection models, with accuracies ranging from 91.8 to 95.02 %. The improved YOLOX-P algorithm demonstrated higher mean average precision scores (5.30-5.99 %) and F1 scores (0.06) than did the YOLOX algorithm when it was applied to the same subsets. Three new SN loci, namely, QSN . caas-4A2, QSN . caas-4D and QSN . caas-5B2 , were identified using the 50k SNP arrays. Two kompetitive allele-specific PCR markers linked with QSN . caas-4A2 and QSN . caas-5B2 were developed, and their genetic effects were validated in a diverse panel of 166 cultivars. These findings provide useful tools for high-throughput identification of SNs and novel loci in wheat.
Why it matches plant phenotyping methodsRGB画像とコンピュータビジョンによりコムギの穂数を自動推定する手法を開発・比較し、精度を評価しているため、植物フェノタイピング手法が中心である。
abstractHence, there is an urgent need to develop efficient and accurate methodologies for SN counting.
Reproduction assets foundThe paper publicly releases two paper-specific assets: (1) a wheat spike number image dataset (subsets CD&DD&XX) on GitHub, and (2) the YOLOX-P analysis code on Google Drive. Both have explicit availability statements with author-provided URLs.Dataset · publicThe image set for CD&DD&XX is publicly available on GitHub ( https://github.com/lileimax/YOLOXP-wheat-spike-identification ).Open asset ↗https://github.com/lileimax/YOLOXP-wheat-spike-identificationlines:223-246Code · publicThe code for YOLOX-P is publicly available on Google Drive ( https://drive.google.com/drive/folders/1urCDUdyrq14FuwG2I3YwCZraUGEEl_8X?usp=sharing ).Open asset ↗lines:247-250Plant phenotyping relevance match · UnverifiedEurope PMC · Crossref · checked 6 Sept 2026
Introduction The number of rice leaves largely reflects the growth stage and health status of rice. However, the current rice leaf counting method is time-consuming and laborious, with low accuracy and poor efficiency, which is difficult to meet the needs of rice growth monitoring. Methods This study proposes a field rice leaf detection method based on an improved YOLOv5s model. First, we added a high-resolution layer and removed the original low-resolution detection layer, using the K-Means++ clustering algorithm to reset the anchor box sizes, enhancing the model's ability to identify small leaf tip targets while reducing the number of parameters. Second, we introduced a coordinate attention mechanism (CA) to strengthen focus on leaf tip features under weed interference and leaf occlusion conditions. Finally, we employed a content-aware reassembly of feature (CARAFE) upsampling operator to enhance the detail reconstruction capability of leaf tip features. Results and discussion Experimental results showed that the improved rice leaf tip detection model achieved precision, recall, and mean average precision rates of 93.7%, 87%, and 93.5%, respectively, with a parameter count of 5.02 million (M), improving by 6.5%, 22.1%, and 18.5% compared to the YOLOv5s baseline model, while reducing the parameter count by 28.4%. The improved model effectively reduced the missed detection rate of rice leaves and enhanced the accuracy and robustness of field rice leaf tip detection, providing strong technical support for rice phenotype feature extraction and growth monitoring.
Why it matches plant phenotyping methods水稲葉先の検出・カウントを通じた葉数という植物形質の抽出手法を改良し、精度・頑健性をYOLOv5sと比較検証しており、フェノタイピング手法が中心である。
abstractThis study proposes a field rice leaf detection method based on an improved YOLOv5s model.
Among various types of forages, Alfalfa (Medicago sativa) is a crucial forage crop that plays a vital role in livestock nutrition and sustainable agriculture. As a result of its ability to adapt to different weather conditions and its high nitrogen fixation capability, this crop produces high-quality forage that contains between 15 and 22 % protein. It is fortunately possible to improve the overall prediction of forage biomass and quality prior to harvest through remote sensing technologies. The recent advent of deep Convolution Neural Networks (deep CNNs) enables researchers to utilize these incredible algorithms. This study aims to build a model to count the number of alfalfa stems from proximal images. To this end, we first utilized a deep CNN encoder-decoder to segment alfalfa and other background objects in a field, such as soil and grass. Subsequently, we employed the alfalfa cover fractions derived from the proximal images to develop and train machine learning regression models for estimating the stem count in the images. This study uses many proximal images taken from significant number of fields in four provinces of Canada over three consecutive years. A combination of real and synthetic images has been utilized to feed the deep neural network encoder-decoder. This study gathered roughly 3447 alfalfa images, 5332 grass images, and 9241 background images for training the encoder-decoder model. With data augmentation, we prepared about 60,000 annotated images of alfalfa fields containing alfalfa, grass, and background utilizing a pre-trained model in less than an hour. Several convolutional neural network encoder-decoder models have also been utilized in this study. Simple U-Net, Attention U-Net (Att U-Net), and ResU-Net with attention gates have been trained to detect alfalfa and differentiate it from other objects. The best Intersections over Union (IoU) for simple U-Net classes were 0.98, 0.93, and 0.80 for background, alfalfa and grass, respectively. Simple U-Net with synthetic data provides a promising result over unseen real images and requires an RGB iPad image for field-specific alfalfa detection. It was also observed that simple U-Net has slightly better accuracy than attention U-Net and attention ResU-Net. Finally, we built regression models between the alfalfa cover fraction in the original images taken by iPad, and the mean alfalfa stems per square foot. Random forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGB) methods have been utilized to estimate the number of stems in the images. RF was the best model for estimating the number of alfalfa stems relative to other machine learning algorithms, with a coefficient of determination (R²) of 0.82, root-mean-square error of 13.00, and mean absolute error of 10.07.
Why it matches plant phenotyping methods近接画像からアルファルファをセグメンテーションし、茎数という植物形質を推定する画像解析・機械学習手法の開発と評価が中心である。
abstractThis study aims to build a model to count the number of alfalfa stems from proximal images.
Grapevine phenotyping, that is the process of determining the physical properties (e.g., size, shape, and number) of grape bunches, provides valuable information for growth and health monitoring, yield estimation and efficient crop management in precision viticulture. Currently, grape bunch counting and sizing is done manually, which is labor intensive and often impractical for large-scale field applications. This paper describes a novel framework to automatically detect, count and estimate the volume/weight of grape bunches using RGB and depth data acquired in the field by a farmer robot. The proposed pipeline starts with the semantic segmentation of RGB images based on a pre-trained MANet architecture with EfficientnetB3 backbone to separate fruit from non-fruit regions. The segmented fruit mask is then projected onto the co-registered depth image to recover a depth mask, allowing for three-dimensional (3D) data association. After a pre-processing step to correct anomalies, such as corrupted and missing values, and to remove outliers, a depth gradient-based clustering algorithm is applied that detects individual grape bunch clusters. This enables the separation of adjacent and partially overlapping bunches. In addition, a method to reconstruct the whole 3D shape of a bunch is introduced, so as to provide an estimate of volume and weight. Experiments performed in a commercial vineyard in Italy are presented showing that, despite the low quality and high variability of the input images, the proposed approach is able to count grape bunch clusters with an average error of about 12% with respect to visual ground-truth and an average error less than 30% with respect to manual weight measurements. It is also shown that the processing framework can be applied to geo-referenced image sequences acquired by the farmer robot while traversing vineyard rows, thus providing an automated pipeline for the generation of high-resolution yield maps for precision viticulture applications.
Why it matches plant phenotyping methodsブドウ房の検出・計数・3D形状復元により体積・重量を推定する画像・深度ベースの表現型取得手法を開発し、精度検証も行っているため、方法が中心的である。
abstractThis paper describes a novel framework to automatically detect, count and estimate the volume/weight of grape bunches using RGB and depth data acquired in the field by a farmer robot.
The number of cotton bolls is an important phenotyping trait not only for breeders but also for growers. It can provide information on the physiological and genetic mechanisms of plant growth and aid decision-making in crop management. However, traditional visual inspection in the field is time-consuming and laborious. With the application of drones in the agricultural domain, there is promising potential to collect data expediently. In this paper, we integrated the improved Distribution Matching for crowd Counting (DM-Count) and Segment Anything Model (SAM) to predict cotton boll number, size, and yield in aerial images. The cotton plots were first extracted from the raw aerial images using boundaries derived from orthophotos. Then, a convolutional neural network (DM-Count) was introduced as a baseline and customized by replacing the VGG19 backbone and adding a pixel loss. The customized network was first pretrained on ground images and then fine-tuned on aerial images to predict the density map, where the number and locations of cotton bolls can be obtained. The zero-shot foundation model SAM was investigated to segment cotton bolls with the point prompts provided by customized DM-Count. The respective numbers of bolls and segmented pixels were compared for seed cotton yield estimation. The experimental results showed that the customized model obtained a mean absolute error (MAE) of 1.78 per square meter and a mean absolute percentage error (MAPE) of 4.39 % on the testing dataset, with a high correlation between the predicted boll number and ground truth (R² = 0.91). The AP50 of SAM for cotton boll segmentation was 0.63. The segmented masks were used to delineate the boll size differences among the four genotypes, and it was found that the average boll size of Pima was 452 pixels, which was significantly smaller than Acala Maxxa, UA 48 and Tamcot Sphinx. Moreover, the yield estimation using the boll number was better than that using the pixel number, with an R² = 0.70. Combing the boll number and the pixel number can achieve a slightly higher R² of 0.72 for yield estimation. Overall, the customized model can count cotton bolls in aerial images accurately and estimate seed cotton yield effectively, which could significantly benefit breeders in developing genotypes with high yields, as well as help growers in yield estimation and crop management.
Why it matches plant phenotyping methods綿花のボール数・サイズ・収量を航空画像から推定する画像解析手法を開発・評価しており、植物形質取得が研究の中心である。
abstractIn this paper, we integrated the improved Distribution Matching for crowd Counting (DM-Count) and Segment Anything Model (SAM) to predict cotton boll number, size, and yield in aerial images.
The effective tiller number of wheat (ETNW) is one of the three main factors affecting wheat yield. Most traditional methods for counting tillers are manual, which is inefficient and challenging to implement on a wide scale. Existing remote sensing techniques for tiller counting are typically focused on individual plants or small plot areas, often utilizing high-resolution sensors for close-range monitoring. This approach limits the scalability and applicability for large-scale field environments. Moreover, most studies on wheat tiller monitoring have concentrated on specific crop varieties or single-variable conditions, with little research conducted on estimating or monitoring ETNW in large-scale field scenarios using consumer-grade unmanned aerial vehicle (UAV) platforms. To address these issues, we propose a multi-modal fusion-driven machine learning method to enhance the performance of wheat tillering monitoring. This study was conducted using two experimental setups to ensure robust and comprehensive data collection. Experiment 1 (Exp.1) focused on water and nitrogen coupling conditions, while Experiment 2 (Exp.2) included both nitrogen-deficient and nitrogen-sufficient treatments. Each experiment involved multiple wheat varieties to account for genotypic variability. UAV data, including multispectral and RGB imagery, was collected across different growth stages under varying irrigation and nitrogen conditions to ensure the generalizability of the proposed model. This method employs spectral correlation analysis (SCA) to select features strongly correlated with the number of tillers and subsequently fuses these features to construct a multi-modal machine learning model based on vegetation index (VI), color index (CI), multispectral texture features (TF1), and RGB texture features (TF2) derived from UAV data. Three machine learning models are applied: Random Forest Regression (RFR), Partial Least Squares Regression (PLSR), and Support Vector Regression (SVR). The results demonstrated that the vegetation index-driven machine learning model (VI-RFR) achieved the best performance, with an R² of 0.85, RMSE of 348.60, and NRMSE of 0.367, followed by the color index model (CI-RFR, R² = 0.83, RMSE = 313.18, NRMSE = 0.330), the multispectral texture feature model (TF1-RFR, R² = 0.81, RMSE = 357.12, NRMSE = 0.376), and the RGB texture feature model (TF2-RFR, R² = 0.80, RMSE = 373.69, NRMSE = 0.393). Compared to using a single data source, data fusion significantly improved model accuracy, particularly when complementary data sources were combined. Specifically, the VI&CI combination achieved an R² improvement of 6.4 %-19.1 %, an RMSE reduction of 2.8 %-9%, and an NRMSE decrease of 1.56 %-8.24 %. The VI&TF1 combination exhibited an R² increase of 7.8 %–32.6 %, an RMSE reduction of 8.5 %-28.8 %, and an NRMSE decrease of 2.20 %-8.24 %. The VI&TF2 combination showed an R² increase of 3.8 %-40.5 %, an RMSE reduction of 2.8 %-27.6 %, and an NRMSE decrease of 2.68 %-21.61 %. The CI&TF1 combination resulted in an R² improvement of 5.2 %-25.6 %, an RMSE reduction of 7.4 %-24.1 %, and an NRMSE decrease of 2.04 %-19.41 %. The CI&TF2 combination achieved an R² increase of 12.5 %–33.3 %, an RMSE reduction of 5.4 %–23.4 %, and an NRMSE decrease of 1.17 %-18.94 %. The TF1&TF2 combination achieved an R² improvement of 13.9 %–33.3 %, an RMSE reduction of 10.9 %-27.6 %, and an NRMSE decrease of 5.01 %-21.61 %. However, increasing the number of data sources does not necessarily lead to higher model accuracy. The VI&CI&TF1 fusion model (RFR, R² = 0.90, RMSE = 288.39, NRMSE = 0.304) demonstrated the best performance, surpassing the combination of all four features. The machine learning models exhibited systematic variations, with the performance ranking as follows: RFR > SVR > PLSR, regardless of the data fusion strategy employed. Moreover, the best model (VI&CI&TF1-RFR) demonstrated adaptability across different growth stages, irrigation conditions, and nitrogen fertilizer applications. Notably, the model maintained robustness even in extreme environments with complete nitrogen deficiency. The findings of this study provide technical support for crop phenotyping, variety selection, and precision agriculture management.
Why it matches plant phenotyping methodsUAVのマルチスペクトル・RGB画像から小麦の有効分げつ数を推定する特徴融合・機械学習手法を開発し、複数条件・品種で性能評価しており、植物表現型の取得・推定が研究の中心である。
abstractTo address these issues, we propose a multi-modal fusion-driven machine learning method to enhance the performance of wheat tillering monitoring.
Citizen science is an effective approach for collecting extensive data scalable for deep learning, although data quality is debatable. However, few studies have determined the factors associated with data collection that affect model performance and potential sampling bias. This study aims to identify the factors that significantly influence the performance of a deep learning object detection model in agricultural prediction tasks. To do so, we analyzed errors in a You Only Look Once (YOLO v8) model trained for counting the number of coffee cherries in mobile pictures. The model was trained with 436 images taken in Colombia and Peru collected by local farmers as a citizen science approach. We analyzed the prediction errors of the model using 637 additional pictures. We then applied a linear mixed model (LMM) and a decision tree machine learning model to regress the model’s error against predictor variables related to the following categories: photographer influence, geographic location, mobile phone characteristics, picture characteristics, and coffee varieties. Our results show the strong influence of photographer identity and adherence (whether the image collection protocol was followed or not) on model prediction error. Following the protocol can increase model performance from an R2 of 0.48 to 0.73. Additionally, model performance varied significantly depending on photographer identity, with R2 ranging from 0.45 to 0.93. In contrast, factors such as mobile phone characteristics (e.g., frontal camera resolution, flash type, and screen size), using the screen behind the branch to obscure other cherries, coffee varieties, and geographic location did not significantly affect prediction error. These findings demonstrate that data quality in citizen science–based data collection for enhancing model prediction can be achieved through straightforward and comprehensive protocols, customized volunteer training, and regular feedback from experts. Such measures collectively support the robust application of deep learning models in agriculture. Furthermore, this study demonstrated that any mobile device with a camera can contribute to citizen science initiatives, underscoring the potential and scalability of this approach in agricultural research.
Why it matches plant phenotyping methodsコーヒーチェリー数という植物器官の形質を画像から推定する深層学習モデルを対象に、予測誤差、撮影者、プロトコル遵守などを分析して技術性能を検証しており、フェノタイピング手法が中心である。
abstractWe analyzed the prediction errors of the model using 637 additional pictures.
Orchard yield estimation is one of the key indicators of precision agriculture. The traditional random sampling yield estimation method has strict requirements for the laborer experience and scale of orchards. Intelligent orchard management enables growers to use resources more effectively and make wiser decisions to optimize orchard inputs. This study proposes a banana bunch counting and yield estimation method based on the DeepSORT tracking algorithm. This method involves obtaining RGB-D images and calculating the weight of an individual bunch of bananas, which was promoted in our previous work. Building on this, the DeepSORT was used to solve the repeated counting based on the Hungarian algorithm and Kalman filtering. Three constraints were set to improve the statistical accuracy, and a yield estimation system was designed for orchard management monitoring. This system provides managers with bunch weight predictions and statistical plant information to achieve real-time yield estimations for banana orchards. The experimental results showed that the accuracy of the yield estimations reached 97.25% and that banana bunch counting had a success rate of 96.82%. This demonstrates that the effective integration of RGB-D technology and the DeepSORT algorithm can be successfully applied to the intelligent management and harvesting of banana orchards.
Why it matches plant phenotyping methodsRGB-D画像とDeepSORTによるバナナ房の計数・重量推定という、植物の収量形質を抽出する画像ベース手法が研究の中心であり、精度評価も実施しているため。
abstractThis study proposes a banana bunch counting and yield estimation method based on the DeepSORT tracking algorithm.
Trichomes in Cannabis sativa are specialized structures responsible for cannabinoid and terpene biosynthesis, making their density a critical factor for both research and industrial applications. Despite their importance, trichome density analysis is hindered by variability across plant structures and the lack of standardized protocols. This study evaluates different plant structures—bracts, sugar leaves, calyxes, and the main cola—to determine the most reliable site for trichome counting. Among these, bracts emerge as the most consistent due to their homogeneous trichome distribution and high cannabinoid concentration. While sugar leaves and calyxes also contribute to trichome yield assessments, their variability necessitates careful sampling. Moreover, trichome shape and size must be taken into consideration when correlating trichome density with secondary metabolite levels. The integration of microscopic imaging and software-assisted counting enhances accuracy and reproducibility in trichome density analysis. Establishing a standardized protocol for trichome assessment will improve cannabinoid yield optimization, quality control, and overall Cannabis research methodologies. Incorporating morphological data (trichome density, distribution, shape, and size) with chemical assays (cannabinoid and terpene identification and quantification) thus provides a more robust assessment of Cannabis potency and value. Future work should refine imaging techniques and sampling strategies to further enhance trichome analysis reliability.
Why it matches plant phenotyping methodsトリコーム密度・形態を顕微鏡画像とソフトウェアで測定する標準化・再現性向上が研究の中心であり、植物器官の形態形質を取得する方法開発・評価に該当する。
abstractThis study evaluates different plant structures—bracts, sugar leaves, calyxes, and the main cola—to determine the most reliable site for trichome counting.
• Achieved high detection accuracy of apple fruitlets in complex orchard environments with rapid phenological changes. • Provided a dataset of videos and RGB-D images, featuring annotated apple fruitlets and manual caliper measurements during early development. • Developed a workflow for rapid in-field monitoring of flower corymbs and fruitlet sizing, validated through experimental trials. Current research in apple-growing focuses on collecting extensive biometric data to better understand physiological processes, improve orchard productivity and predict yields. In this context, fruit thinning has emerged as a key horticultural practice to enhance fruit size and quality while preventing alternate bearing. Despite the growing role of plant imaging technologies in agronomic management, fruitlet sizing remains challenging, particularly in early phenological stages. To address this challenge, we developed an RGB-D-based vision pipeline that combines YOLO models with depth information and relies on the statistical analysis of frame series to detect and cluster fruitlets into flower corymbs, providing both fruitlet counting and diameter estimates for each video acquisition. After obtaining an AP@0.5 and AP@[0.5:0.95] of respectively 0.894 and 0.77 in fruitlet detection, along with a precision of 0.881 and a recall of 0.846, our approach efficiently processed video frames, extracting the most reliable data for each labeled cluster. While the comparison of true positive estimates with calibrated caliper measurements showed a mean RMSE of 1.05 mm, challenges remain in achieving the correct fruitlet count, with a mean counting error of 0.63 fruitlets per video. Additionally, the proposed workflow retrieved the exact number of fruitlets as the ground truth in 56.4% of the videos, increasing to 75% when excluding those videos where the correct fruitlet count was never detected in any frame by the YOLO model. Despite these limitations, our results are promising, proposing a potential data acquisition tool without compromising the reliability of traditional practices. This approach could pave the way for future applications, including the evaluation of plant growth regulator trials and the development of predictive models for yield and productivity optimization.
Why it matches plant phenotyping methodsRGB-D画像と深度情報、YOLO、動画フレーム統計を組み合わせ、リンゴ果実の検出・計数・直径推定を行うワークフローを開発し、ノギス測定で検証しているため、表現型取得手法が中心である。
abstractProvided a dataset of videos and RGB-D images, featuring annotated apple fruitlets and manual caliper measurements during early development.
In recent years, plant counting using data collected by sensors embedded in remotely piloted aircraft systems (RPAS), combined with machine learning algorithms, has become popular in the agroforestry sector, especially in crop planning, crop production estimation, among other applications. This study aimed to perform a systematic review of the literature on plant counting in the agricultural and forestry sector that uses data from sensors embedded in RPAS. We sought to identify the principal bibliometric indicators of scientific production and, through content analysis, the main characteristics and trends of the studies. A total of 33 scientific articles obtained on the Scopus and Web of Science platforms were used. Then, a content qualitative analysis of each article was conducted to identify the main thematic categories: agricultural and forest species, platform and sensors, software, and algorithm. There was an increase in scientific publications as of 2017. The USA presented the higher number of researches performed, with eight publications. There was a significant presence of RGB (Red, Green and Blue) sensors followed by multispectral. The algorithms Convoluctional Neural Network (CNN), Structure from Motion (SfM), and K-means stood out for the recurrence of use, either singularly or associated. Studies with this purpose drive new research development, where this technology utilization is revealed as a potential instrument to understand the usage trends, subsidize and encourage the information acquisition, promoting improvements and progress for research in the agroforestry scope.
Why it matches plant phenotyping methods農林植物のセンサー画像と機械学習による植物個体数推定を主題とする系統的レビューであり、植物形質取得手法のレビューが中心である。
titleThe Use Remotely Piloted Aircraft in Counting Agricultural and Forestry Plants: a Systematic Review
Automated extraction of plant morphological traits is crucial for supporting crop breeding and agricultural management through high-throughput field phenotyping (HTFP). Solutions based on multi-view RGB images are attractive due to their scalability and affordability, enabling volumetric measurements that 2D approaches cannot directly capture. While advanced methods like Neural Radiance Fields (NeRFs) have shown promise, their application has been limited to counting or extracting traits from only a few plants or organs. Furthermore, accurately measuring complex structures like individual wheat heads-essential for studying crop yields-remains particularly challenging due to occlusions and the dense arrangement of crop canopies in field conditions. The recent development of 3D Gaussian Splatting (3DGS) offers a promising alternative for HTFP due to its high-quality reconstructions and explicit point-based representation. In this paper, we present Wheat3DGS, a novel approach that leverages 3DGS and the Segment Anything Model (SAM) for precise 3D instance segmentation and morphological measurement of hundreds of wheat heads automatically, representing the first application of 3DGS to HTFP. We validate the accuracy of wheat head extraction against high-resolution laser scan data, obtaining per-instance mean absolute percentage errors of 15.1%, 18.3%, and 40.2% for length, width, and volume. We provide additional comparisons to NeRF-based approaches and traditional Muti-View Stereo (MVS), demonstrating superior results. Our approach enables rapid, non-destructive measurements of key yield-related traits at scale, with significant implications for accelerating crop breeding and improving our understanding of wheat development.
Why it matches plant phenotyping methods3D Gaussian SplattingとSAMを用いて小麦穂の3Dセグメンテーションと形態形質抽出手法を開発し、レーザースキャン、NeRF、MVSと比較検証しているため、植物フェノタイピング手法が中心である。
abstractwe present Wheat3DGS, a novel approach that leverages 3DGS and the Segment Anything Model (SAM) for precise 3D instance segmentation and morphological measurement of hundreds of wheat heads automatically
Precisely identifying missing virus-free strawberry mother plants in nutrient pots post-transplantation is crucial for optimizing seedling management and maximizing yields in glass greenhouses. Thus, we present an automated method for detecting and counting missing seedlings based on SSP-MambaNet. Challenges in this process include the variable growth morphology of seedlings and complex environmental conditions in the greenhouse. Our approach starts with SPDFFA (Spatial-to-Depth Feature Fusion Attention) to enhance feature representation while retaining critical information, ensuring the preservation of key details. Additionally, the multi-scale CVSSB(Complex Visual State Space) and CVSSB-E(Expanded CVSSB) modules combine multi-scale and multi-directional spatial features, augmenting the model's capacity to recognize inter-image dependencies. Secondly, the MPDIoU is a novel loss function to tackle the optimization challenge of bounding boxes with similar shapes but different sizes, which enhances the accuracy of localizing strawberry seedlings and nutrient pots. Finally, Distance Intersection over Union is utilized for establishing a belongingness relationship between strawberry seedlings and pots, accurately identifying missing seedlings and counting the corresponding pots. Experimental results demonstrate that SSP-MambaNet achieves 94.9 %in average precision, 92.8 % in recall rate,88.1 % in precision, and 90.4 % F1 score for strawberry seedlings and pots. It outperforms the YOLOv7 by 4.7 % in average precision, and 2.6 % in recall rate while reducing 66.7 f/s in FPS. Furthermore, the proposed method shows 94.29 % accuracy in detecting missing seedlings and 97.14 % accuracy in counting nutrient pots with missing seedlings. These results showcase its effectiveness in improving overall seedling quality and providing timely replanting guidance in glass greenhouses.
Why it matches plant phenotyping methods幼苗の欠損状態を画像から検出・計数する自動化手法の開発が研究の中心であり、植物の状態を直接推定しているため。
abstractwe present an automated method for detecting and counting missing seedlings based on SSP-MambaNet.
Reproduction assets foundThe paper's Data Availability statement explicitly provides the authors' source code and the strawberry seedling/nutrient pot image dataset via a public GitHub repository, matching an allowed URL.Code · publicThis study's source code and datasets can be accessed at https://github.com/STRABf5/SSPMambaNet.git.Open asset ↗STRABf5/SSPMambaNet · STRABf5/SSPMambaNethtml-lines:448-494Dataset · publicThis study's source code and datasets can be accessed at https://github.com/STRABf5/SSPMambaNet.git.Open asset ↗STRABf5/SSPMambaNet · STRABf5/SSPMambaNethtml-lines:538-610Plant phenotyping relevance match · UnverifiedEurope PMC · OpenAlex · Crossref · checked 15 Sept 2026
Precision agriculture methods can achieve the highest yield by applying the optimum amount of water, selecting appropriate pesticides, and managing crops in a way that minimises environmental impact. A rapidly emerging advanced research area, computer vision and deep learning, plays a significant role in effective crop management, such as superior genotype selection, plant classification, weed and pest detection, root localization, fruit counting and ripeness detection, and yield prediction. Also, phenotyping of plants involves analysing characteristics of plants such as chlorophyll content, leaf size, growth rate, leaf surface temperature, photosynthesis efficiency, leaf count, emergence time, shoot biomass, and germination time. This article presents an exhaustive study of recent techniques in computer vision and deep learning in plant science, with examples. The study provides the frequently used imaging parameters for plant image analysis with formulae, the most popular deep neural networks for plant classification and detection, object counting, and various applications. Furthermore, we discuss the publicly available plant image datasets for disease detection, weed control, and fruit detection with the evaluation metrics, tools and frameworks, future advancements and challenges in machine learning and deep learning models.
Why it matches plant phenotyping methods植物フェノタイピングにおけるコンピュータビジョン・深層学習手法、画像パラメータ、データセット、評価指標を体系的に扱うレビューであり、方法論が中心です。
abstractThis article presents an exhaustive study of recent techniques in computer vision and deep learning in plant science, with examples.
ABSTRACT The flowering date of sunflowers is a crucial trait that significantly influences crop management practices and product placement. Traditional ground methods for data collection are labor-intensive and subjective, requiring field scientists to manually estimate and record data in the field. This trait can be measured by counting the number of days from planting until 50% of plants in each research plot have reached flowering at R5 developmental growth stage. However, this method is time-consuming and may overlook valuable information related to flowering rates and duration. Flowering time of sunflower also can be approximated by counting the number of heads (flowers) across multiple dates. We propose a method for rapidly counting sunflower heads to model flower counts over time and estimate flowering time using RGB images acquired by Unmanned Aerial Vehicles (UAVs). The method developed employs a deep learning model trained to detect sunflower heads from UAV imagery and modeling these counts over time using a logistic function to estimate the 50% flowering date. The experimental results obtained from this method enabled estimation of the flowering date with a high correlation to ground measurements ( r > 0.91). Significantly, this approach not only reduces labor but also improves the precision of data collection. Moreover, an increase of 6% in heritability across trials, compared to traditional methods, suggests that our approach contributes to a deeper genetic understanding of flowering dynamics. This includes enhanced insights into the timing and rates of flowering, essential for optimizing breeding strategies and understanding genetic responses to environmental conditions. This innovative approach offers a promising avenue for enhancing the efficiency and accuracy of sunflower phenotyping.
Why it matches plant phenotyping methodsUAV画像からヒマワリ頭花を深層学習で検出し、開花時期という植物形質を推定する手法が研究の中心であり、地上測定との相関による検証も行っている。
abstractWe propose a method for rapidly counting sunflower heads to model flower counts over time and estimate flowering time using RGB images acquired by Unmanned Aerial Vehicles (UAVs).
Introduction Rice is one of the world's leading food crops, with nearly half of the world's population eating rice as their staple food. Rice yield is directly related to varieties, and the most intuitive agronomic trait of varietal yield is the number of grains per panicle. Methods In this study, rice panicles are taken as the research object, and images of the panicles are captured using a smartphone. The CSRNet counting model based on deep learning is then improved and applied to the problem of counting the number of grains per panicle in rice. Results and discussion The results show that the method of this study has a mean error value of 3.83% on the final validation set. On this basis, the development of rice per panicle counting APP based on Android terminal and batch counting software RiceGrainCounter based on PC terminal realizes real-time counting on Android terminal and batch counting on PC terminal, which can provide theoretical basis and technical support for rice per panicle counting.
Why it matches plant phenotyping methodsスマートフォン画像と改良CSRNetによりイネの穂当たり籾数を推定・計数する手法を開発し、検証するとともに、AndroidアプリとPCソフトウェアを実装しており、表現型取得が研究の中心である。
abstractimages of the panicles are captured using a smartphone. The CSRNet counting model based on deep learning is then improved and applied to the problem of counting the number of grains per panicle in rice.
Wheat spike detection holds significant importance for agricultural production as it enhances the efficiency of crop management and the precision of operations. This study aims to improve the accuracy and efficiency of wheat spike detection, enabling efficient crop monitoring under resource-constrained conditions. To this end, a wheat spike dataset encompassing multiple growth stages was constructed, leveraging the advantages of MobileNet and ShuffleNet to design a novel network module, SeCUIB. Building on this foundation, a new wheat spike detection network, LGWheatNet, was proposed by integrating a lightweight downsampling module (DWDown), spatial pyramid pooling (SPPF), and a lightweight detection head (LightDetect). The experimental results demonstrate that LGWheatNet excels in key performance metrics, including Precision, Recall, and Mean Average Precision (mAP50 and mAP50-95). Specifically, the model achieved a Precision of 0.956, a Recall of 0.921, an mAP50 of 0.967, and an mAP50-95 of 0.747, surpassing several YOLO models as well as EfficientDet and RetinaNet. Furthermore, LGWheatNet demonstrated superior resource efficiency with a parameter count of only 1,698,529 and GFLOPs of 5.0, significantly lower than those of competing models. Additionally, when combined with the Slicing Aided Hyper Inference strategy, LGWheatNet further improved the detection accuracy of wheat spikes, especially for small-scale targets and edge regions, when processing large-scale high-resolution images. This strategy significantly enhanced both inference efficiency and accuracy, making it particularly suitable for image analysis from drone-captured data. In wheat spike counting experiments, LGWheatNet also delivered exceptional performance, particularly in predictions during the filling and maturity stages, outperforming other models by a substantial margin. This study not only provides an efficient and reliable solution for wheat spike detection but also introduces innovative methods for lightweight object detection tasks in resource-constrained environments.
Why it matches plant phenotyping methods小麦穂の画像検出・計数という植物器官形質の抽出手法を開発し、データセット構築、モデル比較、精度・資源効率評価まで行っており、フェノタイピング手法が中心である。
abstractThis study aims to improve the accuracy and efficiency of wheat spike detection
Field / plotWhole plant / canopy / plot / fieldCountingSegmentationGrowth / development / phenology
Agriculture research is particularly essential since crop production is a challenge for farmers in India and around the world. 37% of the crop is impacted by invasive plants (weeds). Those unwelcome plants that interbreed with cultivated crops and decrease the purity of the crops are referred to here as weeds. A total of 2100 weed images were utilized to train the DCNN model in this study, including 500 images from the original dataset and 1600 images from the Crop Weed Field Image Dataset (CWFID), which includes broadleaf, a monocot, and dicot weeds. This research has proposed proposes hybrid Convolutional Neural Network models (HCNN) which have amalgamated the feature of the SegNet and U-Net CNN model for weed image segmentation. This work uses segmentation masks to exclude background and foreground vegetation to investigate weed growth and density estimation. To boost the identification weight of the weed leaf, furthermore, it has presented four distinct modified pooling layers and reduced the pooling layer of the classic segmentation model and loss function. According to the experimental results, our proposed algorithms achieved the best accuracy of 98.95%. The evaluation of financial misfortunes and impact due to weeds in farming is a critical perspective of considering which makes a difference in formulating suitable management methodologies against weeds.
Why it matches plant phenotyping methods雑草画像のセグメンテーションを基盤に、雑草の成長と密度を推定する深層学習手法を開発しており、植物状態の取得・抽出が研究の中心である。
abstractThis research has proposed proposes hybrid Convolutional Neural Network models (HCNN) which have amalgamated the feature of the SegNet and U-Net CNN model for weed image segmentation.
Counting soybean plants is a crucial strategy for assessing sowing quality and supporting high production. Despite its importance, the laborious nature of traditional assessment methods makes them unreliable and not scalable. Additionally, innovative image-based solutions have demonstrated limitations in detecting dense crops such as soybeans. Therefore, in this study, we developed neural network models to analyze a set of RGB and multispectral images and perform plant classification in a comprehensive dataset, which included data collected at three vegetative stages of soybean (VC, V1, and V2). Our results demonstrated high accuracy in classifying plants using either RGB (98%) or multispectral images (92%). A significant strength of this study is the ability to classify highly dense plants, without a trend for misclassification. Clearly, our findings provide stakeholders with a timely and effective approach to counting soybean plants, reducing labor and time, while increasing reliability.
Why it matches plant phenotyping methods大豆個体数をRGB・マルチスペクトル画像とニューラルネットワークで推定する手法を開発しており、植物フェノタイピング手法が研究の中心である。
abstractwe developed neural network models to analyze a set of RGB and multispectral images and perform plant classification
Counting soybean plants is a crucial strategy for assessing sowing quality and supporting high production. Despite its importance, the laborious nature of traditional assessment methods makes them unreliable and not scalable. Additionally, innovative image-based solutions have demonstrated limitations in detecting dense crops such as soybeans. Therefore, in this study, we developed neural network models to analyze a set of RGB and multispectral images and perform plant classification in a comprehensive dataset, which included data collected at three vegetative stages of soybean (VC, V1, and V2). Our results demonstrated high accuracy in classifying plants using either RGB (98%) or multispectral images (92%). A significant strength of this study is the ability to classify highly dense plants, without a trend for misclassification. Clearly, our findings provide stakeholders with a timely and effective approach to counting soybean plants, reducing labor and time, while increasing reliability.
Why it matches plant phenotyping methodsRGB・マルチスペクトル画像とニューラルネットワークによるダイズ個体数の推定手法を開発・評価しており、植物形質取得が研究の中心である。
abstractwe developed neural network models to analyze a set of RGB and multispectral images and perform plant classification
We present a novel method for soybean [ Glycine max (L.) Merr.] yield estimation leveraging high-throughput seed counting via computer vision and deep learning techniques. Traditional methods for collecting yield data are labor-intensive, costly, and prone to equipment failures at critical data collection times and require transportation of equipment across field sites. Computer vision, the field of teaching computers to interpret visual data, allows us to extract detailed yield information directly from images. By treating it as a computer vision task, we report a more efficient alternative, employing a ground robot equipped with fisheye cameras to capture comprehensive videos of soybean plots from which images are extracted in a variety of development programs. These images are processed through the P2PNet-Yield model, a deep learning framework, where we combined a feature extraction module (the backbone of the P2PNet-Soy) and a yield regression module to estimate seed yields of soybean plots. Our results are built on 2 years of yield testing plot data-8,500 plots in 2021 and 650 plots in 2023. With these datasets, our approach incorporates several innovations to further improve the accuracy and generalizability of the seed counting and yield estimation architecture, such as the fisheye image correction and data augmentation with random sensor effects. The P2PNet-Yield model achieved a genotype ranking accuracy score of up to 83%. It demonstrates up to a 32% reduction in time to collect yield data as well as costs associated with traditional yield estimation, offering a scalable solution for breeding programs and agricultural productivity enhancement.
Why it matches plant phenotyping methodsロボット動画とコンピュータビジョン／深層学習によりダイズの種子収量を推定する手法を開発・評価しており、表現型取得・推定が研究の中心である。
abstractWe present a novel method for soybean [ Glycine max (L.) Merr.] yield estimation leveraging high-throughput seed counting via computer vision and deep learning techniques.
Ground control points (GCPs) are used in forest surveys employing unmanned aerial vehicle (UAV)-based structure from motion (SfM). In that context, the influence of the surrounding environment on GCP placement requires further analysis. This study investigated the effects of GCP placement and the surrounding environment on the estimation of forest information by UAV-SfM. Forest resource estimation was performed using UAV (Inspire2) aerial images and SfM analysis (via Pix4Dmapper) under varying environmental conditions around GCPs within the same forest stand. The results indicated that GCP placement had no significant effect on SfM processing, tree top extraction (the number of extracted target trees was 151 or 150), or tree crown area estimation (RMSEs ranged from approximately 5 to 6.5 m2). However, when GCPs were placed in open areas, the tree height estimation accuracy improved, without significant differences between estimated and measured values (patterns A, B, D and E, had RMSEs of 1.60 to 3.09 m; patterns C and D had RMSEs of 5.69 to 7.92 m). These findings suggest that in UAV-SfM-based forest resource surveys, particularly for tree height estimation, both the number and placement of GCPs, as well as the surrounding environment, are crucial in enhancing estimation accuracy.
Why it matches plant phenotyping methodsUAV-SfMによる樹高・樹冠面積推定について、GCP配置と環境条件が推定精度に与える影響を比較評価しており、植物形質取得法の技術的検証が中心である。
abstractThis study investigated the effects of GCP placement and the surrounding environment on the estimation of forest information by UAV-SfM.
Background Seed testing plays a crucial role in improving crop yields.In actual seed testing processes, factors such as grain sticking and complex imaging environments can significantly affect the accuracy of wheat grain counting, directly impacting the effectiveness of seed testing. However, most existing methods primarily focus on simple counting tasks and lack general applicability. Results To enable fast and accurate counting of wheat grains under severe adhesion and complex scenarios, this study collected images of wheat grains from different varieties, backgrounds, densities, imaging heights, adhesion levels, and other natural conditions using various imaging devices and constructed a comprehensive wheat grain dataset through data enhancement techniques. We propose a wheat grain detection and counting model called GrainNet, which significantly improves the counting performance and detection speed across diverse conditions and adhesion levels by incorporating lightweight and efficient feature fusion modules. Specifically, the model incorporates an Efficient Multi-scale Attention (EMA) mechanism, effectively mitigating the interference of background noise on detection results. Additionally, the ASF-Gather and Distribute (ASF-GD) module optimizes the feature extraction component of the original YOLOv7 network, improving the model's robustness and accuracy in complex scenarios. Ablation experiments validate the effectiveness of the proposed methods.Compared with classic models such as Faster R-CNN, YOLOv5, YOLOv7, and YOLOv8, the GrainNet model achieves better detection performance and computational efficiency in various scenarios and adhesion levels. The mean Average Precision reached 93.15%, the F1 score was 0.946, and the detection speed was 29.10 frames per second (FPS). A comparative analysis with manual counting results revealed that the GrainNet model achieved the highest coefficient of determination and Mean Absolute Error values for wheat grain counting tasks, which were 0.93 and 5.97, respectively, with a counting accuracy of 94.47%. Conclusions Overall, the GrainNet model presented in this study enables accurate and rapid recognition and quantification of wheat grains, which can provide a reference for effective seed examination of wheat grains in real scenarios. Related content can be accessed through the following link: https://github.com/1371530728/grainnet.git .
Why it matches plant phenotyping methods小麦粒の画像検出・計数モデルとデータセットを開発し、複数条件で性能検証しているため、植物器官形質の取得手法が中心である。
abstractWe propose a wheat grain detection and counting model called GrainNet
Accurate soybean pod counting remains a significant challenge in field-based phenotyping due to complex factors such as occlusion, dense distributions, and background interference. We present SmartPod, an advanced deep learning framework that addresses these challenges through three key innovations: (1) a novel vision Transformer architecture for enhanced feature representation, (2) an efficient attention mechanism for the improved detection of overlapping pods, and (3) a semi-supervised learning strategy that maximizes performance with limited annotated data. Extensive evaluations demonstrate that SmartPod achieves state-of-the-art performance with an Average Precision at an IoU threshold of 0.5 (AP@IoU = 0.5) of 94.1%, outperforming existing methods by 1.7–4.6% across various field conditions. This significant improvement, combined with the framework’s robustness in complex environments, positions SmartPod as a transformative tool for large-scale soybean phenotyping and precision breeding applications.
Why it matches plant phenotyping methods大豆莢数を圃場画像から自動抽出する深層学習フレームワークの開発・評価であり、植物フェノタイピング手法が中心です。
abstractWe present SmartPod, an advanced deep learning framework that addresses these challenges through three key innovations
In the process of smart breeding, the rapid statistics of soybean emergence rate, as an important part of breeding screening, face challenges under environmental constraints, especially the selection and breeding of soybean varieties in dense environments. Due to the influence of environmental factors, the existing methods have shortcomings, such as low throughput, low efficiency, and insufficient precision. Therefore, an effective and precise statistical method is required. In this study, UAV (Unmanned Aerial Vehicle)-scale data combined with ground measurement data were used as the research object to explore the feasibility of improving the throughput, efficiency, and accuracy of breeding screening under intensive soybean planting. To this end, a set of technical solutions, including background removal, object detection, and accurate counting, were designed. Firstly, a combined background segmentation method based on contrast enhancement filtering combined with ultra-green eigenvalues and the Otsu algorithm was proposed to remove the complex background in remote sensing images and retain the morphological information of soybean seedlings. Secondly, the deep learning object detection model was used to infer and predict the processed images to label soybean seedlings. Then, a soybean seedling counting algorithm was constructed: by establishing a soybean seedling growth model, the idea of "growth normalization" was proposed, and the expansion-compression factor was defined to eliminate the influence of soybean seedling growth inconsistency on counting. After statistical and in-depth analysis of the growth and planting characteristics of soybean seedlings under overlapping conditions, the "inter-seedling occlusion counting algorithm" was proposed to solve the problem of overlapping counting between seedlings. In order to solve the problem of an overlapping bounding box, a soft strategy is specially designed to avoid the redundant values brought by it. Finally, according to the calculation results, the statistical thematic map of soybean emergence rate based on plot plots was displayed. After experiments, the proposed method can effectively count the number of soybean seedlings in the image, with an overall accuracy of 99.18% and an error rate of 0.82%. In addition, Yolov8n had the best recognition effect in the soybean seedling detection task, with a mAP (0.5-0.95) of 85.15%. The proposed background segmentation method increased the mAP (0.5-0.95) of the detection results by 4.06%. It has been demonstrated through experimental tests and verifications that solid support for the statistical work concerning the soybean emergence rate under the condition of intensive planting is provided by this method. This innovative method has played a facilitating role in accelerating the breeding process and has also provided some new ideas and reference directions for further exploration of efficient screening.
Why it matches plant phenotyping methods大豆苗の出芽率を高スループットに推定するため、画像分割・物体検出・重複個体計数を開発し、精度検証まで行った植物フェノタイピング手法研究である。
abstracta set of technical solutions, including background removal, object detection, and accurate counting, were designed.
Crop yield estimation is a relevant problem in agriculture, because an accurate yield estimate can support farmers' decisions on harvesting or precision intervention. Robots can help to automate this process. To do so, they need to be able to perceive the surrounding environment to identify target objects such as trees and plants. In this paper, we introduce a novel approach to address the problem of hierarchical panoptic segmentation of apple orchards on 3D data from different sensors. Our approach is able to simultaneously provide semantic segmentation, instance segmentation of trunks and fruits, and instance segmentation of trees (a trunk with its fruits). This allows us to identify relevant information such as individual plants, fruits, and trunks, and capture the relationship among them, such as precisely estimate the number of fruits associated to each tree in an orchard. To efficiently evaluate our approach for hierarchical panoptic segmentation, we provide a dataset designed specifically for this task. Our dataset is recorded in Bonn, Germany, in a real apple orchard with a variety of sensors, spanning from a terrestrial laser scanner to a RGB-D camera mounted on different robots platforms. The experiments show that our approach surpasses state-of-the-art approaches in 3D panoptic segmentation in the agricultural domain, while also providing full hierarchical panoptic segmentation. Our dataset is publicly available at https://www.ipb.uni-bonn.de/data/hops/. The open-source implementation of our approach is available at https://github.com/PRBonn/hapt3D.
Why it matches plant phenotyping methodsリンゴ樹・果実・幹を3Dセグメンテーションし、樹ごとの果実数を推定する手法と専用データセットを中心に開発・評価しており、植物の器官形態・収量関連形質の取得に該当する。
abstractwe introduce a novel approach to address the problem of hierarchical panoptic segmentation of apple orchards on 3D data from different sensors.
Reproduction assets foundThe paper introduces the HOPS dataset of annotated 3D apple orchard point clouds (TLS, UAV, UGV, SfM) for hierarchical panoptic segmentation, publicly available at the authors' IPB Bonn page, and releases the open-source implementation (hapt3D) on GitHub. Both are paper-specific, public, and actionable.Code · publicThe open-source implementation of our approach is available at https://github.com/PRBonn/hapt3D .Open asset ↗PRBonn/hapt3Dlines:1-59Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
Published12 Mar 2025Investigaciones Geográficas Boletín del Instituto de GeografíaCited by 1 · OpenAlex ↗
This study employs photogrammetry to evaluate and monitor the recovery of the cypress forest on Guadalupe Island, Mexico, an ecosystem significantly impacted by fires and overgrazing. Two drone surveys were conducted over the forest area during the summers of 2016 and 2019 using natural color (RGB) and near-infrared (NIR) cameras. This work presents the first complete 3D reconstruction of the cypress forest on the island. The image processing products include the canopy height model (CHM), digital surface model (DSM), and digital terrain model (DTM), which were utilized to calculate the number, density, height and crown projected areas of trees. The CHM showed a high correlation with the forest's structure (R = 0.92), based on field measurements of tree heights. Our study accounted for approximately 67,340 trees taller than two meters in 2019. Over 90% of the cypress population consisted of young trees between 2 and 3 meters tall, which have recovered significantly following a fire in 2008 that burned 70% of its extent. A horizontal expansion of 134 hectares was observed from 2016 to 2019 in the regeneration process.
Why it matches plant phenotyping methodsドローン画像のフォトグラメトリによる3D再構成を用いて樹木の高さ・密度・樹冠面積を推定し、現地測定との相関で検証しているため、植物形質の取得手法が中心です。
abstractThis study employs photogrammetry to evaluate and monitor the recovery of the cypress forest on Guadalupe Island, Mexico
Reproduction assets foundThe paper's photogrammetric phenotyping products (2016/2019 point clouds, orthomosaics, DSMs, CHMs) are publicly downloadable via a DOI data repository, and supplemental crown/treetop features are in CICESE's institutional repository. Both URLs appear in allowed_urls.Dataset · publicees. This phenomenon can be seen
in the three years observation window (2016-2019)
using photogrammetry.
AVAILABILITY OF DATA
AND MATERIALS
Point clouds from the 2016 and 2019 photogram-
metric reconstructions, as well as orthomosaics,
digital surface models (DSMs), and canopy height
models (CHMs), are available for download in
https://doi.org/10.5069/G9668BDD and https://
doi.org/10.5069/G92J693D. Supplemental infor-
mation such as Features related to crown and tree-
tops are accessible through CICESE’s institutional
repository (https://repositoriobiblioteca.cicese.mx/jspui/handle/123456789/44)
REFERENCES
Aljos-Farjon. (2017). A handbook of the world’s conifers
(second ed., vol. 1).Open asset ↗10.5069/G9668BDD · 10.5069/G9668BDDpdf-raw-page:15 lines:1-89Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Growing consumer demand for high-quality strawberries has highlighted the need for accurate, efficient, and non-destructive methods to assess key postharvest quality traits, such as weight, size uniformity, and quantity. This study proposes a multi-objective learning algorithm that leverages RGB-D multimodal information to estimate these quality metrics. The algorithm develops a fusion expert network architecture that maximizes the use of multimodal features while preserving the distinct details of each modality. Additionally, a novel Heritable Loss function is implemented to reduce redundancy and enhance model performance. Experimental results show that the coefficient of determination (R²) values for weight, size uniformity and number are 0.94, 0.90 and 0.95 respectively. Ablation studies demonstrate the advantage of the architecture in multimodal, multi-task prediction accuracy. Compared to single-modality models, non-fusion branch networks, and attention-enhanced fusion models, our approach achieves enhanced performance across multi-task learning scenarios, providing more precise data for trait assessment and precision strawberry applications.
Why it matches plant phenotyping methodsRGB-D画像を用いてイチゴ果実の重量・サイズ均一性・個数を推定する融合ネットワークの開発と性能評価が研究の中心であり、植物器官の形質取得手法に該当する。
abstractThis study proposes a multi-objective learning algorithm that leverages RGB-D multimodal information to estimate these quality metrics.
Understanding plant growth dynamics is essential for applications in agriculture and plant phenotyping. We present the Growth Modelling (GroMo) challenge, which is designed for two primary tasks: (1) plant age prediction and (2) leaf count estimation, both essential for crop monitoring and precision agriculture. For this challenge, we introduce GroMo25, a dataset with images of four crops: radish, okra, wheat, and mustard. Each crop consists of multiple plants (p1, p2, ..., pn) captured over different days (d1, d2, ..., dm) and categorized into five levels (L1, L2, L3, L4, L5). Each plant is captured from 24 different angles with a 15-degree gap between images. Participants are required to perform both tasks for all four crops with these multiview images. We proposed a Multiview Vision Transformer (MVVT) model for the GroMo challenge and evaluated the crop-wise performance on GroMo25. MVVT reports an average MAE of 7.74 for age prediction and an MAE of 5.52 for leaf count. The GroMo Challenge aims to advance plant phenotyping research by encouraging innovative solutions for tracking and predicting plant growth. The GitHub repository is publicly available at https://github.com/mriglab/GroMo-Plant-Growth-Modeling-with-Multiview-Images.
Why it matches plant phenotyping methods植物のマルチビュー画像から葉数と植物齢を推定するデータセット・ベンチマークおよびモデルを提示しており、植物表現型取得・推定が中心である。
abstractWe present the Growth Modelling (GroMo) challenge, which is designed for two primary tasks: (1) plant age prediction and (2) leaf count estimation
Accurate tomato yield estimation and ripeness monitoring are critical for optimizing greenhouse management. While manual counting remains labor-intensive and error-prone, this study introduces a novel vision-based framework for automated tomato counting in standardized greenhouse environments. The proposed method integrates YOLOv8-based detection, depth filtering, and an inter-frame prediction algorithm to address key challenges such as background interference, occlusion, and double-counting. Our approach achieves 97.09% accuracy in tomato cluster detection, with mature and immature single-fruit recognition accuracies of 92.03% and 91.79%, respectively. The multi-target tracking algorithm demonstrates a MOTA (Multiple Object Tracking Accuracy) of 0.954, outperforming conventional methods like YOLOv8+DeepSORT. By fusing odometry data from an inspection robot, this lightweight solution enables real-time yield estimation and maturity classification, offering practical value for precision agriculture.
Why it matches plant phenotyping methods温室ロボット向けの画像解析フレームワークを中心に、トマト果実の計数、成熟度分類、収量推定という植物器官・状態の定量手法を開発・評価しているため。
abstractthis study introduces a novel vision-based framework for automated tomato counting in standardized greenhouse environments.
Downy mildew is a critical disease in viticulture, typically identified through manual inspection of individual leaves in the field by experts. The combination of artificial intelligence techniques with mobile platforms can optimise non-invasive detection. This work focused on employing semantic segmentation deep neural networks to detect visual symptoms of downy mildew in high-resolution grapevine images under field conditions. Vineyard canopy images were collected from 14 plots using both manual and mobile platform methods. The study compared six architectures and six encoders using transfer learning, as well as two SegNet AdHoc architectures. To address imbalance problems, simple data augmentation, MixUp, oversampling, and undersampling techniques were employed. The results were adjusted through test-time augmentation. The study found that the U-Net architecture, using the MobileVit-S encoder and the Dice loss function, was particularly efficient. The U-Net architecture with light-weight encoders exhibited potential for real-time applications. The robustness of the model was improved by combining oversampling and undersampling with simple data augmentation during training. The classification of areas with and without disease symptoms achieved an accuracy of 86% and an f1-score of 82%. Additionally, the number of symptoms in grapevine canopy images was detected with an NRMSE of 12%. In conclusion, the proposed methodology shows promise for efficiently early assessing grapevine downy mildew under field conditions. This approach could be applied to other crop diseases and pests, taking advantage of the complexity of the dataset to strengthen the robustness of the model in real-world scenarios.
Why it matches plant phenotyping methodsブドウ葉・キャノピー画像から病徴領域と症状数を推定するセマンティックセグメンテーション手法が研究の中心であり、植物病害状態の表現型計測と技術評価に該当する。
abstractThis work focused on employing semantic segmentation deep neural networks to detect visual symptoms of downy mildew in high-resolution grapevine images under field conditions.
Machine learning models for crop image analysis and phenomics are highly important for precision agriculture and breeding and have been the subject of intensive research. However, the lack of publicly available high-quality image datasets with detailed annotations has severely hindered the development of these models. In this work, we present a comprehensive multicultivar and multiview rice plant image dataset (CVRP) created from 231 landraces and 50 modern cultivars grown under dense planting in paddy fields. The dataset includes images capturing rice plants in their natural environment, as well as indoor images focusing specifically on panicles, allowing for a detailed investigation of cultivar-specific differences. A semiautomatic annotation process using deep learning models was designed for annotations, followed by rigorous manual curation. We demonstrated the utility of the CVRP by evaluating the performance of four state-of-the-art (SOTA) semantic segmentation models. We also conducted 3D plant reconstruction with organ segmentation via images and annotations. The database not only facilitates general-purpose image-based panicle identification and segmentation but also provides valuable resources for challenging tasks such as automatic rice cultivar identification, panicle and grain counting, and 3D plant reconstruction. The database and the model for image annotation are available at https://bic.njau.edu.cn/CVRP.html.
Why it matches plant phenotyping methodsイネ画像データセットとアノテーションモデルを開発・評価し、セグメンテーション、器官再構成、穂・粒数計測などの再利用可能な表現型解析を中心に扱っているため。
abstractwe present a comprehensive multicultivar and multiview rice plant image dataset (CVRP)
Reproduction assets foundThe paper's own CVRP rice image dataset (images + annotations), accompanying code, and trained Mask2Former annotation model are explicitly stated as publicly available on Hugging Face and the authors' NJAU site.Dataset · publicThe CVRP dataset is publicly available on Hugging Face at https://huggingface.co/datasets/CVRPDataset/CVRP for academic use under the specified license.Open asset ↗CVRPDataset/CVRPhtml-lines:236-252Code · publicThe accompanying code and trained models are available at https://huggingface.co/CVRPDataset/Model.Open asset ↗CVRPDataset/Modelhtml-lines:236-252Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Location-based methods for counting rice panicles have often been underestimated, primarily due to their perceived inferior performance when compared to detection-based techniques. However, we argue that the potential of these location-based methods has not been fully realized, largely owing to the limitations of existing model architectures. In response to this challenge, we introduce LKNet, an innovative model developed on the foundation of the location-based framework P2Pnet. To enhance the performance of panicle counting across diverse types and growth stages, we implemented several key strategies. Firstly, we reconstructed the localization loss function as a predictive probability distribution to reduce the influence of manual labeling. Additionally, we dynamically adapted the receptive field to better accommodate different panicle types through the use of large kernel convolutional blocks. We evaluated LKNet on several publicly available counting task datasets and achieved state-of-the-art performance on the Diverse Rice Panicle Detection dataset. Furthermore, we employed a rice panicle dataset collected at an altitude of 7 m, which includes various panicle types and growth stages for model training and evaluation. The results showed that LKNet effectively accommodates variations in panicle morphology, with R 2 values ranging from 0.903 to 0.989. These findings highlight LKNet's potential to enhance precision in panicle counting in rice breeding programs.
Why it matches plant phenotyping methodsイネ穂の画像ベース計数モデルを開発し、複数データセットで評価しており、植物表現型取得・抽出手法が中心である。
abstractwe introduce LKNet, an innovative model developed on the foundation of the location-based framework P2Pnet.
Reproduction assets foundThe paper's authors explicitly state that the LKNet analysis code is publicly available on GitHub, matching an allowed URL. No separate phenotype dataset deposit by the authors is stated (the 7 m rice panicle dataset and public benchmarks like DPRD/MTC/SHTech are described but no authors' dataset URL is given).Code · publicCode is available at https://github.com/L129921/LKnet .Open asset ↗LKnetlines:297-319Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
This study systematically performed an extensive real-world evaluation of the performances of all configurations of YOLOv8, YOLOv9, YOLOv10, YOLO11( or YOLOv11), and YOLOv12 object detection algorithms in terms of precision, recall, mean Average Precision at 50\% Intersection over Union (mAP@50), and computational speeds including pre-processing, inference, and post-processing times immature green apple (or fruitlet) detection in commercial orchards. Additionally, this research performed and validated in-field counting of the fruitlets using an iPhone and machine vision sensors. Among the configurations, YOLOv12l recorded the highest recall rate at 0.90, compared to all other configurations of YOLO models. Likewise, YOLOv10x achieved the highest precision score of 0.908, while YOLOv9 Gelan-c attained a precision of 0.903. Analysis of mAP@0.50 revealed that YOLOv9 Gelan-base and YOLOv9 Gelan-e reached peak scores of 0.935, with YOLO11s and YOLOv12l following closely at 0.933 and 0.931, respectively. For counting validation using images captured with an iPhone 14 Pro, the YOLO11n configuration demonstrated outstanding accuracy, recording RMSE values of 4.51 for Honeycrisp, 4.59 for Cosmic Crisp, 4.83 for Scilate, and 4.96 for Scifresh; corresponding MAE values were 4.07, 3.98, 7.73, and 3.85. Similar performance trends were observed with RGB-D sensor data. Moreover, sensor-specific training on Intel Realsense data significantly enhanced model performance. YOLOv11n achieved highest inference speed of 2.4 ms, outperforming YOLOv8n (4.1 ms), YOLOv9 Gelan-s (11.5 ms), YOLOv10n (5.5 ms), and YOLOv12n (4.6 ms), underscoring its suitability for real-time object detection applications.
Why it matches plant phenotyping methods果実数という植物器官形質の画像ベース推定を対象に、複数YOLOモデルの性能比較とiPhone・RGB-Dセンサーによる圃場カウント検証を中心的に行っているため。
abstractThis study systematically performed an extensive real-world evaluation of the performances of all configurations of YOLOv8, YOLOv9, YOLOv10, YOLO11( or YOLOv11), and YOLOv12 object detection algorithms
Accurate counting of Amorphophallus konjac (Konjac) plants can offer valuable insights for agricultural management and yield prediction. While current studies have primarily focused on detecting and counting crop plants during the early stages of low coverage, there is limited investigation into the later stages of high coverage, which could impact the accuracy of forecasting yield. High canopy coverage and severe occlusion in later stages pose significant challenges for plant detection and counting. Therefore, this study evaluated the performance of the Count Crops tool and a deep learning (DL) model derived from early-stage unmanned aerial vehicle (UAV) imagery in detecting and counting Konjac plants during the high-coverage growth stage. Additionally, the study proposed an approach that integrates the DL model with Konjac location information from both early-stage and high canopy coverage stage imagery to improve the accuracy of recognizing Konjac plants during the high canopy coverage stage. The results indicated that the Count Crops tool outperformed the DL model constructed solely from early-stage imagery in detecting and counting Konjac plants during the high-coverage period. However, given the single stem and erect growth characteristics of Konjac, incorporating the DL model with the location information of the Konjac plants achieved the highest accuracy (Precision = 98.7%, Recall = 86.7%, F1-score = 92.3%). Our findings indicate that combining DL detection results from the early growth stages of Konjac, along with plant positional information from both growth stages, not only significantly improved the accuracy of detecting and counting plants but also saved time on annotating and training DL samples in the later stages. This study introduces an innovative approach for detecting and counting Konjac plants during high-coverage periods, providing a new perspective for recognizing and counting other crop plants at high-overlapping growth stages.
Why it matches plant phenotyping methodsUAV画像と深層学習を用いて植物個体の検出・計数手法を評価・改良しており、植物表現型の取得方法が研究の中心です。
abstractthis study evaluated the performance of the Count Crops tool and a deep learning (DL) model derived from early-stage unmanned aerial vehicle (UAV) imagery in detecting and counting Konjac plants during the high-coverage growth stage.
Introduction Quantitative wood anatomy is critical for establishing climate reconstruction proxies, understanding tree hydraulics, and quantifying carbon allocation. Its accuracy depends upon the image acquisition methods, which allows for the identification of the number and dimensions of vessels, fibres, and tracheids within a tree ring. Angiosperm wood is analysed with a variety of different image acquisition methods, including surface pictures, wood anatomical micro-sections, or X-ray computed micro-tomography. Despite known advantages and disadvantages, the quantitative impact of method selection on wood anatomical parameters is not well understood. Methods In this study, we present a systematic uncertainty analysis of the impact of the image acquisition method on commonly used anatomical parameters. We analysed four wood samples, representing a range of wood porosity, using surface pictures, micro-CT scans, and wood anatomical micro-sections. Inter-annual patterns were analysed and compared between methods from the five most frequently used parameters, namely mean lumen area ( MLA ), vessel density ( VD ), number of vessels ( VN ), mean hydraulic diameter ( D h ), and relative conductive area ( RCA ). A novel sectorial approach was applied on the wood samples to obtain intra-annual profiles of the lumen area ( A l ), specific theoretical hydraulic conductivity ( K s ), and wood density ( ρ ). Results Our quantitative vessel mapping revealed that values obtained for hydraulic wood anatomical parameters are comparable across different methods, supporting the use of easily applicable surface picture methods for ring-porous and specific diffuse-porous tree species. While intra-annual variability is well captured by the different methods across species, wood density ( ρ ) is overestimated due to the lack of fibre lumen area detection. Discussion Our study highlights the potential and limitations of different image acquisition methods for extracting wood anatomical parameters. Moreover, we present a standardized workflow for assessing radial tree ring profiles. These findings encourage the compilation of all studies using wood anatomical parameters and further research to refine these methods, ultimately enhancing the accuracy, replication, and spatial representation of wood anatomical studies.
Why it matches plant phenotyping methods木材解剖学的形質を抽出する画像取得法を比較・不確実性分析し、標準化ワークフローも提示しており、植物フェノタイピング手法が中心です。
abstractwe present a systematic uncertainty analysis of the impact of the image acquisition method on commonly used anatomical parameters.
Wheat ( Triticum aestivum L.) is one of the significant food crops in the world, and the number of wheat ears serves as a critical indicator of wheat yield. Accurate quantification of wheat ear counts is crucial for effective scientific management of wheat fields. To address the challenges of missed detections, false detections, and diminished detection accuracy arising from the dense distribution, small size, and high overlap of wheat ears in Unmanned Aerial Vehicle (UAV) imagery, we propose a lightweight model, PSDS-YOLOv8 (P2-SPD-DySample-SCAM-YOLOv8), on the basis of the improved YOLOv8 framework, for the accurate detection of wheat ears in UAV images. First, the high resolution micro-scale detection layer (P2) is introduced to enhance the model's ability to recognize and localize small targets, while the large-scale detection layer (P5) is eliminated to minimize computational redundancy. Then, the Spatial Pyramid Dilated Convolution (SPD-Conv) module is employed to improve the ability of the network to learn features, thereby enhancing the representation of weak features of small targets and preventing information loss caused by low image resolution or small target sizes. Additionally, a lightweight dynamic upsampler, Dynamic Sample (DySample), is introduced to decrease computational complexity of the upsampling process by dynamically adjusting interpolation positions. Finally, the lightweight module Spatial Context-Aware Module (SCAM) is utilized to accurately map the connection between small targets and global features, enhancing the discrimination of small targets from the background. Experimental results demonstrate that the improved PSDS-YOLOv8 model achieves Mean Average Precision(mAP) 50 and mAP50:95 scores of 96.5% and 55.2%, which increases by 2.8% and 4.4%, while the number of parameters is reduced by 40.6% in comparison with the baseline YOLOv8 model. Compared to YOLOv5, YOLOv7, YOLOv9, YOLOv10, YOLOv11, Faster RCNN, SSD, and RetinaNet, the improved model demonstrates superior accuracy and fewer parameters, exhibiting the best overall performance. The methodology proposed in this study enhances model accuracy while concurrently reducing resource consumption and effectively addressing the issues of missed and false detections of wheat ears, thereby providing technical support and theoretical guidance for intelligent counting of wheat ears in UAV imagery.
Why it matches plant phenotyping methodsUAV画像から小麦穂数という植物形態・収量関連形質を抽出する改良YOLOモデルを開発し、複数モデルとの性能比較で技術的に検証しているため、方法開発が中心である。
abstractwe propose a lightweight model, PSDS-YOLOv8 (P2-SPD-DySample-SCAM-YOLOv8), on the basis of the improved YOLOv8 framework, for the accurate detection of wheat ears in UAV images.
The soybean-cyst nematode (SCN; Heterodera glycines) is one of the most destructive pests affecting soybean crops. Effective management of SCN is imperative for the sustainability of soybean agriculture. A promising approach to achieving this goal is the development and breeding of new resistant soybean varieties. Researchers and breeders typically employ exploratory methods such as Genome-Wide Association Studies or Quantitative Trait Loci mapping to identify genes linked to resistance. These methods depend on extensive phenotypic screening. The primary phenotypic measure for assessing SCN resistance is often the number of cysts that form on a plant's root system. Manual counting hundreds of cysts on a given root system is not only laborious but also subject to variability due to individual assessor differences. Additionally, while measuring cyst size could provide valuable insights due to its correlation with cyst development, this aspect is frequently overlooked because it demands even more hands-on work. To address these challenges, we have created Nemacounter, an intuitive software designed to detect, count, and measure the size of cysts autonomously. Nemacounter boasts a user-friendly graphical interface, simplifying the process for users to obtain reliable results. It enhances productivity by delivering annotated images and compiling data into csv files for easy analysis and reporting.
Why it matches plant phenotyping methodsダイズ根上の線虫シスト数とサイズという植物病害抵抗性関連形質を、画像から自動検出・計測するソフトウェアを開発しており、表現型取得手法が研究の中心です。
abstractwe have created Nemacounter, an intuitive software designed to detect, count, and measure the size of cysts autonomously.
Reproduction assets foundThe paper's SCN cyst phenotyping assets are publicly available: the authors' Nemacounter analysis software on GitHub, two annotated cyst image datasets on Roboflow (bounding-box and segmentation/area annotations), and the authors' trained YOLOv5-xl model (cystmodel.pt) on Iowa State's Box. The SAM weights and ultralyptCode · publicThe Nemacounter software can be downloaded here: https://github.com/DjampaKozlowski/NemaCounter and we provide an installation manual and utilization manual as supplementary data.Open asset ↗DjampaKozlowski/NemaCounterlines:65-70Dataset · publicThe complete dataset is accessible on the Roboflow website at: https://universe.roboflow.com/iowa-state-university-cwvqa/cystnewboundingboxv2Open asset ↗lines:118-138Dataset · publicAll training datasets are available on Roboflow website at : https://universe.roboflow.com/iowa-state-university-cwvqa/cystnewboundingboxv2 and https://universe.roboflow.com/iowa-state-university-cwvqa/cyst-detectors-area.Open asset ↗lines:139-197Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Viticulture benefits significantly from rapid grape bunch identification and counting, enhancing yield and quality. Recent technological and machine learning advancements, particularly in deep learning, have provided the tools necessary to create more efficient, automated processes that significantly reduce the time and effort required for these tasks. On one hand, drone, or Unmanned Aerial Vehicles (UAV) imagery combined with deep learning algorithms has revolutionised agriculture by automating plant health classification, disease identification, and fruit detection. However, these advancements often remain inaccessible to farmers due to their reliance on specialized hardware like ground robots or UAVs. On the other hand, most farmers have access to smartphones. This article proposes a novel approach combining UAVs and smartphone technologies. An AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training. By leveraging UAV-captured data for training, the proposed model not only accelerates the detection process but also enhances the accuracy and adaptability of grape bunch detection across different devices, surpassing the efficiency of traditional and purely UAV-based methods. To this end, using a dataset of UAV videos recorded during early growth stages in July (BBCH77-BBCH79), the X-Decoder segments vegetation in the front of the frames from their background and surroundings. X-Decoder is particularly advantageous because it can be seamlessly integrated into the AI pipeline without requiring changes to how data is captured, making it more versatile than other methods. Then, YOLO is trained using the videos and further applied to images taken by farmers with common smartphones (Xiaomi Poco X3 Pro and iPhone X). In addition, a web app was developed to connect the system with mobile technology easily. The proposed approach achieved a precision of 0.92 and recall of 0.735, with an F1 score of 0.82 and an Average Precision (AP) of 0.802 under different operation conditions, indicating high accuracy and reliability in detecting grape bunches. In addition, the AI-detected grape bunches were compared with the actual ground truth, achieving an R 2 value as high as 0.84, showing the robustness of the system. This study highlights the potential of using smartphone imaging and web applications together, making an effort to integrate these models into a real platform for farmers, offering a practical, affordable, accessible, and scalable solution. While smartphone-based image collection for model training is labour-intensive and costly, incorporating UAV data accelerates the process, facilitating the creation of models that generalise across diverse data sources and platforms. This blend of UAV efficiency and smartphone precision significantly cuts vineyard monitoring time and effort.
Why it matches plant phenotyping methodsスマートフォン画像とUAVデータを用いてブドウ房を検出・計数するAIパイプラインを開発・評価し、実測値との比較も行っているため、植物器官形質の取得手法が中心である。
abstractAn AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training.
Reproduction assets foundThe paper's Data Availability Statement points to a public, paper-specific dataset (EscaYard: geotagged smartphone vineyard images, phytosanitary status, UAV 3D point clouds and orthomosaics) published as a Data Brief with a DOI, which directly underpins the smartphone/UAV grape detection phenotyping analysis. No code,Dataset · publicData is available at https://doi.org/10.1016/j.dib.2024.110497 [ 55 ].Open asset ↗10.1016/j.dib.2024.110497lines:202-204Code / dataset availability confirmedEurope PMC · checked 6 Sept 2026
Accurate counting of crop plants is essential for agricultural science, particularly for yield forecasting, field management, and experimental studies. Traditional methods are labor-intensive and prone to errors. Unmanned Aerial Vehicle (UAV) technology offers a promising alternative; however, varying UAV altitudes can impact image quality, leading to blurred features and reduced accuracy in early maize seedling counts. To address these challenges, we developed RC-Dino, a deep learning methodology based on DINO, specifically designed to enhance the precision of seedling counts from UAV-acquired images. RC-Dino introduces two innovative components: a novel self-calibrating convolutional layer named RSCconv and an adaptive spatial feature fusion module called ASCFF. The RSCconv layer improves the representation of early maize seedlings compared to non-seedling elements within feature maps by calibrating spatial domain features. The ASCFF module enhances the discriminability of early maize seedlings by adaptively fusing feature maps extracted from different layers of the backbone network. Additionally, transfer learning was employed to integrate pre-trained weights with RSCconv, facilitating faster convergence and improved accuracy. The efficacy of our approach was validated using the Early Maize Seedlings Dataset (EMSD), comprising 1,233 annotated images of early maize seedlings, totaling 83,404 individual annotations. Testing on this dataset demonstrated that RC-Dino outperformed existing models, including DINO, Faster R-CNN, RetinaNet, YOLOX, and Deformable DETR. Specifically, RC-Dino achieved improvements of 16.29% in Average Precision (AP) and 8.19% in Recall compared to the DINO model. Our method also exhibited superior coefficient of determination (R²) values across different datasets for seedling counting. By integrating RSCconv and ASCFF into other detection frameworks such as Faster R-CNN, RetinaNet, and Deformable DETR, we observed enhanced detection and counting accuracy, further validating the effectiveness of our proposed method. These advancements make RC-Dino particularly suitable for accurate early maize seedling counting in the field. The source code for RSCconv and ASCFF is publicly available at https://github.com/collapser-AI/RC-Dino, promoting further research and practical applications.
Why it matches plant phenotyping methodsUAV画像からトウモロコシ幼苗数を抽出する深層学習手法を開発し、公開データセット上で既存手法と比較検証しており、植物表現型取得法が研究の中心です。
abstractwe developed RC-Dino, a deep learning methodology based on DINO, specifically designed to enhance the precision of seedling counts from UAV-acquired images.
Reproduction assets foundThe paper's EMSD UAV image dataset is explicitly not publicly available, but the authors' RSCconv and ASCFF analysis code for the RC-Dino model is publicly released on GitHub.Code · publicHowever, our code is open to the public. The RSCconv and ASCFF code mentioned in this paper can be found here: https://github.com/collapser-AI/RC-Dino .Open asset ↗collapser-AI/RC-Dinolines:727-739Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Sowing uniformity is an important evaluation indicator of mechanical sowing quality. In order to achieve accurate evaluation of sowing uniformity in hybrid rice mechanical sowing, this study takes the seeds in a seedling tray of hybrid rice blanket-seedling nursing as the research object and proposes a method for evaluating sowing uniformity by combining image processing methods and the ODConv_C2f-ECA-WIoU-YOLOv8n (OEW-YOLOv8n) network. Firstly, image processing methods are used to segment seed image and obtain seed grids. Next, an improved model named OEW-YOLOv8n based on YOLOv8n is proposed to identify the number of seeds in a unit seed grid. The improved strategies include the following: (1) Replacing the Conv module in the Bottleneck of C2f modules with the Omni-Dimensional Dynamic Convolution (ODConv) module, where C2f modules are located at the connection between the Backbone and Neck. This improvement can enhance the feature extraction ability of the Backbone network, as the new modules can fully utilize the information of all dimensions of the convolutional kernel. (2) An Efficient Channel Attention (ECA) module is added to the Neck for improving the network's capability to extract deep semantic feature information of the detection target. (3) In the Bbox module of the prediction head, the Complete Intersection over Union (CIoU) loss function is replaced by the Weighted Intersection over Union version 3 (WIoUv3) loss function to improve the convergence speed of the bounding box loss function and reduce the convergence value of the loss function. The results show that the mean average precision (mAP) of the OEW-YOLOv8n network reaches 98.6%. Compared to the original model, the mAP improved by 2.5%. Compared to the advanced object detection algorithms such as Faster-RCNN, SSD, YOLOv4, YOLOv5s YOLOv7-tiny, and YOLOv10s, the mAP of the new network increased by 5.2%, 7.8%, 4.9%, 2.8% 2.9%, and 3.3%, respectively. Finally, the actual evaluation experiment showed that the test error is from -2.43% to 2.92%, indicating that the improved network demonstrates excellent estimation accuracy. The research results can provide support for the mechanized sowing quality detection of hybrid rice and the intelligent research of rice seeder.
Why it matches plant phenotyping methods画像処理と改良YOLOネットワークにより、育苗トレイ内の種子数からハイブリッドイネの播種均一性を定量評価する手法を開発・検証しており、植物状態の取得・推定が研究の中心である。
abstractproposes a method for evaluating sowing uniformity by combining image processing methods and the ODConv_C2f-ECA-WIoU-YOLOv8n (OEW-YOLOv8n) network.
Accurate counting of plant organs at various growth stages is crucial for crop growth monitoring, phenotypic assessment, and yield prediction. Traditional plant counting models, typically designed for specific plant types, often exhibit poor generalization capabilities. In this research, an improved model, Field-CounTR, was proposed to accurately count multiple plant organs. The Field-Count dataset, utilizing a Few-Shot method, was introduced to enhance accuracy in plant counting. Additionally, a mixed salient module was developed to improve the feature fusion capability of sample images by exploiting their high similarity in Few-Shot counting tasks. During the downsampling process, Shuffle Attention was integrated to reduce redundant information in the feature fusion process. Furthermore, a hybrid convolution module combining Do-Conv and dilated convolutions was developed to increase the speed of convolutional inference and expand the receptive field through over-parameterized and dilation operations. To assess the effectiveness of the proposed approach, tests were conducted using the Field-Count dataset, which includes Sorghum, Wheat, Maize, and Rice. The results demonstrated a mean absolute error (MAE) of 14.49 and a root mean square error (RMSE) of 21.14. Compared with the CounTR model, the Field-CounTR model reduced the MAE and RMSE by 2.01 and 1.33, respectively. The enhanced Field-CounTR model exhibited superior feature extraction performance, high detection accuracy, and excellent generalization capabilities. This model can accurately count multiple plant types in complex field or orchard conditions and offers a wide range of applications.
Why it matches plant phenotyping methods植物器官数という観測可能な形質を画像から推定するFew-Shot計数モデルとデータセットを開発し、複数作物で性能検証しており、表現型取得・抽出手法が中心です。
abstractAccurate counting of plant organs at various growth stages is crucial for crop growth monitoring, phenotypic assessment, and yield prediction.
Tiller number is a key agronomic trait of the crop population, reflecting the adaptability of crop to the environment and the status of plant growth, as well as grain yield. As a state-of-the-art form of active remote sensing that penetrates the vegetation canopy and provides a detailed representation of 3D structures, terrestrial laser scanning (TLS) is beginning to show great potential in precisely counting the tiller number. However, current research of TLS-derived wheat canopy tiller number is commonly affected by mutual occlusion among plants and noise problem. In this study, we proposed a novel Mean Shift Clustering algorithm based on Voxel Interpolation (MSCVI) to effectively mitigate those effects by removing excess noise and interpolating unsampled voxels. The findings demonstrated that there was a strong exponential relationship between the gap fraction (Pgₐₚ) and point cloud density for wheat plots. In addition, MSCVI was effective to count the tiller number of wheat under different field treatments (R² = 0.69, RMSE = 79 tilllers/m²), producing better results than previous adaptive layering and hierarchical clustering (ALHC) algorithm. MSCVI could obtain more precise and detailed wheat canopy information by denoising and compensating the point cloud data, which greatly improved the accuracy of detecting tiller numbers under the condition of high plant density, planophile plant type and tiller stage data. This study provides new insights into effectively mitigating noise and occlusion between plants and within dense canopies, and has potential for accurate calculation of the tiller number in the assessment of crop yield phenotype.
Why it matches plant phenotyping methodsLiDAR点群のノイズ・遮蔽を補正し、コムギの分げつ数を推定する新規アルゴリズムを開発・比較検証しており、植物表現型取得手法が中心である。
abstractwe proposed a novel Mean Shift Clustering algorithm based on Voxel Interpolation (MSCVI) to effectively mitigate those effects by removing excess noise and interpolating unsampled voxels.
Tomato ( Solanum lycopersicum L.) is one of the most important vegetables in the world economy, consequently, it presents as a model organism for biotechnology research and plant breeding programs. The phenotyping by image is a technique that can be applied in these programs and allows the analysis of the experiment quickly, accurately, objectively and without destroying samples. In this sense, the objective of this study was to establish methodologies for phenotyping of the productivity in tomato plant, through computational analysis of images using the Mask-RCNN algorithm and to test its efficiency in a balanced diallel without reciprocals with the tomato plant. They were evaluated the F1′s at the time of fruit harvest by the traditional phenotyping for productivity and, in parallel, images were captured for evaluation by Mask-RCNN. They were estimated six variables by Mask-RCNN, namely the number of green fruits, number of ripe fruits, number of total fruits, percentage of the area in the image occupied by ripe fruits (PFM), percentage of the area occupied by green fruits (PFV) and percentage of area occupied by all fruits (PFT). Accuracy for recognition was 85 % for ripe fruits and 88 % for green fruits. The model achieved recall of 92 % for ripe fruits and 84 % for green fruits. It was observed a significant correlation between the characters evaluated in traditional and computational way, all with positive values and close to one. The ranks correlation between fruit productivity, obtained by traditional phenotyping and the variables estimated by Mask-RCNN, was high, especially for the number of ripe and total fruits. It was concluded from the results that image analysis is efficient and can be used in high-throughput phenotyping in tomato plant genetic improvement.
Why it matches plant phenotyping methodsトマト果実の生産性を画像とMask-RCNNで推定する表現型計測法を開発し、従来法との精度・相関を検証しており、方法が研究の中心である。
abstractAccuracy for recognition was 85 % for ripe fruits and 88 % for green fruits. The model achieved recall of 92 % for ripe fruits and 84 % for green fruits.
Viticulture benefits significantly from rapid grape bunch identification and counting, enhancing yield and quality. Recent technological and machine learning advancements, particularly in deep learning, have provided the tools necessary to create more efficient, automated processes that significantly reduce the time and effort required for these tasks. On one hand, drone, or Unmanned Aerial Vehicles (UAV) imagery combined with deep learning algorithms has revolutionised agriculture by automating plant health classification, disease identification, and fruit detection. However, these advancements often remain inaccessible to farmers due to their reliance on specialized hardware like ground robots or UAVs. On the other hand, most farmers have access to smartphones. This article proposes a novel approach combining UAVs and smartphone technologies. An AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training. By leveraging UAV-captured data for training, the proposed model not only accelerates the detection process but also enhances the accuracy and adaptability of grape bunch detection across different devices, surpassing the efficiency of traditional and purely UAV-based methods. To this end, using a dataset of UAV videos recorded during early growth stages in July (BBCH77-BBCH79), the X-Decoder segments vegetation in the front of the frames from their background and surroundings. X-Decoder is particularly advantageous because it can be seamlessly integrated into the AI pipeline without requiring changes to how data is captured, making it more versatile than other methods. Then, YOLO is trained using the videos and further applied to images taken by farmers with common smartphones (Xiaomi Poco X3 Pro and iPhone X). In addition, a web app was developed to connect the system with mobile technology easily. The proposed approach achieved a precision of 0.92 and recall of 0.735, with an F1 score of 0.82 and an Average Precision (AP) of 0.802 under different operation conditions, indicating high accuracy and reliability in detecting grape bunches. In addition, the AI-detected grape bunches were compared with the actual ground truth, achieving an R² value as high as 0.84, showing the robustness of the system. This study highlights the potential of using smartphone imaging and web applications together, making an effort to integrate these models into a real platform for farmers, offering a practical, affordable, accessible, and scalable solution. While smartphone-based image collection for model training is labour-intensive and costly, incorporating UAV data accelerates the process, facilitating the creation of models that generalise across diverse data sources and platforms. This blend of UAV efficiency and smartphone precision significantly cuts vineyard monitoring time and effort.
Why it matches plant phenotyping methodsスマートフォン画像とUAVデータを用いてブドウ房を検出・計数するAIパイプラインを開発し、実測値との比較で性能検証しているため、植物形質取得法が中心である。
abstractAn AI-based framework is introduced, integrating a 5-stage AI pipeline combining object detection and pixel-level segmentation algorithms to automatically detect grape bunches in smartphone images of a commercial vineyard with vertical trellis training.
Crop field monitoring using unmanned aerial vehicles (UAVs) is one of the most important technologies for plant growth control in modern precision agriculture. One of the important and widely used tasks in field monitoring is plant stand counting. The accurate identification of plants in field images provides estimates of plant number per unit area, detects missing seedlings, and predicts crop yield. Current methods are based on the detection of plants in images obtained from UAVs by means of computer vision algorithms and deep learning neural networks. These approaches depend on image spatial resolution and the quality of plant markup. The performance of automatic plant detection may affect the efficiency of downstream analysis of a field cropping pattern. In the present work, a method is presented for detecting the plants of five species in images acquired via a UAV on the basis of image segmentation by deep learning algorithms (convolutional neural networks). Twelve orthomosaics were collected and marked at several sites in Russia to train and test the neural network algorithms. Additionally, 17 existing datasets of various spatial resolutions and markup quality levels from the Roboflow service were used to extend training image sets. Finally, we compared several texture features between manually evaluated and neural-network-estimated plant masks. It was demonstrated that adding images to the training sample (even those of lower resolution and markup quality) improves plant stand counting significantly. The work indicates how the accuracy of plant detection in field images may affect their cropping pattern evaluation by means of texture characteristics. For some of the characteristics (GLCM mean, GLRM long run, GLRM run ratio) the estimates between images marked manually and automatically are close. For others, the differences are large and may lead to erroneous conclusions about the properties of field cropping patterns. Nonetheless, overall, plant detection algorithms with a higher accuracy show better agreement with the estimates of texture parameters obtained from manually marked images.
Why it matches plant phenotyping methodsUAV画像から植物個体をセグメンテーションし、株数・欠株などの植物状態を推定する画像解析手法を開発・評価しており、手法が研究の中心です。
abstractIn the present work, a method is presented for detecting the plants of five species in images acquired via a UAV on the basis of image segmentation by deep learning algorithms (convolutional neural networks).
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Supplement · publicThe following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/jimaging11010028/s1 , “Supplementary Material.pdf” contains the following Supplementary Materials: Table S1. The field location of the crop image dataset from Russia (2019–2023); Table S2. Public datasets from Roboflow used for the analysis (accessed on 25 November 2023); Table S3. The row spacing (for different crops) used in the work to mark up images from the additional datasets (not ours); Table S4. Description of the ResNet neural network architectures for models RN18, RN34, and RN50; Table S5. Description of the texture characteristics; Table S6. Estimates of the four texture characteristicsOpen asset ↗lines:344-359Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · checked 15 Sept 2026
Accurate leaf segmentation and counting are critical for advancing crop phenotyping and improving breeding programs in agriculture. This study evaluates YOLOv11-based models for automated leaf detection and segmentation across spring barley, spring wheat, winter wheat, winter rye, and winter triticale. The key focus is assessing whether a unified model trained on a combined multi-crop dataset can outperform crop-specific models. Results show that the unified model achieves superior performance in bounding box tasks, with mAP@50 exceeding 0.85 for spring crops and 0.7 for winter crops. Segmentation tasks, however, reveal mixed results, with individual models occasionally excelling in recall for winter crops. These findings highlight the benefits of dataset diversity in improving generalization, while emphasizing the need for larger annotated datasets to address variability in real-world conditions. While the combined dataset improves generalization, the unique characteristics of individual crops may still benefit from specialized training.
Why it matches plant phenotyping methods葉の検出・セグメンテーション・計数を対象とする統一画像解析モデルを開発・評価しており、植物形態形質の抽出手法が研究の中心である。
abstractAccurate leaf segmentation and counting are critical for advancing crop phenotyping and improving breeding programs in agriculture.
Quantifying planting layouts during the seedling stage of mung beans (Vigna radiata L.) is crucial for assessing cultivation conditions and providing support for precise management. Traditional information extraction methods are often hindered by engineering workloads, time consumption, and labor costs. Applying deep-learning technologies for information extraction reduces these burdens and yields precise and reliable results, enabling a visual analysis of seedling distribution. In this work, an unmanned aerial vehicle (UAV) was employed to capture visible light images of mung bean seedlings in a field across three height gradients of 2 m, 5 m, and 7 m following a time series approach. To improve detection accuracy, a small target detection layer (p2) was integrated into the YOLOv8-obb model, facilitating the identification of mung bean seedlings. Image detection performance and seedling information were analyzed considering various dates, heights, and resolutions, and the K-means algorithm was utilized to cluster feature points and extract row information. Linear fitting was performed via the least squares method to calculate planting layout parameters. The results indicated that on the 13th day post seeding, a 2640 × 1978 image captured at 7 m above ground level exhibited optimal detection performance. Compared with YOLOv8, YOLOv8-obb, YOLOv9, and YOLOv10, the YOLOv8-obb-p2 model improved precision by 1.6%, 0.1%, 0.3%, and 2%, respectively, and F1 scores improved by 2.8%, 0.5%, 0.5%, and 3%, respectively. This model extracts precise information, providing reliable data for quantifying planting layout parameters. These findings can be utilized for rapid and large-scale assessments of mung bean seedling growth and development, providing theoretical and technical support for seedling counting and planting layouts in hole-seeded crops.
Why it matches plant phenotyping methodsUAV画像と改良YOLOモデルを用いて、ムング豆苗の検出・計数および植栽配置パラメータを抽出する手法を開発・評価しており、植物形態・分布の定量化が研究の中心です。
abstractApplying deep-learning technologies for information extraction reduces these burdens and yields precise and reliable results, enabling a visual analysis of seedling distribution.
Accurately obtaining both the number and the location of rice plants plays a critical role in agricultural applications, such as precision fertilization and yield prediction. With the rapid development of deep learning, numerous models for plant counting have been proposed. However, many of these models contain a large number of parameters, making them unsuitable for deployment in agricultural settings with limited computational resources. To address this challenge, we propose a novel pruning method, Cosine Norm Fusion (CNF), and a lightweight feature fusion technique, the Depth Attention Fusion Module (DAFM). Based on these innovations, we modify the existing P2PNet network to create P2P-CNF, a lightweight model for rice plant counting. The process begins with pruning the trained network using CNF, followed by the integration of our lightweight feature fusion module, DAFM. To validate the effectiveness of our method, we conducted experiments using rice datasets, including the RSC-UAV dataset, captured by UAV. The results demonstrate that our method achieves a MAE of 3.12 and an RMSE of 4.12 while utilizing only 33% of the original network parameters. We also evaluated our method on other plant counting datasets, and the results show that our method achieves a high counting accuracy while maintaining a lightweight architecture.
Why it matches plant phenotyping methodsUAV画像からイネ個体数と位置を推定する軽量な画像ベース計数手法を開発・検証しており、植物表現型の取得方法が研究の中心です。
titleDevelopment of a Lightweight Model for Rice Plant Counting and Localization Using UAV-Captured RGB Imagery
High-throughput phenotyping (HTP) has become a powerful tool for gaining insights into the genetic and environmental factors that affect cotton (Gossypium spp.) growth and yield. With the recent advances in the field of computer vision, namely the integration of deep learning algorithms, the accuracy and efficiency of HTP systems have improved dramatically, enabling them to automatically quantify such fundamental phenotypic traits as fruit identification and enumeration. However, there is currently no HTP system available for counting all the reproductive phases of cotton crop that can be deployed in agronomic field conditions throughout the growing season. This study presents CottonSense, an advanced HTP system that overcomes the challenges of deployment across multiple growth periods by effectively segmenting and enumerating cotton fruits at four stages of growth, including square, flower, closed boll, and open boll. Consequently, CottonSense enhances agronomic management through increased opportunities for data collection and analysis. Using RGB-D cameras, it captures and processes both two and three-dimensional data, facilitating a wider range of phenotypic trait extractions such as crop biomass and plant architecture. To segment the cotton fruits, a Mask-RCNN model is trained and optimized for faster inference using TensorRT. The model yields an average AP score of 79% in segmentation across the four fruit categories. Moreover, the model's accuracy in estimating total fruit count per image is validated by a strong agreement with the counts given by ten domain experts, as reflected by an R² value of 0.94. Furthermore, to accurately count the segmented fruits over large populations of plants, an enumeration algorithm based on a tracking strategy is developed that achieves an R² value of 0.93 when compared to hand-counted fruits in the field. The proposed HTP system, which is implemented entirely on an edge computing device, is cost-effective and power-efficient, making it an effective tool for high-yield cotton breeding and crop improvement. The code for CottonSense is publicly available at https://github.com/FeriBolour/CottonSense.
Why it matches plant phenotyping methods綿花の果実を画像から分割・計数する高スループット表現型計測システムを開発し、専門家および手作業計数で検証しているため、方法が研究の中心である。
abstractThis study presents CottonSense, an advanced HTP system that overcomes the challenges of deployment across multiple growth periods by effectively segmenting and enumerating cotton fruits at four stages of growth, including square, flower, closed boll, and open boll.
The increasing popularity of blueberry has led to expanded blueberry production in many parts of the world. Berry size and average berry weight are key factors in determining the price and marketability of blueberries and therefore are important traits for breeders and researchers to evaluate. Manual measurement of berry size and average berry weight is labor-intensive and prone to human error. This study developed an automated algorithm and smartphone application for accurate blueberry count and size estimation. Two different computer vision pipelines based on traditional methods and deep neural networks were implemented to detect and segment individual blueberries from Red-Green-Blue (RGB) images. The first pipeline used traditional algorithms such as Hough Transform, Watershed, and filtering. The second pipeline deployed YOLOv5 models with additional modifications using the Ghost module and bi-Feature Pyramid Network (biFPN). A total of 198 images of blueberries, together with manually measured berry count and average berry weight, were used to train and test the model performance. The YOLOv5-based model miscounted four berries in 4,604 total berries across the 198 images. The mean average precision was 92.3%, averaged across an intersection-over-union threshold between 0.50−0.95. The model-derived average berry size was highly correlated with measured average berry weight (R2 > 0.93), which translated to a mean absolute error of around 0.14 g (8.3%). An Android application was also developed in this study to allow easier access to implemented models for berry size and weight phenotyping.
Why it matches plant phenotyping methodsブルーベリーの個数・サイズ・重量を画像から推定するコンピュータビジョン手法とスマートフォンアプリを開発・検証しており、植物形質取得が研究の中心である。
abstractThis study developed an automated algorithm and smartphone application for accurate blueberry count and size estimation.
Subcellular RNA localization is an underexplored regulatory layer crucial for properly adapting cells to cellular or environmental conditions. Most studies describing RNA localization have been performed by cell fractionation and subsequent RNA quantification from pools of cells, thereby missing information about cell-to-cell variability. RNA single-molecule fluorescent in situ hybridization (smFISH) is an effective technique for detecting single RNA molecules and identifying subcellular accumulation patterns. Nevertheless, obtaining quantitative results from smFISH can be challenging in tissues with high autofluorescence, like in plants. Here, we describe an automated pipeline to detect and quantify nucleocytoplasmic RNA levels from Arabidopsis root smFISH images. This pipeline utilizes free image preprocessing, segmentation, and RNA detection software. The method permits users with any programming skills to analyze batches of images. Suggestions and recommendations for image acquisition, processing, and data analysis are included. This pipeline allows quantitative differences in nucleocytoplasmic distribution at the single-cell level to be studied under different cellular, environmental, and genetic contexts.
Why it matches plant phenotyping methodsArabidopsis根のsmFISH画像から細胞内RNA分布を自動検出・定量する画像解析パイプラインを開発しており、植物の状態を測定する方法が中心である。
abstractHere, we describe an automated pipeline to detect and quantify nucleocytoplasmic RNA levels from Arabidopsis root smFISH images.
Estimating three-dimensional (3D) fruit-level phenotypic traits of apple trees can potentially improve apple orchard breeders' management strategy. However, the phenotypic traits of apple fruits (including the quantity, 3D distribution, and volume of apples) constitute important parameters that influence yield but are difficult to quantify manually. Therefore, it is necessary to effectively and efficiently quantify apple phenotyping to monitor apple yield and support a better management system. This study developed a novel method for extracting individual 3D apple traits and 3D mapping for three apple training systems. The 3D point cloud of apples was reconstructed from multi-view images collected via a multi-camera system-based unmanned-aerial vehicle. Individual apples in the 3D point cloud were extracted via a 3D instance segmentation algorithm that included generalized sparse convolutional neural networks, a weighted discriminative loss function, and a varying density-based 3D clustering method. The developed apple trait extraction algorithm can help compute the position and volume of an individual apple. The R² values with the average weighted-mean-absolute-percentage error (WMAPE) of apple counting and apple volume estimation were 0.84 - 0.99 (VMAPE = 3.41 - 12.75%) and 0.84 - 0.90 (WMAPE = 4.74 - 7.03%), respectively. The 3D spatial and volumetric distribution of apples were obtained and analyzed. This study developed an effective method combining 3D photography and 3D instance segmentation that can accurately estimate individual apple phenotypic traits from different types of apple training systems in orchards and can also be utilized for the analysis of other fruit traits.
Why it matches plant phenotyping methodsリンゴ果実の個体数・3D分布・体積という植物形質を、マルチビュー画像、3D再構成、インスタンスセグメンテーションで抽出する手法を開発・検証しており、フェノタイピング手法が研究の中心である。
abstractThis study developed a novel method for extracting individual 3D apple traits and 3D mapping for three apple training systems.
Numerous studies have reported a significant positive correlation between wheat yield and the quantity of wheat heads. However, collecting data on wheat heads in the field poses a challenge for several reasons, including the uncontrollable nature of the environment, inconsistent data quality, and ambiguous data truth. To address these challenges, we developed a simulation strategy to replicate the conditions of a real wheat field, which enabled the data collection process to be conducted indoors over a short period. After applying grayscale image processing to process the simulated wheat images, we trained and tested nine deep learning models: Faster-RCNN, YOLOv7, YOLOv8, CenterNet, SSD, RetinaNet, EfficientDet, Deformable-DETR and DINO. Our results indicated that YOLOv7 performed the best (R² = 0.963, RMSE = 2.463). We then compared our model trained on simulated wheat data to a model trained on real wheat data (R² = 0.963 vs 0.972, RMSE = 2.463 vs 2.692). We also achieved good model performance on five test sets: GWHD, SDAU2021-SDAU2024. The results demonstrated the efficacy of our simulation, which provides an efficient and convenient strategy for the precision agriculture community.
Why it matches plant phenotyping methods小麦穂数という植物形態形質の画像取得・推定手法を、シミュレーション画像、画像処理、複数検出モデルの比較、実画像および公開データセットでの評価を通じて開発・検証しており、フェノタイピング手法が中心である。
abstractwe developed a simulation strategy to replicate the conditions of a real wheat field, which enabled the data collection process to be conducted indoors over a short period.
This paper investigates the application of a VR-controlled robotic system for yield monitoring in strawberry farming within a greenhouse environment. The study aims to evaluate the effectiveness of the system in identifying and counting ripe strawberries, categorized by size (small and large) and variety (Seascape and Albion), and compares with the obtained results by an onsite human expert. We designed experiments, in a controlled environment agriculture center, and conducted in two trials. The yield monitoring performance of the developed robotic system was evaluated based on two primary metrics of cycle completion times and fruit detection accuracy, 32 strawberry plants which grew 336 ripe fruits. In the first experiment, the system achieved detection rates of 63 % for small strawberries and 72 % for large strawberries, with cycle completion times ranging from 12.5 to 16 s. In the second experiment, improvements were observed, with detection rates increasing to 74 % for both sizes and cycle completion times reduced to between 11.9 and 15.7 s. The developed robotic system demonstrated high accuracy and efficiency, particularly with larger strawberries. However, some limitations were identified, including challenges related to occlusion. These findings suggest that while the VR-controlled robotic system has the potential to complement and even surpass traditional yield monitoring methods managed by human experts, further refinements are necessary. Future research should focus on optimizing the system’s performance and adapting the system to broader applications in agriculture.
Why it matches plant phenotyping methodsVR制御ロボットによるイチゴ果実の検出・計数とサイズ分類を、検出精度および処理時間で評価しており、果実収量関連形質の取得方法と技術性能が研究の中心である。
abstractThis paper investigates the application of a VR-controlled robotic system for yield monitoring in strawberry farming within a greenhouse environment.
Abstract has not been obtained from indexed metadata or an accessible article page.
Why it matches plant phenotyping methods果実幼果の検出・計数という植物器官形質を対象に、複数の物体検出手法の性能評価を行う研究であり、画像ベースの表現型取得・抽出手法が中心である。
titleComprehensive Performance Evaluation of Yolov12, Yolo11, Yolov10, Yolov9 and Yolov8 on Detecting and Counting Fruitlet in Complex Orchard Environments
Abstract has not been obtained from indexed metadata or an accessible article page.
Why it matches plant phenotyping methods果実幼果の検出・計数という植物器官の形質推定を対象に、複数の画像認識モデルを包括的に性能評価しており、手法の比較検証が中心である。
titleComprehensive Performance Evaluation of Yolov12, Yolo11, Yolov10, Yolov9 and Yolov8 on Detecting and Counting Fruitlet in Complex Orchard Environments
Precision agriculture leverages data and machine learning so that farmers can monitor their crops and target interventions precisely. This enables the precision application of herbicide only to weeds, or the precision application of fertilizer only to undernourished crops, rather than to the entire field. The approach promises to maximize yields while minimizing resource use and harm to the surrounding environment. To this end, we propose a hierarchical panoptic segmentation method that simultaneously determines leaf count (as an identifier of plant growth)and locates weeds within an image. In particular, our approach aims to improve the segmentation of smaller instances like the leaves and weeds by incorporating focal loss and boundary loss. Not only does this result in competitive performance, achieving a PQ+ of 81.89 on the standard training set, but we also demonstrate we can improve leaf-counting accuracy with our method. The code is available at https://github.com/madeleinedarbyshire/HierarchicalMask2Former.
Why it matches plant phenotyping methods植物・葉の階層的パノプティックセグメンテーション法を開発し、葉数という植物成長形質の推定精度を評価しているため、フェノタイピング手法が中心である。
abstractwe propose a hierarchical panoptic segmentation method that simultaneously determines leaf count (as an identifier of plant growth)and locates weeds within an image.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 6 Sept 2026
Accurate crop density estimation is critical for effective agricultural resource management, yet existing methods face challenges due to data acquisition difficulties and low model usability caused by inconsistencies between optical and radar imagery. This study presents a novel approach to maize density estimation by integrating optical and radar data, addressing these challenges with a unique mapping strategy. The strategy combines available data selection, key feature extraction, and optimization to improve accuracy across diverse growth stages. By identifying critical features for maize density and incorporating machine learning to explore optimal feature combinations, we developed a multi-temporal model that enhances estimation accuracy, particularly during leaf development, stem elongation, and tasseling stages (R2 = 0.602, RMSE = 0.094). Our approach improves performance over single-temporal models, and successful maize density maps were generated for the three typical demonstration counties. This work represents an advancement in large-scale crop density estimation, with the potential to expand to other regions and support precision agriculture efforts, offering a foundation for future research on optimizing agricultural resource management.
Why it matches plant phenotyping methods光学・レーダー時系列データと機械学習によるトウモロコシ密度推定モデルを開発・評価しており、植物形質の取得手法が中心です。
abstractThis study presents a novel approach to maize density estimation by integrating optical and radar data
Although season-long cotton flower counts have value to breeders and growers, a manual data collection process is too laborious to be practical in most cases. In recent years, several fully automated flower counting approaches have been proposed. However, such approaches are typically designed to run offline and require a significant amount of computation. Furthermore, little thought has gone into developing convenient interfaces and integrations so that a layperson can use such systems without extensive training. The goal of this study is to develop a lightweight flower tracking system that is deployable on a ground robot and can operate in real-time. We modify a previous GCNNMatch++ approach to increase the inference speed. Additionally, we fuse data from multiple cameras in order to avoid canopy occlusions, and extract three-dimensional flower locations by integrating GPS data from the robot. We show that our approach significantly outperforms UAV-based counting and single-camera counting while running at above 40 FPS on an edge device, achieving a counting error of 15% and an average localization error of 19 cm. This level of performance is enough to observe significant differences in flowering behavior between genotypes. Overall, we believe that our highly-integrated, automated, and simplified flower counting solution makes significant strides towards a practical commercial cotton phenotyping platform.
Why it matches plant phenotyping methodsリアルタイムの花数・三次元位置推定を行う画像解析・ロボット統合手法を開発し、精度と処理速度を評価した綿花フェノタイピング研究であり、方法が中心である。
abstractThe goal of this study is to develop a lightweight flower tracking system that is deployable on a ground robot and can operate in real-time.
ABSTRACT Phenotypic trait identification is crucial in cultivating new soybean varieties with high yield and quality. The traditional soybean phenotypic trait identification relies on manual pod counting and plant height measuring with a ruler. The heavy workload causes the data collected by human resources to be extremely prone to error. Therefore, developing an efficient and high‐quality method to obtain phenotypic data of soybean pods and branches is urgently needed. Three network models including ResNet‐101, Swin‐S and ConvNeXt‐S are compared in this study, and the ConvNeXt‐S model is identified as optimal, with a mAP@0.5 of 0.95, which could reach 74.2%. A deep learning–based approach soybean plant phenotype detection and data storage system is developed, including information on plants, pods and branches. The R2 between the number of pods detected by the system and the true value reached 0.995. These results indicate that the system is more accurate and stable than the manual phenotype identification. Our study paves the way for reducing economic and time costs as well as improving phenotype identification efficiency and accuracy in detecting soybean phenotypes.
Why it matches plant phenotyping methods深層学習によるダイズの莢数・枝などの表現型検出システムを開発し、手作業との精度比較・検証を行っており、表現型取得手法が研究の中心である。
abstractTherefore, developing an efficient and high‐quality method to obtain phenotypic data of soybean pods and branches is urgently needed.
Time to maturity and yield are important traits for highbush blueberry (Vaccinium corymbosum) breeding. Proper determination of the time to maturity of blueberry varieties and breeding lines informs the harvest window, ensuring that the fruits are harvested at optimum maturity and quality. On the other hand, high-yielding crops bring in high profits per acre of planting. Harvesting and quantifying the yield for each blueberry breeding accession are labor-intensive and impractical. Instead, visual ratings as an estimation of yield are often used as a faster way to quantify the yield, which is categorical and subjective. In this study, we developed and shared a high-throughput phenotyping method using neural networks to predict blueberry time to maturity and to provide a proxy for yield, overcoming the labor constraints of obtaining high-frequency data. We aim to facilitate further research in computer vision and precision agriculture by publishing the labeled image dataset and the trained model. In this research, true-color images of blueberry bushes were collected, annotated, and used to train a deep neural network object detection model [You Only Look Once (YOLOv11)] to detect mature and immature berries. Different versions of YOLOv11 were used, including nano, small, and medium, which had similar performance, while the medium version had slightly higher metrics. The YOLOv11m model shows strong performance for the mature berry class, with a precision of 0.90 and an F1 score of 0.90. The precision and recall for detecting immature berries were 0.81 and 0.79. The model was tested on 10 blueberry bushes by hand harvesting and weighing blueberries. The results showed that the model detects approximately 25% of the berries on the bushes, and the correlation coefficients between model-detected and hand-harvested traits were 0.66, 0.86, and 0.72 for mature fruit count, immature fruit count, and mature ratio, respectively. The model applied to 91 blueberry advance selections and categorized them into groups with diverse levels of maturity and productivity using principal component analysis (PCA). These results inform the harvest window and yield of these breeding lines with precision and objectivity through berry classification and quantification. This model will be helpful for blueberry breeders, enabling more efficient selection, and for growers, helping them accurately estimate optimal harvest windows. This open-source tool can potentially enhance research capabilities and agricultural productivity.
Why it matches plant phenotyping methodsブルーベリーの成熟度・収量 proxy を画像とニューラルネットワークで推定する高スループット表現型計測法を開発・検証し、モデルとラベル付きデータセットを共有しているため、方法が研究の中心である。
abstractwe developed and shared a high-throughput phenotyping method using neural networks to predict blueberry time to maturity and to provide a proxy for yield
Reproduction assets foundThe paper publishes its labeled blueberry image dataset on Zenodo (record 14014858) and its trained YOLOv11-based blueberry fruit counting model/code on GitHub (jeromemaleski/blueberry), both directly supporting the paper's phenotyping measurements and analysis.Dataset · public32. Zhang, J. Blueberry Images and Labels for YOLO Model Training. Zenodo. 2024. Available online: https://zenodo.org/records/Open asset ↗zenodopdf-page:14 lines:1-36Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
This study aims to improve the precision of wheat spike counting and disease detection, exploring the application of deep learning in the agricultural sector. Addressing the shortcomings of traditional detection methods, we propose an advanced feature extraction strategy and a model based on the probability density attention mechanism, designed to more effectively handle feature extraction in complex backgrounds and dense areas. Through comparative experiments with various advanced models, we comprehensively evaluate the performance of our model. In the disease detection task, our model performs excellently, achieving a precision of 0.93, a recall of 0.89, an accuracy of 0.91, and an mAP of 0.90. By introducing the density loss function, we are able to effectively improve the detection accuracy when dealing with high-density regions. In the wheat spike counting task, the model similarly demonstrates a strong performance, with a precision of 0.91, a recall of 0.88, an accuracy of 0.90, and an mAP of 0.90, further validating its effectiveness. Furthermore, this paper also conducts ablation experiments on different loss functions. The results of this research provide a new method for wheat spike counting and disease detection, fully reflecting the application value of deep learning in precision agriculture. By combining the probability density attention mechanism and the density loss function, the proposed model significantly improves the detection accuracy and efficiency, offering important references for future related research.
Why it matches plant phenotyping methodsコムギ穂数の計数と植物病害の検出を対象に、深層学習モデルと特徴抽出・損失関数を開発し、比較およびアブレーション実験で評価しており、表現型取得手法が中心である。
abstractwe propose an advanced feature extraction strategy and a model based on the probability density attention mechanism
var. capitata L.) quantification cultivated under different types of mulching, using aerial images captured by RPAS (Remotely Piloted Aircraft System). Design/methodology/approach: The cabbage plantation used for the study was established under a completely randomized block design with different types of mulch as treatments: black plastic, white plastic, straw, and bare soil. Manual plant counts and automated estimates were performed using two agricultural artificial intelligence platforms (Platforms A and B). The relationship was evaluated using linear regression correlation (R²), and the following indicators were subsequently used: estimation accuracy (Ps), estimation error percentage (Es), mean absolute error (MAE), and root mean square error (RMSE). Results: Platform A showed a correlation coefficient range of R²=0.41 to 0.91. Platform B obtained R² values ranging from 0.77 to 0.88. Platform A exhibited the highest estimation accuracy (Ps) with 98.3% and an estimation error (Es) of -1.7% for straw mulch, with a mean absolute error (MAE) of 2.0% and a root mean square error (RMSE) of 1 for bare soil. Both platforms showed underestimations in the number of detected plants, ranging from -6.7% to -1.7%. Limitations on study/implications: The use of RPAS was limited by atmospheric conditions such as wind and rain. Findings/conclusions: The effectiveness of counting cabbage plants using RPAS was validated.
Why it matches plant phenotyping methodsRPAS画像とAIプラットフォームによるキャベツ個体数の自動推定を開発・評価し、手動計数との相関や誤差で検証しているため、植物表現型取得手法が中心である。
abstractManual plant counts and automated estimates were performed using two agricultural artificial intelligence platforms (Platforms A and B).
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · Crossref · checked 6 Sept 2026
Abstract Accurate leaf segmentation and counting are critical for advancing crop phenotyping and improving breeding programs in agriculture. This study evaluates YOLOv11-based models for automated leaf detection and segmentation across spring barley, spring wheat, winter wheat, winter rye, and winter triticale. The key focus is assessing whether a unified model trained on a combined multi-crop dataset can outperform crop-specific models. Results show that the unified model achieves superior performance in bounding box tasks, with mAP@50 exceeding 0.85 for spring crops and above 0.7 for winter crops. Segmentation tasks, however, reveal mixed results, with individual models occasionally excelling in recall for winter crops. These findings highlight the benefits of dataset diversity in improving generalization, while emphasizing the need for larger annotated datasets to address variability in real-world conditions. This highlights that while the combined dataset improves generalization, the unique characteristics of individual crops may still benefit from specialized training. This work demonstrates the potential of AI-driven models to advance automated phenotyping for large-scale precision agriculture.
Why it matches plant phenotyping methods複数作物の葉を自動検出・セグメンテーションし、統合モデルと作物別モデルの性能を比較する画像解析手法が中心で、葉の抽出・計数という植物表現型取得に直接関係する。
abstractAccurate leaf segmentation and counting are critical for advancing crop phenotyping and improving breeding programs in agriculture.
Silvicultural operations such as planting, pruning, and thinning are vital for the forest value chain, requiring efficient monitoring to prevent value loss. While effective, traditional field plots are time-consuming, costly, spatially limited, and rely on assumptions that they adequately represent a wider area. Alternatively, unmanned aerial vehicles (UAVs) can cover large areas while keeping operators safe from hazards including steep terrain. Despite their utility, optimal flight parameters to ensure flight efficiency and data quality remain under-researched. This study evaluated the impact of forward and side overlap and flight altitude on the quality of two- and three-dimensional spatial data products from UAV photogrammetry (UAV-SfM) for assessing stand density in a recently thinned Pinus radiata D. Don plantation. A contemporaneously acquired UAV laser scanner (ULS) point cloud provided reference data. The results indicate that the optimal UAV-SfM flight parameters are 90% forward and 85% side overlap at a 120 m altitude. Flights at an 80 m altitude offered marginal resolution improvement (2.2 cm compared to 3.2 cm ground sample distance/GSD) but took longer and were more error-prone. Individual tree detection (ITD) for stand density assessment was then applied to both UAV-SfM and ULS canopy height models (CHMs). Manual cleaning of the detected ULS tree peaks provided ground truth for both methods. UAV-SfM had a lower recall (0.85 vs. 0.94) but a higher precision (0.97 vs. 0.95) compared to ULS. Overall, the F-score indicated no significant difference between a prosumer-grade photogrammetric UAV and an industrial-grade ULS for stand density assessments, demonstrating the efficacy of affordable, off-the-shelf UAV technology for forest managers. Furthermore, in addressing the knowledge gap regarding optimal UAV flight parameters for conducting operational forestry assessments, this study provides valuable insights into the importance of side overlap for orthomosaic quality in forest environments.
Why it matches plant phenotyping methodsUAV-SfMの飛行条件を評価・最適化し、個体樹検出と林分密度という植物群落形質をULSと比較検証しており、フェノタイピング手法が中心です。
abstractThis study evaluated the impact of forward and side overlap and flight altitude on the quality of two- and three-dimensional spatial data products from UAV photogrammetry (UAV-SfM) for assessing stand density
Wheat yield is positively correlated with the number of wheat spikes in the field, which is an essential index for plant breeders. To efficiently calculate this index, there is a high demand for precise and automatic plant phenotyping methods that quantify images. As a tool for estimating crop yields, machine vision is rapidly advancing. In the majority of crop phenotyping fields, however, annotating thousands of small objects using bounding boxes or polygons is extremely time- and labor-intensive. This study aims to develop a state-of-the-art framework for localizing and counting wheat spikes using dotted annotation datasets in conjunction with Gaussian and constant density map generation algorithms. In addition, we developed hybrid UNet architectures as the computational component, including VGG16-UNet, ResNet34-UNet, ResNet50-UNet, and ResNeXt-UNet. Furthermore, we improved the performance of wheat counting by employing a hybrid density map estimation-non-maximal supervision algorithm. Wheat images from the ACID and GWHD datasets are used to evaluate the proposed models. With F1 score and Mean Absolute Percentage Error (MAPE) of 0.96 and 1.68%, respectively, the results from the ACID dataset demonstrate a significant improvement in spike localization and counting compared to previous research studies. Additionally, to assess the proposed models’ generalizability in real-world modeling scenarios, the ACID-based pretrained models are used to predict the more complex in-field GWHD dataset. With an F1 score of 0.57 and a MAPE of 2.302%, the pretrained models were able to localize and count the wheat spikes in the new dataset. Following that, the ACID-based pretrained models are retrained for a few epochs using the GWHD dataset’s small training sample size. The results demonstrate a significant improvement in the proposed model’s performance, with an F1 score of 0.91 and a MAPE of 1.56%.
Why it matches plant phenotyping methods小麦穂数の画像ベース局在化・計数という植物形質推定法を開発し、複数データセットで性能と汎化性を評価しており、フェノタイピング手法が中心である。
abstractThis study aims to develop a state-of-the-art framework for localizing and counting wheat spikes using dotted annotation datasets in conjunction with Gaussian and constant density map generation algorithms.
StrawberryFruitClassificationCountingTrackingGrowth / development / phenology
Accurately counting fruit in orchards is a critical step for effective digital farming management. However, the variability in fruit size, overlapping shadows, and light interference present significant challenges to applying computer vision during the strawberry growth phase. To address these challenges, we propose StraTracker, a multi-object tracking (MOT) algorithm specifically designed to identify and count strawberries at various growth stages. StraTracker transforms the counting task into a frame-by-frame tracking problem, integrating both motion and appearance features. The algorithm is composed of three key components: a strawberry detector based on YOLOv8n, a feature association module, and a dual-area counting (DC) module. First, the strawberry detector accurately recognizes five growth stages, achieving an average accuracy of 91.93 % at 38.3 FPS. Next, the feature association module, incorporating the Feature Slicing Attention (FSA) and Adaptive Kalman Filtering (AKF) modules, mitigates issues such as light interference, impractical tracking frames, and ID switching (IDs). As a result, StraTracker achieves a Multi-Object Tracking Accuracy (MOTA) of 83.28 % and a Higher-Order Tracking Accuracy (HOTA) of 77.26 %, with only 259 IDs, outperforming existing baseline models. Finally, the DC module categorizes fruit counts based on the unique IDs assigned during tracking. The algorithm’s coefficient of determination (R2 = 0.91) and GEH of 2.33 indicate a strong correlation between predicted and actual counts. In conclusion, StraTracker offers a promising solution for farmers to optimize planting strategies and develop more precise harvesting plans.
Why it matches plant phenotyping methodsイチゴ果実数という植物器官の形質を、画像検出・多対象追跡・成長段階認識で自動抽出する手法を開発し、精度検証しているため、方法中心の植物フェノタイピング研究である。
abstractwe propose StraTracker, a multi-object tracking (MOT) algorithm specifically designed to identify and count strawberries at various growth stages.
Accurately counting the number of grains per panicle is crucial for evaluating rice yield and selecting superior germplasm resources. Traditional measurement methods are labor-intensive, time-consuming, and prone to errors. To address this challenge, computer vision-based methods have emerged as a promising approach for seed counting. However, achieving precise grain counting is particularly challenging due to their natural morphology, which involves occlusion and substantial variations in size, shape and orientation. This often requires additional steps, such as manual shaping or threshing. Therefore, we propose an innovative approach for precisely counting rice grains in their natural form by integrating object detection, image classification and regression equations. Initially, we trained the Yolov7-tiny model for grain counting. Subsequently, we introduced a classification system based on the variability in the natural morphology of rice panicles using the EfficientNetV2 network, enabling the classification of rice panicles into five distinct classes. Furthermore, we devised a set of univariate linear regression equations for the different classes of rice panicles, utilizing data from 2920 diverse rice germplasm to establish correlations predicted and actual values. Experimental findings demonstrated a counting accuracy of 92.60% with an average absolute percentage error of 7.69%. And, this study revealed that utilizing two-sided images of panicles did not significantly improve counting accuracy. This study represents a successful endeavor in achieving precise and efficient counting of rice panicles within their natural morphology, offering a novel solution for detecting and counting dense objects.
Why it matches plant phenotyping methodsイネ穂の粒数という植物形質を、画像分類・物体検出・回帰により非破壊推定する手法を開発し、精度評価も行っており、フェノタイピング手法が中心である。
abstractwe propose an innovative approach for precisely counting rice grains in their natural form by integrating object detection, image classification and regression equations.
One of the major challenges for the agricultural industry today is the uncertainty in manual labor availability and the associated cost. Automated flower and fruit density estimation, localization, and counting could help streamline harvesting, yield estimation, and crop-load management strategies such as flower and fruitlet thinning. This article proposes a deep regression-based network, AgRegNet, to estimate density, count, and location of flower and fruit in tree fruit canopies without explicit detection or polygon annotation. Inspired by popular U-Net architecture, AgRegNet is a U-shaped network with an encoder-to-decoder skip connection and modified ConvNeXt-T as an encoder feature extractor. AgRegNet can be trained based on information from point annotation and leverages segmentation information and attention modules (spatial and channel) to highlight relevant flower and fruit features while suppressing non-relevant background features. Experimental evaluation in apple flower and fruit canopy images under an unstructured orchard environment showed that AgRegNet achieved promising accuracy as measured by Structural Similarity Index (SSIM), percentage Mean Absolute Error (pMAE) and mean Average Precision (mAP) to estimate flower and fruit density, count, and centroid location, respectively. Specifically, the SSIM, pMAE, and mAP values for flower images were 0.938, 13.7%, and 0.81, respectively. For fruit images, the corresponding values were 0.910, 5.6%, and 0.93. Since the proposed approach relies on information from point annotation, it is suitable for sparsely and densely located objects. This simplified technique will be highly applicable for growers to accurately estimate yields and decide on optimal chemical and mechanical flower thinning practices.
Why it matches plant phenotyping methods花および果実の密度・個数・位置を画像から推定する深層学習手法を提案し、実画像で性能評価しており、植物表現型取得法が研究の中心である。
abstractThis article proposes a deep regression-based network, AgRegNet, to estimate density, count, and location of flower and fruit in tree fruit canopies without explicit detection or polygon annotation.
Monitoring plant growth is crucial for cultivation management. Agronomists can assess the health status of lettuce seedlings based on monitoring results to implement relevant management measures for improving the quality and yield of lettuce seedlings. This study developed a non-destructive, high-throughput growth monitoring method suitable for large-scale assessment of lettuce seedling quality in nurseries. The method utilizes a plant high-throughput phenotyping platform to acquire 10-day time-series imagery data. An Mask2Former network model enhanced by multidimensional collaborative attention mechanism, combined with sliding window and morphological operations, achieves precise recognition and localization of seedling trays, varieties, and individual seedling plants in a progressive manner. Based on individual seedling localization and segmentation results, the method estimates emergence numbers and rates for each variety, and further achieves instance segmentation and counting of individual seedling leaves, innovatively constructing leaf segmentation results of different varieties across the entire seedling tray. Applied to time-series images, the method automatically monitored seedling emergence changes and growth trends for 1,086 lettuce varieties. In monitoring these varieties, the method achieved a coefficient of determination (R²) of 0.96 for emergence number estimation. The extraction of all six key phenotypic parameters demonstrated exceptionally high correlations: projected area, projected perimeter, convex hull area, and convex hull perimeter all showed R² above 0.99, while leaf compactness R² was 0.9698, and leaf count R² was 0.91. Results demonstrate that this high-throughput, reliable method can effectively monitor the growth status of large-scale lettuce seedlings and provide technical support for lettuce nursery quality assessment.
Why it matches plant phenotyping methods画像解析モデルとハイスループット表現型プラットフォームを開発し、個体・葉の形態形質や出芽を自動抽出することが研究の中心であるため。
abstractThis study developed a non-destructive, high-throughput growth monitoring method suitable for large-scale assessment of lettuce seedling quality in nurseries.
The automatic detection and counting of wheat spike images are of great significance for yield prediction and variety evaluation. Therefore, accurate and timely estimation of spike numbers is crucial for wheat production. However, in actual production, due to the susceptibility of wheat spike images to factors such as lighting conditions, shooting angles, occlusion, and overlap, the contour and features of wheat spike is unclear, which affects the accuracy of automatic detection and counting of wheat spike. In order to solve the above problems and further improve the accuracy of wheat spike counting, an improved wheat spike counting model DMseg-Count was proposed by enhancing local contextual supervision information based on existing target object counting model DM-Count. Firstly, wheat spike local segmentation branch was introduced to improve the network architecture of DM-Count, so as to extract the local contextual supervision information of wheat spike. Secondly, an element-by-element point multiplication mechanism was designed to fuse global and local contextual supervision information of wheat spike. Finally, the total loss function was constructed to optimize the model. The test results showed that the mean absolute error (MAE) and root mean square error (RMSE) of the proposed DMseg-Count model were 5.79 and 7.54, respectively, which were 9.76 and 10.91 higher than the standard distribution matching for crowd counting (DM-Count) model. Compared with other deep learning models, the proposed DMseg-Count model can detect wheat spike image in challenging situations, and has better computer vision processing capabilities and performance evaluation detection effect. In summary, the proposed DMseg-Count model can effectively detect wheat spike and has good counting performance, which provides a new method for automatic counting of wheat spike and yield prediction in complex field environments.
Why it matches plant phenotyping methods小麦穂の画像から穂数を自動検出・計数するモデルを開発しており、植物形質（穂数）の取得手法が研究の中心であるため。
abstractan improved wheat spike counting model DMseg-Count was proposed
Common beanAerial / UAVField / plotLiDAR / point cloudRGB / grayscaleWhole plant / canopy / plot / fieldCountingMorphology / geometry measurementGrowth / time-series analysisGrowth / development / phenology
Substantial effort has been made in manually tracking plant maturity and to measure early-stage plant density and crop height in experimental fields. In this study, RGB drone imagery and deep learning (DL) approaches are explored to measure relative maturity (RM), stand count (SC), and plant height (PH), potentially offering higher throughput, accuracy, and cost-effectiveness than traditional methods. A time series of drone images was utilized to estimate dry bean RM employing a hybrid convolutional neural network (CNN) and long short-term memory (LSTM) model. For early-stage SC assessment, Faster RCNN object detection algorithm was evaluated. Flight frequencies, image resolution, and data augmentation techniques were investigated to enhance DL model performance. PH was obtained using a quantile method from digital surface model (DSM) and point cloud (PC) data sources. The CNN-LSTM model showed high accuracy in RM prediction across various conditions, outperforming traditional image preprocessing approaches. The inclusion of growing degree days (GDD) data improved the model's performance under specific environmental stresses. The Faster R-CNN model effectively identified early-stage bean plants, demonstrating superior accuracy over traditional methods and consistency across different flight altitudes. For PH estimation, moderate correlations with ground-truth data were observed across both datasets analyzed. The choice between PC and DSM source data may depend on specific environmental and flight conditions. Overall, the CNN-LSTM and Faster R-CNN models proved more effective than conventional techniques in quantifying RM and SC. The subtraction method proposed for estimating PH without accurate ground elevation data yielded results comparable to the difference-based method. Additionally, the pipeline and open-source software developed hold potential to significantly benefit the phenotyping community.
Why it matches plant phenotyping methodsドローン画像と深層学習を用いて成熟度、株数、草丈を推定する手法を開発・評価し、パイプラインとオープンソースソフトウェアも提示しており、植物表現型取得が研究の中心である。
abstractIn this study, RGB drone imagery and deep learning (DL) approaches are explored to measure relative maturity (RM), stand count (SC), and plant height (PH)
Reproduction assets foundThe paper explicitly states that all R/Python analysis code, apps, and the complete datasets (orthomosaics, shapefiles, ground notes, clipped plots) are publicly available via the authors' GitHub organization and three Zenodo deposits for RM, SC, and PH.Dataset · publicof the manuscript.
Competing interests: The authors declare that they have no competing interests.
Data Availability
Developed software and analysis are available in the GitHub repositories at https://github.com/msudrybeanbreeding and datasets can be download at Zenodo deposit page ( https://zenodo.org/ ) using the links to RM: https://doi.org/10.5281/zenodo.7922565; SC: https://doi.org/10.5281/zenodo.7922584; and PH: https://doi.org/10.5281/zenodo.7922589 .
Supplementary Materials
Supplementary 1
Figs. S1 to S14
Tables S1 and S2
Data files S1 to 21
References
1.
Uebersax
MA , Cichy
KA , Gomez
FE , Porch
TG , Heitholt
J , Osorno
JM , Kamfwa
K , Snapp
SS , Bales
S
.
Dry beans ( Phaseolus vuOpen asset ↗zenodo · 10.5281/zenodo.7922565lines:677-730Dataset · publicauthors declare that they have no competing interests.
Data Availability
Developed software and analysis are available in the GitHub repositories at https://github.com/msudrybeanbreeding and datasets can be download at Zenodo deposit page ( https://zenodo.org/ ) using the links to RM: https://doi.org/10.5281/zenodo.7922565; SC: https://doi.org/10.5281/zenodo.7922584; and PH: https://doi.org/10.5281/zenodo.7922589 .
Supplementary Materials
Supplementary 1
Figs. S1 to S14
Tables S1 and S2
Data files S1 to 21
References
1.
Uebersax
MA , Cichy
KA , Gomez
FE , Porch
TG , Heitholt
J , Osorno
JM , Kamfwa
K , Snapp
SS , Bales
S
.
Dry beans ( Phaseolus vulgaris L.) as a vital component of sustainabOpen asset ↗zenodo · 10.5281/zenodo.7922584lines:677-730Dataset · publicrests.
Data Availability
Developed software and analysis are available in the GitHub repositories at https://github.com/msudrybeanbreeding and datasets can be download at Zenodo deposit page ( https://zenodo.org/ ) using the links to RM: https://doi.org/10.5281/zenodo.7922565; SC: https://doi.org/10.5281/zenodo.7922584; and PH: https://doi.org/10.5281/zenodo.7922589 .
Supplementary Materials
Supplementary 1
Figs. S1 to S14
Tables S1 and S2
Data files S1 to 21
References
1.
Uebersax
MA , Cichy
KA , Gomez
FE , Porch
TG , Heitholt
J , Osorno
JM , Kamfwa
K , Snapp
SS , Bales
S
.
Dry beans ( Phaseolus vulgaris L.) as a vital component of sustainable agriculture and food security—A review
. LeguOpen asset ↗zenodo · 10.5281/zenodo.7922589lines:677-730Code · publics from each individual breeding plot were extracted from the time series of images (6 and 9 flights date), and the RM was estimated using an optimized threshold value of 0.06. To perform the VI extractions from each breeding plot in the field, an open-source Streamlit app in Python was implemented and can be accessed online at: https://msudrybeanbreeding-vegetation-index--vi-extractions-v0-3-9knpzt.streamlit.app/ . Additionally, to accommodate user preferences, an R script is available to perform VI extractions analysis (Data S7 ).
SC DL model
The SC pipeline deployed in this study comprised 6 distinct steps, starting from the raw images and annotations, and ending with the final SC predictiOpen asset ↗lines:139-147Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Tomato harvesting in intelligent greenhouses is crucial for reducing costs and optimizing management. Agricultural robots, as an automated solution, require advanced visual perception. This study proposes a tomato detection and counting algorithm based on YOLOv8 (TCAttn-YOLOv8). To handle small, occluded tomato targets in images, a new detection layer (NDL) is added to the Neck and Head decoupled structure, improving small object recognition. The ColBlock, a dual-branch structure leveraging Transformer advantages, enhances feature extraction and fusion, focusing on densely targeted regions and minimizing small object feature loss in complex backgrounds. C2fGhost and GhostConv are integrated into the Neck network to reduce model parameters and floating-point operations, improving feature expression. The WIoU (Wise-IoU) loss function is adopted to accelerate convergence and increase regression accuracy. Experimental results show that TCAttn-YOLOv8 achieves an mAP@0.5 of 96.31%, with an FPS of 95 and a parameter size of 2.7 M, outperforming seven lightweight YOLO algorithms. For automated tomato counting, the R 2 between predicted and actual counts is 0.9282, indicating the algorithm's suitability for replacing manual counting. This method effectively supports tomato detection and counting in intelligent greenhouses, offering valuable insights for robotic harvesting and yield estimation research.
Why it matches plant phenotyping methodsトマト果実の検出・計数を行う画像解析手法自体を開発・比較しており、果実数という植物器官形質を抽出する中心的研究である。
abstractThis study proposes a tomato detection and counting algorithm based on YOLOv8 (TCAttn-YOLOv8).
Uniform spatial distribution of plants is crucial in arable crops. Seeding quality is affected by numerous parameters, including the working speed and vibrations of the seeder. Therefore, investigating effective and rapid methods to evaluate seeding quality and the parameters affecting the seeders’ performance is of high importance. With the latest advancements in unmanned aerial vehicle (UAV) technology, the potential for acquiring accurate agricultural data has significantly increased, making UAVs an ideal tool for scouting applications in agricultural systems. This study investigates the effectiveness of utilizing different plant recognition algorithms applied to UAV-derived images for evaluating seeder performance based on detected plant spacings. Additionally, it examines the impact of seeding unit vibrations on seeding quality by analyzing accelerometer data installed on the seeder. For the image analysis, three plant recognition approaches were tested: an unsupervised segmentation method based on the Visible Atmospherically Resistant Index (VARI), template matching (TM), and a deep learning model called Mask R-CNN. The Mask R-CNN model demonstrated the highest recognition reliability at 96.7%, excelling in detecting seeding errors such as misses and doubles, as well as in evaluating the quality of feed index and precision when compared to ground-truth data. Although the VARI-based unsupervised method and TM outperformed Mask R-CNN in recognizing double spacings, overall, the Mask R-CNN was the most promising. Vibration analysis indicated that the seeder’s working speed significantly affected seeding quality. These findings suggest areas for potential improvements in machine technology to improve sowing operations.
Why it matches plant phenotyping methodsUAV画像から植物間隔と播種エラーを抽出する複数の認識手法を比較・検証しており、植物表現型の取得方法が研究の中心である。
abstractThis study investigates the effectiveness of utilizing different plant recognition algorithms applied to UAV-derived images for evaluating seeder performance based on detected plant spacings.
Background Quantitative disease resistance (QR) is a complex, dynamic trait that is most reliably quantified in field-grown crops. Traditional disease assessments offer limited potential to disentangle the contributions of different components to overall QR at critical crop developmental stages. Yet, a better functional understanding of QR could greatly support a more targeted, knowledge-based selection for QR and improve predictions of seasonal epidemics. Image-based approaches together with advanced image processing methodologies recently emerged as valuable tools to standardize relevant disease assessments, increase measurement throughput, and describe diseases along multiple dimensions. Results We present a simple, affordable, and easy-to-operate imaging set-up and imaging procedure for in-field acquisition of wheat leaf image sequences. The development of Septoria tritici blotch and leaf rusts was monitored over time via robust methods for symptom detection and segmentation, spatial alignment of images, symptom tracking, and leaf- and symptom characterization. The average accuracy of the spatial alignment of images in a time series was approximately 5 pixels (~ 0.15 mm). Leaf-level symptom counts as well as individual symptom property measurements revealed stable patterns over time that were generally in excellent agreement with visual impressions. This provided strong evidence for the robustness of the methodology to variability typically inherent in field data. Contrasting patterns in the number of lesions resulting from separate infection events and lesion expansion dynamics were observed across wheat genotypes. The number of separate infection events and average lesion size contributed to different degrees to overall disease intensity, possibly indicating distinct and complementary mechanisms of QR. Conclusions The proposed methodology enables rapid, non-destructive, and reproducible measurement of several key epidemiological parameters under field conditions. Such data can support decomposition and functional understanding of QR as well as the parameterization, fine-tuning, and validation of epidemiological models. Details of pathogenesis can translate into specific symptom phenotypes resolvable using time series of high-resolution RGB images, which may improve biological understanding of plant-pathogen interactions as well as interactions in disease complexes.
Why it matches plant phenotyping methods圃場での植物病徴を画像から取得・追跡・定量する撮像および画像解析手法を開発・検証しており、植物表現型測定が中心である。
abstractWe present a simple, affordable, and easy-to-operate imaging set-up and imaging procedure for in-field acquisition of wheat leaf image sequences.
Reproduction assets foundThe paper explicitly states that all image-processing/analysis code is publicly available on the authors' GitHub repository, and that a sample dataset plus the trained reference mark detection model are downloadable from the ETH Research Collection. Both are paper-specific, public, and actionable.Code · publicAll code related to the processing of image time series and leaf- and lesion-level trait extraction is available from https://github.com/and-jonas/sympathique-wheat for documentation.Open asset ↗and-jonas/sympathique-wheatlines:98-107Dataset · publicA sample data set and the trained reference mark detection model can be downloaded from ETH research collection at https://doi.org/10.3929/ethz-b-000659812 .Open asset ↗10.3929/ethz-b-000659812lines:98-107Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
The increase in the global population is leading to a doubling of the demand for protein. Soybean ( Glycine max ), a key contributor to global plant-based protein supplies, requires ongoing yield enhancements to keep pace with increasing demand. Precise, on-plant seed counting and localization may catalyze breeding selection of shoot architectures and seed localization patterns related to superior performance in high planting density and contribute to increased yield. Traditional manual counting and localization methods are labor-intensive and prone to error, necessitating more efficient approaches for yield prediction and seed distribution analysis. To solve this, we propose MSANet: a novel deep learning framework tailored for counting and localization of soybean seeds on mature field-grown soy plants. A multi-scale attention map mechanism was applied to maximize model performance in seed counting and localization in soybean breeding fields. We compared our model with a previous state-of-the-art model using the benchmark dataset and an enlarged dataset, including various soybean genotypes. Our model outperforms previous state-of-the-art methods on all datasets across various soybean genotypes on both counting and localization tasks. Furthermore, our model also performed well on in-canopy 360° video, dramatically increasing data collection efficiency. We also propose a technique that enables previously inaccessible insights into the phenotypic and genetic diversity of single plant vertical seed distribution, which may accelerate the breeding process. To accelerate further research in this domain, we have made our dataset and software publicly available: https://github.com/UTokyo-FieldPhenomics-Lab/MSANet.
Why it matches plant phenotyping methods大豆種子の計数・位置推定と垂直分布という植物形質を対象に、深層学習手法を開発・比較検証し、データセットとソフトウェアも公開しているため、フェノタイピング手法が中心である。
abstractwe propose MSANet: a novel deep learning framework tailored for counting and localization of soybean seeds on mature field-grown soy plants.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicwe have made our dataset and software publicly available: https://github.com/UTokyo-FieldPhenomics-Lab/MSANet .Open asset ↗UTokyo-FieldPhenomics-Lab/MSANetlines:1-25Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Plant population counts are highly valued by crop producers as important early-season indicators of field health. Traditionally, emergence rate estimates have been acquired through manual counting, an approach that is labor-intensive and relies heavily on sampling techniques. By applying deep learning-based object detection models to aerial field imagery, accurate plant population counts can be obtained for much larger areas of a field. Unfortunately, current detection models often perform poorly when they are faced with image conditions that do not closely resemble the data found in their training sets. In this paper, we explore how specific facets of a plant detector's training set can affect its ability to generalize to unseen image sets. In particular, we examine how a plant detection model's generalizability is influenced by the size, diversity, and quality of its training data. Our experiments show that the gap between in-distribution and out-of-distribution performance cannot be closed by merely increasing the size of a model's training set. We also demonstrate the importance of training set diversity in producing generalizable models, and show how different types of annotation noise can elicit different model behaviors in out-of-distribution test sets. We conduct our investigations with a large and diverse dataset of canola field imagery that we assembled over several years. We also present a new web tool, Canola Counter, which is specifically designed for remote-sensed aerial plant detection tasks. We use the Canola Counter tool to prepare our annotated canola seedling dataset and conduct our experiments. Both our dataset and web tool are publicly available.
Why it matches plant phenotyping methods航空画像からカノーラ個体数（個体群密度）を推定する検出モデルの汎化性能を検証し、注釈付きデータセットと専用Webツールを提示しており、植物表現型取得手法が中心である。
abstractBy applying deep learning-based object detection models to aerial field imagery, accurate plant population counts can be obtained for much larger areas of a field.
Reproduction assets foundThe paper's aerial canola seedling dataset (images and annotations) is publicly deposited on Zenodo, and the authors' Canola Counter analysis/annotation tool is open source on GitHub. The arXiv 2108.05789 entry is a cited prior work (CropAndWeed dataset), not a paper-specific asset.Dataset · publicData Availability Statement
The canola seedling dataset used in this study is publicly available and can be found at: https://doi.org/10.5281/zenodo.11055599 . The Canola Counter tool is open source and is available at: https://github.com/eandvaag/agricounter .Open asset ↗zenodo · 10.5281/zenodo.11055599lines:236-237Code · publicData Availability Statement
The canola seedling dataset used in this study is publicly available and can be found at: https://doi.org/10.5281/zenodo.11055599 . The Canola Counter tool is open source and is available at: https://github.com/eandvaag/agricounter .Open asset ↗github · eandvaag/agricounterlines:236-237Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Introduction The emergence rate of crop seedlings is an important indicator for variety selection, evaluation, field management, and yield prediction. To address the low recognition accuracy caused by the uneven size and varying growth conditions of crop seedlings under salt-alkali stress, this research proposes a peanut seedling recognition model, MS-YOLOv8. Methods This research employs close-range remote sensing from unmanned aerial vehicles (UAVs) to rapidly recognize and count peanut seedlings. First, a lightweight adaptive feature fusion module (called MSModule) is constructed, which groups the channels of input feature maps and feeds them into different convolutional layers for multi-scale feature extraction. Additionally, the module automatically adjusts the channel weights of each group based on their contribution, improving the feature fusion effect. Second, the neck network structure is reconstructed to enhance recognition capabilities for small objects, and the MPDIoU loss function is introduced to effectively optimize the detection boxes for seedlings with scattered branch growth. Results Experimental results demonstrate that the proposed MS-YOLOv8 model achieves an AP50 of 97.5% for peanut seedling detection, which is 12.9%, 9.8%, 4.7%, 5.0%, 11.2%, 5.0%, and 3.6% higher than Faster R-CNN, EfficientDet, YOLOv5, YOLOv6, YOLOv7, YOLOv8, and RT-DETR, respectively. Discussion This research provides valuable insights for crop recognition under extreme environmental stress and lays a theoretical foundation for the development of intelligent production equipment.
Why it matches plant phenotyping methodsUAVリモートセンシング画像からピーナッツ幼苗を認識・計数するモデルを開発し、検出性能を比較検証している。幼苗数・出現率という植物状態の推定が研究の中心である。
abstractthis research proposes a peanut seedling recognition model, MS-YOLOv8
Reproduction assets foundThe paper's data availability statement explicitly deposits the peanut seedling UAV image dataset (and associated model resources) in a public GitHub repository, matching an allowed URL.Dataset · publicy close-range remote sensing. It provides a certain theoretical guidance for the development of an intelligent monitoring platform for peanut.
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://github.com/zfvincent1997/MS-YOLOV8 .
Author contributions
FZ: Investigation, Resources, Software, Writing – original draft. LZ: Conceptualization, Supervision, Writing – review & editing. DW: Investigation, Writing – review & editing. JW: Investigation, Writing – review & editing. IS: Software, Visualization, Writing – review & editing. JL: Conceptualization, Open asset ↗https://github.com/zfvincent1997/MS-YOLOV8lines:667-765Plant phenotyping relevance match · UnverifiedEurope PMC · OpenAlex · Crossref · checked 14 Sept 2026
Artificial intelligence and machine learning (AI/ML) can be used to automatically analyze large image datasets. One valuable application of this approach is estimation of plant trait data contained within images. Here we review 39 papers that describe the development and/or application of such models for estimation of stomatal traits from epidermal micrographs. In doing so, we hope to provide plant biologists with a foundational understanding of AI/ML and summarize the current capabilities and limitations of published tools. While most models show human-level performance for stomatal density (SD) quantification at superhuman speed, they are often likely to be limited in how broadly they can be applied across phenotypic diversity associated with genetic, environmental, or developmental variation. Other models can make predictions across greater phenotypic diversity and/or additional stomatal/epidermal traits, but require significantly greater time investment to generate ground-truth data. We discuss the challenges and opportunities presented by AI/ML-enabled computer vision analysis, and make recommendations for future work to advance accelerated stomatal phenotyping.
Why it matches plant phenotyping methodsAI/ML画像解析による気孔形質推定を扱うレビューであり、植物フェノタイピング手法の開発・応用、性能と限界の評価が中心です。
abstractHere we review 39 papers that describe the development and/or application of such models for estimation of stomatal traits from epidermal micrographs.
Plant physiology and metabolism rely on the function of stomata, structures on the surface of above-ground organs that facilitate the exchange of gases with the atmosphere. The morphology of the guard cells and corresponding pore that make up the stomata, as well as the density (number per unit area), are critical in determining overall gas exchange capacity. These characteristics can be quantified visually from images captured using microscopy, traditionally relying on time-consuming manual analysis. However, deep learning (DL) models provide a promising route to increase the throughput and accuracy of plant phenotyping tasks, including stomatal analysis. Here we review the published literature on the application of DL for stomatal analysis. We discuss the variation in pipelines used, from data acquisition, pre-processing, DL architecture, and output evaluation to post-processing. We introduce the most common network structures, the plant species that have been studied, and the measurements that have been performed. Through this review, we hope to promote the use of DL methods for plant phenotyping tasks and highlight future requirements to optimize uptake, predominantly focusing on the sharing of datasets and generalization of models as well as the caveats associated with utilizing image data to infer physiological function.
Why it matches plant phenotyping methods気孔画像から形態・密度などの植物形質を推定する深層学習手法を体系的にレビューしており、フェノタイピング手法が中心である。
abstractHere we review the published literature on the application of DL for stomatal analysis.
Accurately estimating the yield of citrus fruit on individual trees is essential for precise orchard management and the income of producers. However, estimating the yield of citrus fruit from images of trees remains challenging among different processes of tree pruning and image acquisition. This study adopted a deep learning based detection model to count fruit in tree images and machine learning models to estimate the yield of individual trees from the fruit count. Trees under four levels of pruning intensity (no pruning, 0–5 %, 5–10 %, and 10–15 % of new sprouts pruned) and imaged from three different views (two, four, and six images per tree) to determine the optimal conditions for yield estimation. The variables considered for yield estimation included fruit count, pruning intensity and image views. Dataset containing 1200 tree images were used to train and test four machine learning models: random forest, support vector machine, extreme gradient boosting (XGBoost), and generalized linear model. The XGBoost model achieved the lowest errors in both training and testing. The optimal yield estimation occurs when there are two, four, and six image views and trees that have been pruned >10 %, 5–10 %, and ≤5 %, respectively. The findings can enhance the accuracy of image based citrus fruit yield estimation for individual trees and reveal the influences of pruning and image views.
Why it matches plant phenotyping methods個体樹の果実画像から果実数を検出し、機械学習で収量を推定する手法が研究の中心であり、画像枚数と剪定条件による精度も評価している。
abstractThis study adopted a deep learning based detection model to count fruit in tree images and machine learning models to estimate the yield of individual trees from the fruit count.
Wheat head or spike detection is significant for phenotyping because it can be directly correlated to yield and is an indicator of yield potential. Historically, wheat head counting was a labor-intensive and error prone process. Use of Deep Learning (DL) techniques has automated this process allowing automated wheat head detection and counting using high resolution imagery, allowing large-scale, High Throughput Phenotyping (HTP). Despite the use of advanced technologies, wheat head detection is a challenging task due to high environmental variability, cultivar differences, and head overlap. Several attempts have been made to make the DL models more robust and the wheat head datasets more diverse to improve the detection accuracy and reliability. With the introduction of advanced DL architectures, there has been continuous improvement in accuracy of head detection. In this study, we have evaluated the performance of three different cutting-edge DL models - YOLOv10x, RetinaNet, and MM-Grounding DINO for wheat head detection. We have also combined two different wheat datasets, Global Wheat Head Detection (GWHD) 2021 and SPIKE dataset to get a diverse dataset with a wide range of genotypes. This study aims to contribute to the ongoing evolution of wheat head detection techniques and provide an insight into how these three models perform for this task.
Why it matches plant phenotyping methodsコムギ穂の検出・計数という植物形態形質の画像解析手法を対象に、複数の深層学習モデルを評価し、異なるデータセットを統合して性能を比較しているため、フェノタイピング手法が中心である。
abstractUse of Deep Learning (DL) techniques has automated this process allowing automated wheat head detection and counting using high resolution imagery, allowing large-scale, High Throughput Phenotyping (HTP).
Phenotyping of selected traits is an important prerequisite for evaluating genetic resources and providing resistant genotypes for subsequent fruit breeding. In order to establish a high-throughput method for more objective and accurate phenotyping in the future, the aim of our study was to develop a UVA-based digital phenotyping method in the field. European pear rust (Gymnosporangium sabinae) was selected as a model pathogen for this purpose because without pesticide application it is widely distributed in pear orchards and shows conspicuous yellow-orange disease symptoms. In 2021 and 2022, 705 images showing symptoms of European pear rust were taken in the experimental field of the Julius Kühn Institute in Dresden-Pillnitz and the symptoms were labeled using the Computer Vision Annotation Tool (CVAT). Model training was performed based on four pre-trained YOLOv5 algorithms that use an object detector approach and allow unique identification of each symptom in an image. Accurate localization of disease symptoms within the orchard was enabled by a novel photogrammetry approach on georeferenced image data. For subsequent quantification of disease symptoms per genotype, the number of infected leaves was related to the total volume of the tree. In the future, this digital phenotyping system will provide a high-throughput method for evaluating European pear rust in pear genetic resources.
Why it matches plant phenotyping methodsナシさび病の症状を画像から検出・局在化・定量するUAVデジタルフェノタイピング手法の開発が研究の中心であり、植物病害状態を直接測定する。
abstractthe aim of our study was to develop a UVA-based digital phenotyping method in the field
In the production management of agriculture, accurate fruit counting plays a vital role in the orchard yield estimation and appropriate production decisions. Although recent tracking-by-detection algorithms have emerged as a promising fruit-counting method, they still cannot completely avoid fruit occlusion and light variations in complex orchard environments, and it is difficult to realize automatic and accurate apple counting. In this paper, a video-based multiple-object tracking method, MR-SORT (Multiple Rematching SORT), is proposed based on the improved YOLOv8 and BoT-SORT. First, we propose the AD-YOLO model, which aims to reduce the number of incorrect detections during object tracking. In the YOLOv8s backbone network, an Omni-dimensional Dynamic Convolution (ODConv) module is used to extract local feature information and enhance the model's ability better; a Global Attention Mechanism (GAM) is introduced to improve the detection ability of a foreground object (apple) in the whole image; a Soft Spatial Pyramid Pooling Layer (SSPPL) is designed to reduce the feature information dispersion and increase the sensory field of the network. Then, the improved BoT-SORT algorithm is proposed by fusing the verification mechanism, SURF feature descriptors, and the Vector of Local Aggregate Descriptors (VLAD) algorithm, which can match apples more accurately in adjacent video frames and reduce the probability of ID switching in the tracking process. The results show that the mAP metrics of the proposed AD-YOLO model are 3.1% higher than those of the YOLOv8 model, reaching 96.4%. The improved tracking algorithm has 297 fewer ID switches, which is 35.6% less than the original algorithm. The multiple-object tracking accuracy of the improved algorithm reached 85.6%, and the average counting error was reduced to 0.07. The coefficient of determination R2 between the ground truth and the predicted value reached 0.98. The above metrics show that our method can give more accurate counting results for apples and even other types of fruit.
Why it matches plant phenotyping methodsリンゴ果実の検出・追跡・計数手法を開発し、精度や計数誤差を検証しており、果実数という植物形質の取得が中心である。
Accurate identification and estimation of the population densities of microscopic, soil-dwelling plant-parasitic nematodes (PPNs) are essential, as PPNs cause significant economic losses in agricultural production systems worldwide. This study presents a comprehensive review of emerging techniques used for the identification of PPNs, including morphological identification, molecular diagnostics such as polymerase chain reaction (PCR), high-throughput sequencing, meta barcoding, remote sensing, hyperspectral analysis, and image processing. Classical morphological methods require a microscope and nematode taxonomist to identify species, which is laborious and time-consuming. Alternatively, quantitative polymerase chain reaction (qPCR) has emerged as a reliable and efficient approach for PPN identification and quantification; however, the cost associated with the reagents, instrumentation, and careful optimisation of reaction conditions can be prohibitive. High-throughput sequencing and meta-barcoding are used to study the biodiversity of all tropical groups of nematodes, not just PPNs, and are useful for describing changes in soil ecology. Convolutional neural network (CNN) methods are necessary to automate the detection and counting of PPNs from microscopic images, including complex cases like tangled nematodes. Remote sensing and hyperspectral methods offer non-invasive approaches to estimate nematode infestations and facilitate early diagnosis of plant stress caused by nematodes and rapid management of PPNs. This review provides a valuable resource for researchers, practitioners, and policymakers involved in nematology and plant protection. It highlights the importance of fast, efficient, and robust identification protocols and decision-support tools in mitigating the impact of PPNs on global agriculture and food security.
Why it matches plant phenotyping methods植物寄生性線虫による植物ストレス・感染状態の推定に関する画像処理、CNN、リモートセンシング、ハイパースペクトル手法を体系的にレビューしており、植物フェノタイピング手法が中心です。
abstractThis study presents a comprehensive review of emerging techniques used for the identification of PPNs, including morphological identification, molecular diagnostics such as polymerase chain reaction (PCR), high-throughput sequencing, meta barcoding, remote sensing, hyperspectral analysis, and image processing.
The number of panicles per unit area (PNpA) is one of the key factors contributing to the grain yield of rice crops. Accurate PNpA quantification is vital for breeding high-yield rice cultivars. Previous studies were based on proximal sensing with fixed observation platforms or unmanned aerial vehicles (UAVs). The near-canopy images produced in these studies suffer from inefficiency and complex image processing pipelines that require manual image cropping and annotation. This study aims to develop an automated, high-throughput UAV imagery-based approach for field plot segmentation and panicle number quantification, along with a novel classification method for different panicle types, enhancing PNpA quantification at the plot level. RGB images of the rice canopy were efficiently captured at an altitude of 15 m, followed by image stitching and plot boundary recognition via a mask region-based convolutional neural network (Mask R-CNN). The images were then segmented into plot-scale subgraphs, which were categorized into 3 growth stages. The panicle vision transformer (Panicle-ViT), which integrates a multipath vision transformer and replaces the Mask R-CNN backbone, accurately detects panicles. Additionally, the Res2Net50 architecture classified panicle types with 4 angles of 0°, 15°, 45°, and 90°. The results confirm that the performance of Plot-Seg is comparable to that of manual segmentation. Panicle-ViT outperforms the traditional Mask R-CNN across all the datasets, with the average precision at 50% intersection over union (AP 50 ) improved by 3.5% to 20.5%. The PNpA quantification for the full dataset achieved superior performance, with a coefficient of determination ( R 2 ) of 0.73 and a root mean square error (RMSE) of 28.3, and the overall panicle classification accuracy reached 94.8%. The proposed approach enhances operational efficiency and automates the process from plot cropping to PNpA prediction, which is promising for accelerating the selection of desired traits in rice breeding.
Why it matches plant phenotyping methodsイネの穂数・穂型という植物形質を、UAV画像と画像解析・深層学習で自動抽出する手法を開発し、精度検証しているため、方法が研究の中心である。
abstractThis study aims to develop an automated, high-throughput UAV imagery-based approach for field plot segmentation and panicle number quantification, along with a novel classification method for different panicle types
Published20 Oct 20242024 IEEE International Conference on Automation/XXVI Congress of the Chilean Association of Automatic Control (ICA-ACCA)Cited by 2 · OpenAlex ↗
The banana is a crucial crop in tropical regions, facing challenges from diseases such as black Sigatoka, which affect its production and quality due to defoliation. This research proposes the use of drones equipped with high-resolution RGB cameras to capture images of banana plantations, employing a hybrid deep learning model that combines detection and semantic segmentation to accurately identify and count banana leaves. Additionally, the metadata from the images provided the geographical coordinates of each plant, exported in shapefiles compatible with Geographic Information Systems (GIS). The results show high accuracy in detection (98.5%) and leaf counting (93.45%), surpassing previous, more costly methods. This facilitates the identification of areas affected by diseases, evidenced in the detection of potential black Sigatoka outbreaks. The ability to make informed decisions based on this data improves agricultural management, promoting sustainable practices and optimizing crop quality and productivity.
Why it matches plant phenotyping methodsドローン画像と深層学習・セマンティックセグメンテーションによりバナナ葉を検出・計数する手法が中心で、葉数という植物形質を定量化しているため。
abstractemploying a hybrid deep learning model that combines detection and semantic segmentation to accurately identify and count banana leaves