Modern crop breeding demands precise organ-level analysis for trait quantification, making plant point cloud segmentation (PPCS) increasingly important. However, conventional deep learning approaches rely heavily on densely annotated datasets that are labor-intensive to acquire. Unified PPCS adaptation from distribution-shifted examples with minimal additional training remains challenging. To address this, we propose PlantC2USeg, a deep transfer learning framework featuring cross-scale consistency learning to explicitly align features across spatial scales and an information-restricted decoding strategy that prevents reconstruction shortcuts and promotes robust adaptation. The resulting pre-training enables stable few-shot generalization across species and sensing conditions, while unified fine-tuning with inherited thresholds further reduces adaptation overhead. Under full supervision on Soybean3D, PlantC2USeg achieves the highest semantic IoU and instance mWCov among compared methods, at 91.91% and 94.62%. With 20 labeled samples, it leads both metrics at 89.78% and 90.27%; with only 10 samples, it retains the highest mWCov of 83.23% while achieving 83.19% IoU. Across HR3D, 10-shot transfer to tobacco, tomato, and sorghum averages 78.41% IoU and 79.42% mWCov, while 22-shot transfer to SYAU-Maize achieves the highest IoU and mRec at 92.75% and 93.51%. Furthermore, a leading category-averaged mIoU of 85.0% on ShapeNet Part demonstrates the framework's capability to handle diverse shape variations beyond agricultural domains. These results demonstrate that PlantC2USeg reduces overall adaptation effort under distribution shifts, enabling scalable plant phenotyping and transferable 3D representation learning beyond agriculture.
Why it matches plant phenotyping methods植物点群の器官レベル形質定量を目的とするセグメンテーション手法を開発し、複数データセット・作物・ショット条件で性能評価しているため、植物フェノタイピング手法が中心である。
abstractwe propose PlantC2USeg, a deep transfer learning framework featuring cross-scale consistency learning
Hyperspectral sensors have emerged as a promising approach in the study of plant diseases. The objective was to distinguish between healthy and inoculated seeds, and also to distinguish between genera of plant-pathogenic fungi in soybean seeds, using hyperspectral sensors combined with machine learning. The experimental design was a fully randomized factorial design with six algorithms (Simple Logistic Regression, Support Vector Machine, Artificial Neural Network, Random Forest, REPTree and J48 decision trees) and four phytopathogens (Sclerotinia sclerotiorum, Macrophomina phaseolina, Rhizoctonia solani, and Colletotrichum sp.) plus the control. Spectral analysis of the seeds was performed using a spectroradiometer (Ocean Optics) consisting of two sensors: NIR and Flame, covering the spectrum from 350 to 2500 nm. It was possible to distinguish between healthy and inoculated seeds, as well as identify the type of phytopathogen, based on each spectral signature. The Simple Logistic Regression and Support Vector Machine algorithms performed best. Hyperspectral sensors combined with machine learning constitute a promising tool for the detection of phytopathogens in seeds, enabling rapid and non-destructive analysis. This promising tool could serve as a complementary alternative to traditional diagnostic methods, which, although accurate, are time-consuming and rely on specialized labor.
Why it matches plant phenotyping methods種子の健全・感染状態を非破壊的に推定するハイパースペクトルセンシングと機械学習が研究の中心であり、感染植物器官の状態を直接測定する方法として扱える。
abstractThe objective was to distinguish between healthy and inoculated seeds, and also to distinguish between genera of plant-pathogenic fungi in soybean seeds, using hyperspectral sensors combined with machine learning.
To overcome YOLOv11’s limitations in complex field environments, this paper proposed SDD-YOLOv11n, a lightweight real-time detector for soybean diseases. The model reconstructed the backbone using GhostConv to minimize redundancy and integrates a C3k2_Star module to enhance small lesion detection against background noise. Additionally, a Detect Efficient (DE) head further compressed the architecture. Experimental results verified the model's efficiency, achieving a parameter count of 1.88 M and a weight size of 3.9 MB—reductions of 27.3% and 25% compared to YOLOv11n, respectively. Furthermore, the model maintained high detection performance with a Mean Average Precision (mAP50) of 75.6% and an F1-score of 69.3%, demonstrating its effectiveness in balancing architectural efficiency and accuracy in complex field environments.
Why it matches plant phenotyping methods圃場のダイズ病害病斑を画像から検出する軽量YOLOモデルの開発と性能評価が研究の中心であり、植物の病害状態を直接推定するフェノタイピング手法に該当する。
titleREAL-TIME AND PRECISE DETECTION OF FIELD SOYBEAN RUST AND BACTERIAL SPOT BASED ON IMPROVED YOLOV11N
Goal. To substantiate the methodical approaches to analyzing the state of soybean crops based on the results of aerial photography with UAV by comparing manual vectorization, controlled classification according to the algorithm of maximum similarity, and uncontrolled classification of K-Means, as well as to determine the feasibility of their combination with expert visual interpretation to assess the spatial structure of the vegetation cover. Methods. Aerial photography of the test proving ground was performed with the help of the unmanned aerial vehicle DJI Phantom 4 Advanced with the subsequent photogrammetric study of materials and the formation of a highly detailed orthophotoplane. In the QGIS environment, visual decryption and manual vectorization of the main objects of the agrolandscape were carried out with the creation of polygonal layers. For automated mapping, methods of controlled classification according to the algorithm of maximum similarity and uncontrolled classification based on the K-Means algorithm were used. The accuracy of the results was evaluated by comparing the data of automated classifications with the data of manual digitization, which was used as a reference (control) method. On the basis of the results obtained, empirical data were summarized to justify practical recommendations for the application of the studied approaches. Results. The study was conducted on the territory of the research farm of the Separate subdivision of the National University of Life and Environmental Sciences of Ukraine «Berezhany Agrotechnical Institute» (vil. Pavliv, Ternopil district, Ternopil oblast) (49.452057°N; 24,805818°E) in may – august 2025. The obtained cartographic materials made it possible to quantify the areas of the main objects of the agro-landscape and identify problem areas with sparse shoots. Methods of controlled classification showed greater compliance with the digitization data (average deviation — 6.7%) compared to uncontrolled (14.7%), which were effective from the point of view of preliminary assessment of spectrally homogeneous sections, but did not provide an accurate division of shoots by density. Analysis of the spatial structure of coverage made it possible to plan local agrotechnical measures and assess the potential yield. Conclusions. Methodical approaches to analyzing the state of soybean crops based on the results of aerial photography with a UAV equipped with RGB cameras are promising and economically feasible. Automated classification methods are effective for highlighting hard and contrasting objects and small-contoured areas, while a detailed assessment of the structure of the vegetation cover is advisable to carry out using a controlled classification in combination with expert visual interpretation.
Why it matches plant phenotyping methodsUAV画像を用いて大豆作物の植生被覆構造や疎な出芽域を抽出し、複数の分類法を手動ベクトル化と比較・精度評価している。植物状態の取得手法自体が研究の中心である。
abstractTo substantiate the methodical approaches to analyzing the state of soybean crops based on the results of aerial photography with UAV by comparing manual vectorization, controlled classification according to the algorithm of maximum similarity, and uncontrolled classification of K-Means
Abstract Background and aims Plant volatile organic compounds (VOCs) change dynamically with plant development and in response to environmental conditions. However, their potential as non-invasive indicators of phenological progression remains poorly explored. In this study, we developed a framework integrating automated VOC sampling, time-resolved VOC profiling, and machine-learning analysis for the non-invasive assessment of plant phenology. Using soybean ( Glycine max (L.) Merr.), we investigated whether development-associated temporal variation in VOC emissions could delineate and predict developmental phases. Methods We collected VOCs daily under controlled environmental conditions from 16 to 43 days after sowing, spanning the transition from vegetative to reproductive stages, using an automated sampling system coupled with thermal desorption-gas chromatograph-mass spectrometer (TD- GC-MS). To characterise temporal changes in VOC profiles associated with phenological progression, we analysed the daily VOC data using a multi-step pipeline combining statistical filtering and similarity-based network analysis. We defined VOC-derived developmental phases from similarity patterns in the VOC profiles, then developed and evaluated machine-learning models to predict these phases. Key results Seven VOCs exhibited distinct phase-dependent dynamics, including green leaf volatiles and monoterpenes showing characteristic temporal changes during phenological progression. Network-based clustering of VOC profiles resolved five developmental phases closely aligned with conventional developmental stages. A machine-learning model predicted these phases from the VOC profiles with high predictive accuracy on independent test data, demonstrating that phenological progression could be quantitatively inferred from VOC emission patterns. Conclusions Our findings support VOC profiling as a reliable and non-invasive approach for assessing phenological progression in soybean. By extracting temporally structured VOC signals, this framework captures developmental information that may be difficult to obtain through visual observation alone, particularly after canopy closure. VOC profiling offers a practical tool for monitoring crop developmental dynamics and has broader potential for plant phenotyping and precision crop management.
Why it matches plant phenotyping methods自動VOCサンプリング、時系列VOCプロファイリング、機械学習を統合し、VOCから植物の発育段階を非破壊推定する方法を開発・評価しており、フェノタイピング手法が中心である。
abstractwe developed a framework integrating automated VOC sampling, time-resolved VOC profiling, and machine-learning analysis for the non-invasive assessment of plant phenology.
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 peak area matrix obtained from the MS- DIAL analysis (Supplementary Dataset S1) was filtered to remove unreliable features.Open asset ↗lines:66-69Plant phenotyping relevance match · UnverifiedOpenAlex · checked 5 Sept 2026
Introduction Leaf shape is a genetically determined crop phenotype, and its accurate classification underpins soybean germplasm assessment and genetic improvement. Manual classification is highly subjective and struggles to distinguish morphologically similar leaves, while mainstream supervised classification demands large labeled datasets and incurs high development costs. Efficient feature frameworks for soybean leaf categorization are still insufficient. Methods In this study, 581 biologically replicated terminal leaflets sampled from 194 soybean varieties were analyzed at the single-leaflet level using traditional morphological indices and novel leaf contour angular features. Unsupervised K-means clustering was used to classify soybean leaflet morphological phenotypes; t-SNE was applied exclusively for dimensional reduction visualization, while Welch's ANOVA combined with Games-Howell post-hoc tests was adopted to detect inter-cluster phenotypic differences. Clustering stability and external consistency against manual visual labeling were further quantified via Adjusted Rand Index to comprehensively verify the reliability of grouping outputs. Results The results revealed no significant difference in leaflet edge complexity (p = 0.41) between two manually divided leaf groups distinguished by overall leaf outline similarity; these two morphologically similar leaf clusters failed to be fully separated even though the first two principal components accounted for 90.2% of total variance. For K-means clustering, k = 3 achieved better overall performance with a Calinski–Harabasz (CH) index of 395.55, Davies–Bouldin (DB) index of 1.03, and silhouette coefficient (SC) of 0.38, compared with k = 4. Nevertheless, the angular feature attained an F-value of 951.62 in driving sample reallocation across clusters, serving as the core indicator for fine subdivision at k = 4. Under k = 4 clustering, all six morphological indices differed significantly among the four groups (p < 0.05). Additionally, the number of cross-clustered samples increased from 66 to 119 as k rose from 3 to 4, with 96.6% of cross-clustering attributed to the leaflet contour angular feature. Discussion This research provides a novel reference and technical support for the automated identification and fine classification of soybean leaf morphology.
Why it matches plant phenotyping methods大豆小葉の形態表現型を角度特徴量とクラスタリングで自動分類する手法が研究の中心であり、検証指標も明示されているため。
abstractLeaf shape is a genetically determined crop phenotype, and its accurate classification underpins soybean germplasm assessment and genetic improvement.
Abstract Technological advances have expanded the adoption of digital technologies in agriculture, helping to reduce labour effort, increase profitability, improve crop efficiency and productivity, enhance product quality, mitigate environmental impacts, and promote human health. This context also extends to soybean farming, a sector of major economic importance in Brazil. Most importantly, Brazil has the global leadership in soybean production with biological nitrogen fixation (BNF) replacing chemical fertilisers. The research and evaluation of BNF is limited by manual counting of nodules, a time-consuming procedure. This study presents SoyNodules, designed for the automatic identification of soybean nodules, consisting of a dataset of images. The dataset includes 1,701 images acquired under controlled conditions: 1,662 images of soybean roots with nodules and 39 images of isolated nodules without roots. A total of 49,210 nodule instances are manually annotated with bounding boxes. SoyNodules was designed to promote reuse and interoperability in alignment with the FAIR principles (Findable, Accessible, Interoperable, Reusable) and to support the development, training, and evaluation of computer vision and deep learning methods for precision agriculture.
Why it matches plant phenotyping methods大豆根粒を自動識別する画像データセットであり、手作業計数の代替となる植物器官形質の抽出・評価を支援する方法論的データセット。
abstractThis study presents SoyNodules, designed for the automatic identification of soybean nodules, consisting of a dataset of images.
Reproduction assets foundThe paper is a data descriptor for SoyNodules, an annotated dataset of 1,701 soybean root/nodule images with 49,210 bounding-box annotations, publicly deposited on Zenodo with a DOI. The same repository also hosts the authors' annotation-format conversion script (AnyLabeling to Pascal VOC/COCO), per the Code AvailabilDataset · publicThe SoyNodules dataset, released as version 1.0, is publicly available on Zenodo [28]
at https://doi.org/10.5281/zenodo.22081914.Open asset ↗Zenodo · 10.5281/zenodo.22081914pdf-page:9 lines:1-43Plant phenotyping relevance match · UnverifiedOpenAlex · checked 8 Sept 2026
To establish an accurate and interpretable prediction framework for soybean lodging grade and clarify the core regulatory traits and differentiated driving mechanisms of soybean lodging under high-density drip irrigation cultivation, 356 spring soybean germplasm accessions were used as experimental materials in this study. Morphological and mechanical traits including plant height (PH), stem pulling force (SPF), internode number (IN) and petiole length (PL) were measured over two consecutive years of field phenotyping. Two composite evaluation indices, plant height/stem pulling force ratio (PH/SPF) and plant height/internode number ratio (PH/IN), were further constructed. Four machine learning algorithms were adopted to develop multi-classification models for soybean lodging grade prediction. SHAP analysis combined with three global sensitivity approaches (perturbation analysis, Sobol’ method and Morris screening) was applied to decipher the regulatory patterns of key traits. The results showed that lodging grade significantly affected soybean grain yield and explained 25–28% of the phenotypic yield variation; yield reduction tended to plateau under severe lodging. Compared with single indicators such as SPF and PL, the two derived composite indices could stably distinguish soybean accessions with different lodging grades and exhibited stronger discriminatory power. Model comparison revealed that the XGBoost model achieved optimal prediction accuracy and generalization stability for lodging grade, with a weighted F1-score of 95.34% on the test set, significantly outperforming the conventional linear model. Interpretability analysis demonstrated that the PH/IN, PH, and PH/SPF acted as the primary positive traits promoting lodging, while SPF was the sole protective trait. Driving factors of lodging presented obvious gradient heterogeneity: mild lodging was dominated by the imbalance of plant architecture ratio, whereas severe lodging was governed by the cumulative effects of PH and IN. Strong interactions existed among all measured traits. The interpretable machine learning framework established in this study can provide theoretical support and technical references for lodging-resistant germplasm screening and targeted plant architecture regulation for densely planted soybean under drip irrigation systems.
Why it matches plant phenotyping methods大豆の倒伏状態を形態・力学形質から機械学習で推定し、モデル性能比較と解釈性解析を行う枠組みが研究の中心であり、単なる生物学的実験の routine 測定ではない。
abstractTo establish an accurate and interpretable prediction framework for soybean lodging grade and clarify the core regulatory traits and differentiated driving mechanisms of soybean lodging
Background Global soybean production is constrained by scarce arable land, and standardized evaluation tools remain lacking for natural mixed saline-alkali stress, the predominant abiotic stress under field conditions. Objective This study aimed to establish a comprehensive saline-alkali tolerance evaluation system for soybean germplasms via integrated multivariate statistical methods, and screen core and auxiliary indicators for efficient germplasm identification. Methods Seventy-one soybean germplasms were tested under 90 mmol/L mixed saline-alkali stress (NaCl:Na 2 SO 4 :NaHCO 3 :Na 2 CO 3 = 1:9:9:1, pH 8.2) simulating natural saline-alkali soil. We quantified the saline-alkali tolerance coefficients (SATC) of 13 morphological and physiological indicators, followed by coefficient of variation (CV), principal component analysis (PCA), subordinate function, cluster analysis and regression modeling. Results Significant inter-germplasm variations in saline-alkali tolerance were detected, and indicators with CV > 0.35 ( e.g ., root length (RL), root fresh weight (RFW)) were screened as primary indices. PCA extracted five principal components with 87.18% cumulative variance contribution, and the integrated analytical pipeline categorized germplasms into five tolerance grades: eight highly tolerant, 24 moderately tolerant, 13 generally tolerant, 16 sensitive and 10 highly sensitive accessions. A high-precision prediction model was constructed ( D = 0.290 X 1 - 0.026 X 2 + 0.438 X 3 + 0.402 X 4 + 0.180 X 5 + 0.153 X 6 + 0.813 X 7 - 1.123; R 2 = 0.998, where X 1 - X 7 represent the SATC of germination rate (GR), RL, RFW, total fresh weight (TFW), shoot dry weight (SDW), root dry weight (RDW), and total dry weight (TDW), respectively). A Chi-squared Automatic Interaction Detection (CHAID) decision tree model was further developed and validated using 10-fold cross-validation, yielding a cross-validation risk value of 0.003, which was comparable to the resubstitution risk value (0.002), indicating good generalization ability and low risk of overfitting. A novel five-dimensional overlapping analysis identified RFW as the core evaluation indicator, with RDW, TFW and R/S as key auxiliary indicators. Conclusion This study delivers a standardized, reproducible technical framework for large-scale screening of saline-alkali-tolerant soybean germplasms. It facilitates global saline-alkali land utilization, accelerates worldwide soybean stress-tolerance breeding, and provides a transferable paradigm for stress tolerance evaluation in other major crops.
Why it matches plant phenotyping methodsダイズの耐塩・耐アルカリ性を評価するための形態・生理形質の統合評価体系、予測モデル、指標選定、交差検証を中心的に開発・検証しており、再利用可能な植物表現型評価手法に該当する。
abstractThis study aimed to establish a comprehensive saline-alkali tolerance evaluation system for soybean germplasms via integrated multivariate statistical methods, and screen core and auxiliary indicators for efficient germplasm identification.
Lodging is a major yield-limiting factor in soybean, but efficient large-scale phenotyping and genetic dissection of this complex trait remain challenging for breeding programs. To bridge this gap, this study developed an integrated, breeding-oriented framework that links UAV-based high-throughput phenotyping with candidate gene identification. Field experiments involving 741 diverse soybean genotypes were conducted over two years, with UAV remote sensing performed at key reproductive stages (from R5 to R7). We identified UAV-derived structural (relative plant height), textural (homogeneity, dissimilarity, correlation), and spectral (NDVI, EVI, NDRE) features as the most sensitive indices for retrieving lodging severity. The fusion of these complementary features, coupled with the XGBoost algorithm, achieved high classification accuracy (0.81–0.92) across genotypes, growth stages, and years. This reliable phenotyping pipeline enabled the precise selection of contrasting genotypes (lodging-resistant vs. lodging-prone) for transcriptomic analysis. Transcriptome sequencing revealed 13,447 differentially expressed genes, with significant enrichment in phenylpropanoid and starch–sucrose metabolic pathways. Moreover, the haplotype analysis within a natural population identified superior allelic variants of two candidate genes ( Glyma.19G249100 and Glyma.05G142200 ) significantly associated with soybean lodging resistance. This work can effectively bridge the gap between scalable field phenotyping and the discovery of functionally validated breeding targets, providing an efficient and translational framework to accelerate the development of lodging-resistant soybean varieties.
Why it matches plant phenotyping methodsUAV画像・リモートセンシング特徴量とXGBoostを統合し、ダイズの倒伏重症度を大規模に推定・検証する育種向け表現型解析パイプラインが研究の中心である。
abstractthis study developed an integrated, breeding-oriented framework that links UAV-based high-throughput phenotyping with candidate gene identification
Abstract Heat stress causes ultrastructural damage in pollen grains, leading to reduced pollen germination, pollen size and shortened pollen tube length, ultimately lowering seed set and yield. This study presents a high-throughput phenotyping framework that integrates controlled-environment pollen germination assays with deep learning–based object detection for rapid, accurate, and scalable evaluation of reproductive heat tolerance in soybean breeding programs. Sixteen soybean genotypes were grown under controlled environments at optimal (28/18°C; day/night) and high temperature (38/28°C; day/night) regimes during flowering. In vitro pollen germination was quantified using six YOLO (You Only Look Once) object-detection architectures (YOLOv7–YOLOv12) to identify the best-performing model for automated analysis. Among the tested object-detection architectures, YOLOv9 achieved the best overall performance for detecting germinated and non-germinated pollen grains in complex images. High temperature significantly reduced mean pollen germination from an average of 40% under optimal conditions to an average of 21% under heat stress (P < 0.05), with a significant genotype × growth temperature interaction. Invitro incubation temperatures ranging from 10 °C to 45 °C produced a clear thermal response; however, no significant genotype × incubation temperature interaction was detected within either growth temperature regime. Although photosynthetic and physiological traits were measured exploring their relationship with pollen germination, their transient and complex response limited their reliability for predicting reproductive performance. The automated pipeline substantially reduced the time required to evaluate pollen germination. The pipeline processed nearly 5,000 images in approximately one hour, substantially increasing throughput and reducing reliance on manual counting. The findings demonstrate that pollen germination is a promising proxy trait for screening reproductive heat tolerance in soybean. Combining controlled environment phenotyping with YOLO-based object detection enabled efficient, accurate, and scalable pollen analysis, and represents the central methodological advance of this study. YOLOv9 performed best among the tested architectures, although discrepancies from manual counts in some images indicate that additional validation is needed. The weak associations with vegetative physiological traits further support the value of direct pollen-based phenotyping.
Why it matches plant phenotyping methods深層学習による花粉画像解析を中心に、花粉発芽という生殖形質を高速・自動測定するハイスループット表現型解析フレームワークを開発・比較・検証している。
abstractThis study presents a high-throughput phenotyping framework that integrates controlled-environment pollen germination assays with deep learning–based object detection for rapid, accurate, and scalable evaluation of reproductive heat tolerance in soybean breeding programs.
Reproduction assets foundThe authors state that all data supporting the study, including annotated pollen germination images, computational and statistical codes, and analysis tools, were deposited in Zenodo with a public DOI. This is a paper-specific, publicly actionable asset. LabelMe and Ultralytics YOLO are generic third-party tools, not作者Dataset · publicCommission.
Data availability
All data supporting the findings of this study, including annotated images, computational and
statistical codes, and analysis tools, have been deposited in the Zenodo data repository. Additional
data will be made available upon reasonable request following acceptance of the manuscript.
Repository: https://doi.org/10.5281/zenodo.21685593
Ethics approval and consent to participate
Not applicable
Consent for publication
Not applicable
Competing Interests
Authors declared no competing interests
References
1. FAOSTAT: Crops and livestock products: soybean production data. https://www.fao.org/faostat/
(2022). Accessed 15 Feb 2026.
2. Patel D, Franklin KA. TemperaturOpen asset ↗Zenodo · 10.5281/zenodo.21685593pdf-raw-page:28 lines:1-34Plant phenotyping relevance match · UnverifiedbioRxiv · Europe PMC · checked 5 Sept 2026
Red crown rot of soybean (RCR), caused by Calonectria ilicicola, is an emerging soilborne disease whose quantification is challenging due to its complex symptom development across root and foliage levels. This study developed and evaluated a multi-scale framework to improve the assessment of RCR severity from controlled environments to field conditions using root imaging and standardized visual scales. Under controlled conditions, a standard area diagram (SAD) for root necrosis was developed and validated, and SAD-assisted evaluations significantly improved accuracy, precision, and inter-rater agreement compared with unaided assessments. In field conditions, a diagrammatic symptom scale (DSS) was developed using consensus-rated images from experts and showed high reliability, repeatability, and reproducibility across 18 raters, with strong intra- and inter-rater agreement. This study developed and evaluated complementary methods to improve the assessment of RCR severity from controlled environments to field conditions using root imaging and standardized visual scales.
Why it matches plant phenotyping methods根の壊死と地上部症状という植物病害状態の定量評価法を開発・検証しており、画像化と標準視覚尺度が研究の中心であるため。
abstractThis study developed and evaluated a multi-scale framework to improve the assessment of RCR severity from controlled environments to field conditions using root imaging and standardized visual scales.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 15 Sept 2026
Abstract Probe-based spatial transcriptomics platforms use predefined oligonucleotide panels to detect selected RNAs in tissue sections while preserving transcript spatial coordinates. Accurate cell segmentation is required for reliable transcript-to-cell assignments. This analytical process is affected in plant tissues by cell walls, large vacuoles, and strong autofluorescence, which often reduce boundary contrast and elevate background. Nucleus-only segmentation with fixed-distance expansion can be an alternative approach, but it underestimates cellular area and morphology and reduces the number of assignable transcripts per cell. Here, we present a practical workflow for segmentation and downstream processing in plant probe-based spatial transcriptomics. Using the soybean nodule, soybean seed, rice root, and wheat inflorescence, we demonstrate the applicability of our workflow across species, tissues, and technological platforms. In brief, candidate cell masks are generated from available fluorescence signals and then selected and corrected using two napari plugins. Transcript-informed refinement with Baysor is included as an optional step. Upon benchmarking our approach using a collection of metrics (assignment yield, background/negative controls, and per-cell transcript/gene distributions) and linked segmentation choices to expression-matrix quality and downstream clustering, we demonstrate the potential of our workflow to support the analysis of plant probe-based spatial transcriptomics.
Why it matches plant phenotyping methods植物組織の細胞セグメンテーションとトランスクリプト割当てを改善する実用ワークフローを開発し、複数種・組織でベンチマークしている。植物形態そのものの測定ではないが、細胞レベルの空間状態を抽出する解析手法が中心である。
abstractHere, we present a practical workflow for segmentation and downstream processing in plant probe-based spatial transcriptomics.
This study investigates the automatic classification of crop phenological stages from low-altitude UAV RGB imagery. The dataset included 11,489 images of five crops: sunflower, rapeseed, soybean, wheat, and barley. The images were annotated using the Biologische Bundesanstalt, Bundessortenamt und Chemische Industrie (BBCH) scale, with labels corresponding to either single stages or stage ranges to reflect heterogeneous field conditions and transitional crop states. A pretrained ResNet18 model was adapted to the task using transfer learning. Training was conducted in two stages: first, the classification head was optimized while the backbone remained frozen; second, the entire network was fine-tuned. The model achieved strong internal test accuracy across all crops, with 100% test accuracy for rapeseed and barley, more than 99% for the remaining crops, and a mean accuracy of 99.73% under the studied survey conditions. The results also compare favorably with previously reported studies on UAV-based phenological classification. Overall, the findings support the potential of low-altitude UAV imagery and deep learning for localized phenological assessment of selected field zones in precision agriculture, while broader deployment requires validation across independent fields, seasons, regions, and survey conditions.
Why it matches plant phenotyping methodsUAV画像と深層学習により作物の生育（フェノロジー）段階を自動推定する手法が研究の中心であり、植物状態の取得・分類に直接関わる。
abstractThis study investigates the automatic classification of crop phenological stages from low-altitude UAV RGB imagery.
White lupin ( Lupinus albus L.) is a cool-season grain legume with seed crude protein of 33-47%, competitive with soybean ( Glycine max L.) meal. It also fixes nitrogen and mobilizes soil phosphorus. Because soybean is a summer crop, white lupin can occupy Southeastern winter fields as a complementary protein source. Breeding for seed protein is limited by the cost and throughput of reference phenotyping. To determine how each is best deployed, we compared the utility of near-infrared spectroscopy (NIRS)-based phenomic selection with genomic selection based on 246,847 SNPs from low-pass, whole genome sequencing in a panel of Auburn University breeding lines and USDA National Plant Germplasm System germplasm. A handheld NIR calibration against Dumas reference protein reached screening-grade accuracy (R 2 = 0.81). Under common cross-validation, phenomic predictive ability was 0.93 and genomic was 0.12. The low genomic value was consistent with moderate heritability (H 2 = 0.33) and strong genotype-by-year interaction. Beyond predictive ability, NIRS recovered superior accessions the strictest selection intensity, and 40 to 60 reference assays sufficed to calibrate the model. Handheld NIRS is a low-cost tool for protein calibration and early-generation screening, while genomic prediction remains suited to parental selection, together supporting a complementary strategy for legume breeding Plain Language Summary Soybean meal is the main protein source for livestock and fish farms in the United States. Because soybean is a summer crop, many Southeastern fields sit idle or grow low-value cover crops in winter. White lupin, a cool-season legume whose seeds are as protein-rich as soybean meal, makes a good complementary winter crop: it yields high-protein grain while serving as a cover crop that fixes nitrogen and frees up soil phosphorus for later crops. In our early-stage lupin breeding program, measuring seed protein by standard lab methods is slow and costly. We built a calibration that lets a handheld scanner estimate protein from light, and compared it with predicting protein from the plant’s DNA. The scanner gave accurate, low-cost protein screening from only about 40-60 lab tests, while DNA-based prediction remains suited to guiding parent selection. Used together, these tools offer breeders a practical path to develop high-protein white lupin. Core ideas Handheld NIRS provides screening-grade prediction of white lupin seed crude protein. Spectra carried more usable protein signal than markers by measuring seed chemistry directly. NIRS and genomic prediction serve different stages of a white lupin breeding program. About 40 to 60 reference assays sufficed to calibrate NIRS to near-full accuracy.
Why it matches plant phenotyping methods携帯型NIRSによる種子タンパク質形質の推定・校正・精度検証が研究の中心であり、育種スクリーニングへの実質的応用も評価している。
abstractA handheld NIR calibration against Dumas reference protein reached screening-grade accuracy (R 2 = 0.81).
This study investigates the automatic classification of crop phenological stages from low-altitude UAV RGB imagery. The dataset included 11,489 images of five crops: sunflower, rapeseed, soybean, wheat, and barley. The images were annotated using the Biologische Bundesanstalt, Bundessortenamt und Chemische Industrie (BBCH) scale, with labels corresponding to either single stages or stage ranges to reflect heterogeneous field conditions and transitional crop states. A pretrained ResNet18 model was adapted to the task using transfer learning. Training was conducted in two stages: first, the classification head was optimized while the backbone remained frozen; second, the entire network was fine-tuned. The model achieved strong internal test accuracy across all crops, with 100% test accuracy for rapeseed and barley, more than 99% for the remaining crops, and a mean accuracy of 99.73% under the studied survey conditions. The results also compare favorably with previously reported studies on UAV-based phenological classification. Overall, the findings support the potential of low-altitude UAV imagery and deep learning for localized phenological assessment of selected field zones in precision agriculture, while broader deployment requires validation across independent fields, seasons, regions, and survey conditions.
Why it matches plant phenotyping methodsUAV画像と深層学習により作物の生育・フェノロジー段階を自動推定する方法が研究の中心であり、植物状態の抽出性能も評価している。
abstractThis study investigates the automatic classification of crop phenological stages from low-altitude UAV RGB imagery.
Unmanned aerial vehicle (UAV) imagery can support plot-scale crop phenotyping, but spectral, RGB and structural predictors may contribute differently to different traits. We compared six predefined feature groups for predicting soybean SPAD and plant height (PH) in a 1.3 ha field experiment in Sanya, China. The field contained 6197 soybean planting plots, of which 234 had paired SPAD and PH measurements. Multispectral bands, vegetation indices (VIs), RGB descriptors and digital surface model (DSM) metrics were extracted from DJI Mavic 3 Multispectral imagery. Six regression algorithms were evaluated using random fivefold cross-validation, spatial block cross-validation and nested spatial cross-validation. Under random cross-validation, ExtraTrees with multispectral bands, VIs and RGB descriptors produced the numerically highest SPAD performance (R2 = 0.589; RMSE = 6.66), while BayesianRidge with multispectral bands, VIs and DSM metrics produced the highest PH performance (R2 = 0.760; RMSE = 7.14 cm). Nested spatial cross-validation yielded R2 = 0.473 and RMSE = 7.56 for SPAD and R2 = 0.690 and RMSE = 8.13 cm for PH. G4 was selected in four of the five outer folds for SPAD, although the selected algorithm varied, and G5 was selected in all five outer folds for PH. VIs improved prediction of both traits relative to the original bands. Adding RGB descriptors produced only a small and model-dependent improvement for SPAD, whereas adding DSM metrics produced a larger and more consistent improvement for PH. The complete feature set did not outperform G4 for SPAD or G5 for PH. The retained models were applied to all 6197 plots to map SPAD, PH and their field relative combinations. Because all of the validations used one field and one UAV acquisition date, the results describe performance within this experiment and do not establish transferability to other sites, years or growth stages.
Why it matches plant phenotyping methodsUAVマルチスペクトル・RGB・構造特徴からSPADと草丈を推定する特徴抽出および回帰手法を、複数の空間交差検証で比較・評価しており、植物表現型取得が研究の中心である。
titleTrait-Specific Contributions of UAV Multispectral, RGB and Structural Features to Soybean SPAD and Plant Height Phenotyping
Phosphorus (P) deficiency severely limits soybean ( Glycine max L.) productivity. This study proposed a three-stage screening framework to identify reliable traits and P-efficient genotypes. In Experiment I, percent tolerance to phosphorus deficiency (PTPD) was calculated for ten growth parameters across 98 genotypes under P-deficient and control conditions. Principal component analysis and comprehensive evaluation identified six key indicators in Experiment I, which were subsequently refined to five indicators through further analysis: SPAD at V3 and R1, photosynthetic rate at R1, shoot dry weight at R8, and seed number per plant at R8. Experiment II re-evaluated these traits using 12 contrasting genotypes under three P levels, identifying CN 15 as the most P-efficient and SN 22 as the most P-inefficient. Experiment III further revealed that CN 15 maintained superior PSII performance and exhibited a 26.2% increase in grain P-utilization efficiency under 0 µM KH 2 PO 4 treatment. This integrated framework offers a preliminary reference for screening P-efficient soybean genotypes under controlled conditions, pending field evaluation.
Why it matches plant phenotyping methodsリン欠乏耐性を評価するPTPDと三段階の形質選抜フレームワーク自体を提案・検証しており、単なる生物学的処理試験ではなく、植物形質に基づく遺伝子型スクリーニング手法が中心である。
abstractThis study proposed a three-stage screening framework to identify reliable traits and P-efficient genotypes.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 5 Sept 2026
The use of glufosinate-resistant GM soybean has expanded, raising concerns about resistant weed development and unintended transgene flow. To support monitoring for timely management, we propose an early, non-destructive identification method using spectral images acquired from whole soybean plants after glufosinate treatment. We evaluated the potential of spectral imaging, using RGB, infrared (IR) thermal, and chlorophyll fluorescence (CF) sensors, for early detection of glufosinate resistance in soybean. In the dose-response test, the key spectral indices including NDI, temperature difference, F v /F m , and NPQ distinguished between resistant and susceptible soybeans within 4 to 24 hours after treatment (HAT). IR thermal and CF imaging showed higher sensitivity in identifying resistance than RGB imaging by detecting spectral responses associated with physiological changes before visual symptoms appeared. Validation test with a single dose treatment of glufosinate reconfirmed that image analysis by both the naked eye and machine learning (ML) can discriminate between resistant and susceptible soybeans in a single day after glufosinate treatment. ML-based classification using IR thermal index achieved 100% accuracy as early as 6 HAT and the classification by the naked eye using IR thermal images showed 96.6% accuracy at 24 HAT. These results suggest that plant imaging enables early and non-destructive identification of herbicide-resistant individuals by detecting early spectral changes to herbicide treatment. These findings support its use as a potential alternative to conventional diagnostic methods for detecting individuals containing transgenes in herbicide-resistant GM soybean cultivation for future applications in herbicide-resistant weed monitoring.
Why it matches plant phenotyping methodsスペクトル画像（RGB、熱赤外、クロロフィル蛍光）と機械学習を用いて、薬剤処理後の植物の生理応答から耐性を早期識別する方法を開発・検証しており、植物表現型の取得が中心である。
abstractwe propose an early, non-destructive identification method using spectral images acquired from whole soybean plants after glufosinate treatment.
Accurate 3D crop monitoring underpins data-driven precision agriculture by enabling field-scale analysis of plant structure, growth dynamics, and management response. Modern 3D reconstruction methods perform strongly on generic benchmarks, but rendered appearance may not translate into metrically and agronomically useful geometry in crop fields. We introduce UAV3DCrop, a public benchmark of repeated multi-angle unmanned aerial vehicle (UAV) crop surveys. It contains 88,830 RGB images at $5280 \times 3956$ pixels, with a ground sampling distance of 3.6-5.8 mm, from 91 scenes spanning corn, soybean, wheat, and oat. Track A evaluates seven scene-optimized methods -- Neural Radiance Field (NeRF) and 3D Gaussian Splatting (3DGS) variants -- on held-out views, photogrammetry-referenced depth, and canopy-height recovery. Track B tests four pretrained feed-forward models on zero-shot camera-pose and geometry estimation. The scene-optimized methods rank differently across the three targets: Splatfacto-big leads appearance, whereas Scaffold-GS leads depth and is statistically tied with Splatfacto for canopy height. Among feed-forward models, MapAnything leads on seven of the eight metrics, while the remaining models vary more across crops and fail severely on absolute scale in a way that alignment conceals. Repeated acquisitions reveal further sensitivities that differ by output type and by model, associated with position within the acquisition sequence and with tie-point multiplicity. Current 3D reconstruction methods are therefore not yet interchangeable for agronomic use: no single method wins on appearance, geometry, and canopy height at once, and only one of four feed-forward models recovers usable metric scale. The dataset is publicly available at https://link-dev.github.io/UAV3DCrop/
Why it matches plant phenotyping methods植物キャノピー高さという明示的な形質を対象に、UAV 3D再構成手法をベンチマークし、公開データセットとして提供しているため、フェノタイピング手法が中心である。
abstractWe introduce UAV3DCrop, a public benchmark of repeated multi-angle unmanned aerial vehicle (UAV) crop surveys.
Reproduction assets foundThe paper introduces UAV3DCrop, a public benchmark of repeated multi-angle UAV crop surveys (88,830 RGB images, 91 scenes, four crops) with refined poses, photogrammetric depth references, and linked canopy-height and effective-LAI field measurements. The dataset is explicitly stated to be publicly available under CC BDataset · publiche acquisition sequence and with tie-point multiplicity. Current 3D reconstruction methods are therefore not yet interchangeable for agronomic use: no single method wins on appearance, geometry, and canopy height at once, and only one of four feed-forward models recovers usable metric scale. The dataset is publicly available at https://link-dev.github.io/UAV3DCrop/ .
Keywords:
UAV imagery; agricultural datasets; crop-field reconstruction; neural radiance fields;
Gaussian splatting; feed-forward geometry.
1 IntroductionOpen asset ↗UAV3DCroplines:1-90Plant phenotyping relevance match · UnverifiedOpenAlex · checked 6 Sept 2026
Soybean diseases caused by fungal, bacterial, and viral pathogens represent a major constraint to global agricultural productivity. Although molecular phylogenetic analyses have advanced the understanding of pathogen evolution, the extent to which disease phenotypes reflect evolutionary relationships remains poorly understood. In this study, we developed an integrative framework combining deep learning-based phenotypic analysis with phylogenetic inference to investigate the relationship between soybean disease symptoms and pathogen evolution. An EfficientNet-B0 convolutional neural network (CNN) was trained to classify 10 soybean disease classes comprising 703 leaf images and achieved a mean cross-validation accuracy of 98.72 ± 1.17%, a weighted F1-score of 98.74 ± 1.16%, and a macro F1-score of 98.47 ± 1.74%. Evaluation on a held-out test set generated through image-level partitioning yielded an accuracy of 96.19%, a weighted F1-score of 96.28%, and a macro F1-score of 95.86%. Latent feature embeddings revealed a structured phenotypic space with clear separation among most disease classes and enabled quantitative analyses of phenotypic similarity. To provide biological context, taxonomy-derived distance matrices and sequence-based phylogenetic analyses of the fungal subset using 28S rRNA sequences were compared with CNN-derived phenotypic representations. A Mantel test identified a moderate and statistically significant association between phenotypic and phylogenetic distances (Spearman r = 0.3393, p = 0.0050), indicating that pathogen evolutionary history contributes to disease phenotype while explaining only part of the observed phenotypic variation. Overall, the results demonstrate that deep learning effectively captures biologically meaningful phenotypic information while highlighting that disease symptoms arise from the combined influence of pathogen evolution, host responses, and environmental conditions. This study provides an integrative framework for combining image-based phenotyping with phylogenetic analysis to support biologically informed interpretation of plant disease phenotypes.
Why it matches plant phenotyping methods深層学習による画像ベースのダイズ病徴分類・表現型空間抽出が研究の中心であり、植物病害状態を直接推定する手法を開発・評価している。
abstractwe developed an integrative framework combining deep learning-based phenotypic analysis with phylogenetic inference
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.
The precise identification of unsound soybean seeds is a critical step in deep soybean processing and seed selection. The accuracy of this identification directly influences the quality of subsequent processed products, as well as the germination rate and yield of soybean crops. This study proposes a nondestructive identification method for unsound soybean seeds based on hyperspectral imaging (HSI), Gramian Angular Field (GAF), and a Dual-Channel Residual-Squeeze-and-Excitation Network with GAF Fusion (DC-RSEN-GF). According to common damage types, soybeans were categorized into six classes: sound seeds, thermal-damaged seeds, insect-damaged seeds, broken seeds, spotted seeds, and moldy seeds. Spectral data from these six soybean categories were acquired using a hyperspectral camera and transformed into two-dimensional GAF images. The DC-RSEN-GF network integrates one-dimensional spectral data with two-dimensional GAF images. After preprocessing with Savitzky-Golay (SG) smoothing, high-precision classification was achieved through residual blocks, an attention mechanism (using SENet), and feature fusion. Compared to five benchmark models-Extremely Randomized Trees (ERT), Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), VGG19, and ResNet18-the DC-RSEN-GF model achieved superior performance, with accuracy, precision, specificity, and F1-scores of 96.36%, 96.43%, 97.92%, and 96.36%, respectively. The accuracy, precision, and F1-scores are all superior to traditional machine learning and existing deep learning models, demonstrating better classification capabilities. In addition, t-distributed Stochastic Neighbor Embedding (t-SNE) was employed for visual analysis of soybean spectra, further validating the reliability of the DC-RSEN-GF model. The proposed detection method, based on HSI and DC-RSEN-GF, enables accurate and nondestructive identification of unsound soybean seeds and holds significant potential for practical application.
Why it matches plant phenotyping methodsハイパースペクトル画像と深層学習を用いて、種子の損傷・病変状態を非破壊的に分類する取得・解析手法が研究の中心であり、植物状態の表現型測定に該当する。
abstractThis study proposes a nondestructive identification method for unsound soybean seeds based on hyperspectral imaging (HSI), Gramian Angular Field (GAF), and a Dual-Channel Residual-Squeeze-and-Excitation Network with GAF Fusion (DC-RSEN-GF).
In-season fine-scale (i.e., within-field experiment plot scale) crop grain yield (GY) prediction is critical for optimizing inputs, minimizing environmental impacts, and supporting sustainable food production. Traditional approaches, such as field surveys, are often costly and inefficient over large areas. As an alternative, remote sensing combined with crop simulation models (CSMs) has been increasingly applied for in-season GY prediction. This study investigates the potential of integrating Uncrewed Aircraft Systems (UAS)-based remote sensing data, deep learning, and CSMs to predict maize and soybean GY using a data assimilation approach. UAS multispectral imagery was collected, along with field-measured maize above-ground biomass (AGB) and soybean leaf area index (LAI) during the 2022 and 2023 growing seasons at experimental fields in Brookings, South Dakota. Maize AGB was measured at two growth stages, while soybean LAI was collected across four stages. One-dimensional convolutional neural networks (1D-CNNs) were used to estimate maize AGB and soybean LAI from canopy spectral, textural, and structural features derived from UAS imagery. These UAS and deep learning–derived crop traits were assimilated into DSSAT-Maize and DSSAT-Soybean models to optimize parameters, and the optimized models were subsequently used to predict GY. For maize, the DSSAT-Maize model achieved an R² of 0.62, an RMSE of 717.8 kg ha⁻¹, and an rRMSE of 6.7% for GY prediction. For soybean, the DSSAT-Soybean model achieved an R² of 0.81, an RMSE of 207.3 kg ha⁻¹, and an rRMSE of 4.9%. Overall, these results highlight the potential of combining high-resolution UAS data and deep learning–derived crop traits within a CSM framework through data assimilation, enabling fine-scale, in-season yield predictions and supporting precise agricultural management.
Why it matches plant phenotyping methodsUAS画像と深層学習により、作物のAGBおよびLAIという植物形質を推定する取得・解析手法が研究の中心であり、作物モデルへの同化と性能評価も行っている。
abstractOne-dimensional convolutional neural networks (1D-CNNs) were used to estimate maize AGB and soybean LAI from canopy spectral, textural, and structural features derived from UAS imagery.
Soybean pod-related traits are important for seed development, yield formation, and cultivar evaluation. However, conventional measurements are inefficient and have limited ability to characterize complex pod features such as curvature, local enlargement, and continuous color variation. In this study, mature pod images of 187 cultivated soybean accessions collected across two successive years were analyzed using a deep learning-assisted image phenotyping approach. Eight pod-related traits related to size, morphology, and color were extracted from pod images. A genome-wide association study (GWAS) was performed using 61,541 high-quality SNP markers to dissect the genetic basis of these image-derived pod traits. A total of 16 stable loci associated with pod size, morphology, and color traits were identified across 11 chromosomes. Among these loci, eight were not reported in the previous image-based soybean pod GWAS study. Based on SoyBase gene annotation and Gene Ontology biological process information, 32 biologically relevant candidate gene records were prioritized within the corresponding candidate genomic intervals, while pod-related expression profiles and SoyBase association information were used as supporting evidence for candidate gene evaluation. These findings indicate that refined image-derived traits can provide complementary genetic information beyond conventional pod measurements and offer additional opportunities for dissecting soybean pod development, morphology, and mature pod color variation.
Why it matches plant phenotyping methods深層学習画像解析によるダイズ莢形質の抽出が研究の中心であり、8種類の形態・色・サイズ形質を画像から定量化している。
abstractmature pod images of 187 cultivated soybean accessions collected across two successive years were analyzed using a deep learning-assisted image phenotyping approach.
Enhancing photosynthesis is an important approach to improve crop yields. Photosynthesis, as a key factor determining crop yield, is an important approach to increasing crop production and addressing global food security issues. Improving its efficiency is crucial in this regard. However, traditional photosynthetic phenotyping has long been a bottleneck in crop breeding due to time-consuming data collection. In this study, we simultaneously measured the spectral reflectance and the net photosynthetic rate (Pn) of soybean leaves to develop a high-precision model for estimating Pn based on hyperspectral data. By applying this model, we evaluated Pn in 219 soybean materials. A multi-environment genome-wide association study (GWAS) based on multi-environmental prediction Pn was carried out using the 3VmrMLM method, and 24 significant quantitative trait loci (QTLs) and four suggestive QTLs were identified. Among them, 24 QTLs overlapped with multiple previously reported QTL related to photosynthesis, chlorophyll content, quality, etc., or with genes related to key agronomic traits such as yield. Additionally, four new QTLs were discovered, and four candidate genes potentially associated with Pn were identified. Further, haplotype analysis identified their optimal haplotypes. This study presents a robust and nondestructive hyperspectral model for estimating the photosynthetic rate in soybeans, which is successfully applied to genetic analysis, yielding stable and biologically meaningful results. The approach offers an effective means to explore the genetic basis of photosynthesis and provides a solid theoretical foundation for large-scale, monitoring of soybean photosynthetic physiology.
Why it matches plant phenotyping methods大豆葉のハイパースペクトルデータから光合成速度を推定するモデルを開発し、検証・大規模適用しており、植物フェノタイピング手法が研究の中心である。
abstractwe simultaneously measured the spectral reflectance and the net photosynthetic rate (Pn) of soybean leaves to develop a high-precision model for estimating Pn based on hyperspectral data.
SoybeanNeRF / 3D Gaussian SplattingRGB / grayscaleFruitSegmentation
Neural Radiance Fields (NeRF) have been widely adopted for reconstructing high-quality 3D scenes from 2D RGB images. However, achieving accurate 3D object segmentation within these reconstructed scenes remains challenging. Existing NeRF-based segmentation methods either rely on post-processing (SA3D), which produces noisy point clouds due to the absence of density field optimization, or employ joint training with additional segmentation heads (FruitNeRF), which can lead to suboptimal performance due to conflicting learning objectives. In this work, we propose InvNeRF-Seg (Input-substitution NeRF for Segmentation), a two-stage fine-tuning strategy for 3D object segmentation that preserves the original NeRF architecture and loss function entirely. We first train a standard NeRF on RGB images and then fine-tune it using 2D segmentation masks formatted as RGB-like inputs, without introducing any architectural modifications or additional loss functions. This input-substitution approach reshapes the density field to align with object regions while suppressing background density. We validate InvNeRF-Seg through comprehensive ablation studies examining the roles of density and color MLPs, loss function choices, and training strategies. Field density analysis reveals consistent semantic refinement: densities of object regions increase while background densities are suppressed. Experiments on synthetic fruit datasets and real-world soybean imagery demonstrate that InvNeRF-Seg produces cleaner 3D segmented point clouds compared to both SA3D and FruitNeRF, enabling more accurate downstream object counting. The method is further validated on a self-collected soybean dataset to demonstrate its applicability in real-world agricultural scenarios. Our code is available at https://github.com/ZJiangsan/InvNeRF-Seg .
Why it matches plant phenotyping methods植物画像から3D物体領域を抽出するNeRFベース手法の開発・比較検証が中心で、果実・ダイズ画像を対象に物体カウントへ応用しているため、植物器官の形態・数量推定に関わるフェノタイピング手法として含める。
abstractIn this work, we propose InvNeRF-Seg (Input-substitution NeRF for Segmentation), a two-stage fine-tuning strategy for 3D object segmentation
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 5 Sept 2026
Introduction Climate extremes increasingly threaten agricultural production, yet many artificial intelligence systems in agriculture remain local, reactive and narrowly trained for one crop, region or sensing modality. Methods We present AgriFM, a multimodal geospatial foundation model that combines satellite image time series, radar, thermal observations, weather trajectories, soil properties, topography and sparse management variables to estimate crop-stress and yield-failure risk across crops and regions. AgriFM was pretrained using self-supervised objectives on 2.4 million field-season sequences and evaluated on a curated benchmark spanning maize, wheat, soybean, rice and sorghum across five agroclimatic regions. Results In held-out geography and time-split evaluations, AgriFM improved early stress detection and yield-failure prediction over statistical, crop-model and deep-learning baselines. The largest gains occurred during compound drought and heat events, for which AgriFM produced alerts 18 to 24 days earlier than the satellite-only baseline while maintaining improved calibration. Phenology-conditioned fusion improved transfer across planting calendars, and uncertainty calibration reduced false alerts at fixed recall. Discussion Because the study is based on retrospective datasets, these findings establish cross-region retrospective performance rather than prospective field efficacy. The results support further field-based evaluation of multimodal foundation models for climate-resilient crop monitoring.
Why it matches plant phenotyping methods作物ストレス状態と収量失敗リスクを衛星・レーダー・熱画像等から推定する基盤モデルを開発し、複数作物・地域のベンチマークで評価しており、植物状態の取得・推定手法が中心である。
abstractWe present AgriFM, a multimodal geospatial foundation model that combines satellite image time series, radar, thermal observations, weather trajectories, soil properties, topography and sparse management variables to estimate crop-stress and yield-failure risk across crops and regions.
Accurate color representation is critical for UAV-based crop phenotyping, yet UAV images are often distorted by variable illumination and camera exposure settings. Here, we propose Lite U-net FiLM (LUF-net), a lightweight U-net framework integrated with feature-wise linear modulation (FiLM) layers, which incorporates both multispectral-derived irradiance and camera exposure parameters as auxiliary inputs. This metadata-aware design enables dynamic modulation of intermediate image features, allowing the network to disentangle unified canopy color from environmental artifacts. Experiments conducted across diverse crop types (soybean, rice, maize), flight altitudes (6m, 12m, 25m, 40m), and illumination conditions (overcast skies, cloudy, sunny, early morning) demonstrate that the LUF-net model substantially reduces the mean absolute percentage error to below 5.5%, outperforming conventional Gray-world, U-net, AlexNet-MLP, and SIDBlock-MLP methods. Ablation experiments further show that both irradiance and exposure metadata provide complementary information, while FiLM-based conditioning effectively integrates these acquisition parameters into feature learning, jointly contributing to improved color reconstruction performance. Moreover, correlation analysis indicates that the model's color reconstruction errors are weakly dependent on irradiance and exposure settings, confirming that LUF-net reduces sensitivity to external imaging conditions while maintaining physically meaningful and robust corrections.
Why it matches plant phenotyping methodsUAV画像の色校正手法を開発し、複数作物・撮影条件で性能検証しており、作物フェノタイピングの画像取得・補正が中心である。
abstractAccurate color representation is critical for UAV-based crop phenotyping, yet UAV images are often distorted by variable illumination and camera exposure settings.
Reproduction assets foundThe paper's data and code availability statement explicitly points to a public GitHub repository containing training/evaluation/inference code, pretrained model weights, and example data for the LUF-net color calibration method.Code · publicThe data and source code for model training, evaluation, and inference, together with pretrained model weights, example data, and detailed usage instructions, is publicly available at: https://github.com/wangchufeng3652/color-correction.Open asset ↗wangchufeng3652/color-correctionhtml-lines:278-299Plant phenotyping relevance match · UnverifiedEurope PMC · Crossref · checked 15 Sept 2026
Agrophotovoltaic (APV) systems provide a unique opportunity for improving agricultural land-use efficiency by combining crop production with solar energy capture via photovoltaic panels. In-depth information on plant growth patterns within the spatially heterogenous microclimate created by APVs would enable better planning and management within such unconventional systems. Thus, the present study demonstrates the implementation of a customized robot-mounted 3D-multispectral imaging system for monitoring the growth and spectral reflectance patterns of a conventional soybean cultivar "Eiko" (EK) and a chlorophyll-deficient mutant variety MinnGold (MG) under an APV system. Weekly trends in canopy morphometric features revealed significant variations in canopy height, surface area, light penetration, and volume across the APV field depending on the proximity with the overhead solar panels for both EK and MG, with plants receiving adequate rainfall and intermittent shade performing the best. Furthermore, although spectral indices exhibited variations between EK and MG due to intrinsic differences in pigmentation, symptoms of stress could be detected for both genotypes within rain-shaded areas of the APV plot. Hence, the present investigation depicts the potential for complementary usage of robotics and machine vision for high-precision high-throughput crop monitoring under APVs, which would help improve crop management within such non-homogenous cultivation systems.
Why it matches plant phenotyping methodsカスタマイズしたロボット搭載3Dマルチスペクトル画像システムを実装し、植物形態・スペクトル・ストレス状態を高精度に取得することが研究の中心である。
abstractthe present study demonstrates the implementation of a customized robot-mounted 3D-multispectral imaging system for monitoring the growth and spectral reflectance patterns
With climate change and global population growth, accelerating the breeding of superior crop varieties is essential for food security. Genomic prediction, which uses genome-wide genetic markers to predict crop traits, plays an important role in intelligent crop breeding. However, existing methods often lack stable and accurate performance across crops and traits. Here, we propose GEG2P, a genetic algorithm-based ensemble learning method for genotype-to-phenotype prediction, integrates 20 base learners, dynamically selects their combinations through an iterative optimization strategy, and optimizes their weights using the genetic algorithm. Compared with the best-performing single base learners, GEG2P improves prediction accuracy by 4.02% on average across maize, wheat, rice, chickpea, and soybean. We use SHAP to quantify the contribution of SNPs to phenotype prediction and find that SNPs with large effects captured by different base learners are functionally complementary. This study provides a robust and accurate genomic prediction method for crop breeding.
Why it matches plant phenotyping methods作物形質の遺伝子型から表現型を予測するアンサンブル計算法を開発し、複数作物で精度比較・検証しており、表現型推定手法が研究の中心である。
abstractHere, we propose GEG2P, a genetic algorithm-based ensemble learning method for genotype-to-phenotype prediction, integrates 20 base learners, dynamically selects their combinations through an iterative optimization strategy, and optimizes their weights using the genetic algorithm.
Reproduction assets foundThe paper provides public author code (GitHub GEG2P repository and Docker Hub image), a Zenodo deposit of significant SNP interaction pairs generated in this study, and a Figshare link with the wheat genotypic and phenotypic data used in the analyses. These are paper-specific, publicly available, and actionable.Code · publicScripts used in this study are available at GitHub [ https://github.com/Deep-Breeding/GEG2P ] 89 .Open asset ↗GitHub · Deep-Breeding/GEG2Plines:236-266Dataset · publicThe genotypic and phenotypic data of wheat are available at Figshare [ https://figshare.com/s/287c2c7f1623008487a5 ] 68 .Open asset ↗Figsharelines:236-266Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
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 · UnverifiedOpenAlex · checked 15 Sept 2026
Accurate yield prediction for major grain and oilseed crops, including soybean, corn, wheat, and rice, is essential for food-security assessment and precision field management. This study presents a structured integrative review of UAV-based crop yield prediction and follows PRISMA-guided procedures for literature search, screening, and evidence synthesis. Seventy peer-reviewed studies published between 2018 and 2025 were synthesized within a "Data-Ground Truth-Model-Decision" framework. Beyond summarizing UAV platforms, sensor configurations, feature-engineering strategies, and model architectures, the review explicitly distinguishes among microplot, field, and regional prediction scales, and evaluates the characteristics and limitations of yield-label acquisition methods, including manual harvest, plot-combine harvest, and combine yield-monitor data. Existing evidence indicates that the reliability of UAV-based yield prediction depends not only on optimal image acquisition windows, multi-source feature fusion, and model architecture, but also on scale-consistent yield labels, spatially aware validation strategies, and clearly defined model outputs, such as plot-level scalar yield, field-scale yield maps, and regional yield estimates. Major bottlenecks include scale mismatch between UAV imagery and yield labels, error propagation during yield-map generation, limited cross-year and cross-region transferability, weak causal interpretability, and difficulties in deploying models under complex operational field conditions. Future research should emphasize scale-explicit benchmark datasets, quality-controlled ground-truth yield acquisition, UAV-satellite-ground data fusion, spatiotemporal deep learning, and edge-cloud collaborative systems that can translate prediction outputs into agronomic decisions. This review provides a practical pathway for developing robust, interpretable, and deployable UAV-based yield prediction systems for major grain and oilseed crops.
Why it matches plant phenotyping methodsUAV画像から作物の収量という植物形質を推定する手法を中心に、プラットフォーム、特徴量、モデル、検証尺度、グラウンドトゥルースを体系的にレビューしているため、方法レビューとして収載。
abstractThis study presents a structured integrative review of UAV-based crop yield prediction and follows PRISMA-guided procedures for literature search, screening, and evidence synthesis.
The selection of genotypes adapted to water stress requires experimental facilities that allow environmental control without compromising physiological and yield relevance. The objective of this study was to design and validate an outdoor phenotyping semi-controlled platform, PlaFe, which comprised sixty-two high-volume prismatic lysimeters arranged in rows 1.2 m long and spaced 0.6 m apart. Soil water dynamics were monitored weekly using a weighting system. To validate PlaFe, two soybean genotypes were exposed to two water scenarios for forty days from R2 + 7d, during two growing seasons. Two irrigation treatments were applied: irrigation to keep soil water content over 60–70 % of field capacity (EH0), and irrigation equivalent to 35 % of that applied in EH0 (EH1). Water consumption, crop biomass, and pod number were determined at maturity. On average, water stress reduced both biomass and pod numbers by 40 %. However, reproductive efficiency varied among genotypes. Canopy temperature increased by 0.56 °C as daily water consumption decreased, demonstrating its potential to assess drought. These results demonstrate PlaFe’s potential for the accurate evaluation of crop response and adaptation to diverse water scenarios without compromising the complex plant-environment interactions inherent to field conditions.
Why it matches plant phenotyping methodsPlaFeという屋外半制御型フェノタイピングプラットフォームを設計・検証しており、植物の水消費、バイオマス、莢数、群落温度などの表現型評価が研究の中心である。
abstractThe objective of this study was to design and validate an outdoor phenotyping semi-controlled platform, PlaFe
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · Crossref · checked 5 Sept 2026
Abstract Precision field management and high-throughput plant phenotyping increasingly rely on remote sensing to capture spatial and temporal variability in crop performance. Unmanned aerial vehicle (UAV) – based sensing offers unique advantages for field-scale data collection, including high spatial resolution, flexible deployment, and scalable throughput. However, the full potential of UAV platforms remains constrained by labor-intensive operations across flight execution, data transfer, and processing workflows. This study presents a systematic evaluation of an automatic UAV-based crop sensing platform through a season-long, multi-crop field experiment. Data acquisition was conducted over a maize irrigation trial and a soybean breeding experiment, resulting in 176 completed flights over 28 days during the growing season. High-frequency flights on selected days captured diurnal dynamics in key canopy traits, including maize leaf rolling under drought stress and genotype-dependent plot temperature variation in soybean. In the maize irrigation experiment, significant differences in diurnal canopy cover ratio (CCR) were observed among irrigation treatments. The predictive relationship between CCR and final grain yield strengthened throughout the day, with the coefficient of determination (R 2 ) increasing from 0.05 in the early morning (RMSE = 3.05 Mg ha − 1 ) to 0.65 at midday (RMSE = 1.87 Mg ha − 1 ), highlighting the importance of temporal optimization in UAV-based sensing. Temperature measurements from the onboard thermal infrared camera showed a strong overall linear correlation with ground truth measurements (R 2 = 0.85). In the soybean trial, the highest plot temperature was observed on the fast-wilting genotype. Additionally, regression models were developed to estimate key crop traits, including canopy height (CH) and leaf area index (LAI), demonstrating the platform’s quantitative sensing capability. Overall, this study demonstrates that automatic UAV systems enable high-temporal-resolution crop monitoring while substantially reducing operational cost. The results highlight their potential for precise crop management and scalable field phenotyping. Future work will focus on integrating automated data processing pipelines to support near-real-time analytics and decision-making.
Why it matches plant phenotyping methods自動UAVのRGB・熱画像センシング platform を圃場で系統的に評価し、温度・キャノピー被覆率・高さ・LAIなどの植物形質を定量化しているため、フェノタイピング手法が研究の中心である。
abstractThis study presents a systematic evaluation of an automatic UAV-based crop sensing platform through a season-long, multi-crop field experiment.
A bstract Quantifying root traits such as root length (RL) and root surface area (RSA) from minirhizotron imagery is a valuable approach for overcoming the phenotyping bottleneck that limits understanding and improvement of crop productivity, resource use efficiency and resilience in field experiments. However, current approaches remain labor-intensive, and deep learning (DL) methods suffer from limited generalization ability. We present RootQuant, an end-to-end DL model that simultaneously predicts RL and RSA directly from minirhizotron images using only whole-image trait values as supervision, thereby eliminating the need for pixel-level annotations. The model’s generalization ability was evaluated across species and fine-tuning configurations. The practical applicability of the model was further assessed under field conditions by converting image-derived RL estimates into volumetric root length density (vRLD). Using 118,191 maize and soybean images collected between 2009 and 2020, RootQuant trained on both species achieved an R 2 of 0.90 and an RMSE of 2.9 mm for RL, and an R 2 of 0.88 and an RMSE of 4.2 mm 2 for RSA. The same mixed-species model generalized strongly across species, yielding an 8% relative improvement in R 2 and a 30% lower RMSE on maize compared with the same architecture trained on a single species and applied zero-shot. Image-derived RL predictions converted to vRLD showed the expected depth-dependent decline in vRLD, as was also found by coincident destructive quantification of roots washed out of soil cores. By providing a generalist backbone model trained on a large dataset from two major crop species, RootQuant enables high-throughput simultaneous estimation of two relevant root traits directly from raw imagery without task-specific fine-tuning, thereby accelerating in situ root system analysis and phenotyping applications.
Why it matches plant phenotyping methodsミニライゾトロン画像から根長・根表面積を推定する深層学習手法を開発し、種間一般化と圃場適用性を評価しており、植物フェノタイピング手法が研究の中心である。
abstractWe present RootQuant, an end-to-end DL model that simultaneously predicts RL and RSA directly from minirhizotron images using only whole-image trait values as supervision
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · Crossref · checked 5 Sept 2026
(L.) Merrill) is a highly important crop widely used for food, edible oil, animal feed, and microbial fermentation products. Traditional phenotypic measurement methods are often time-consuming, labor-intensive, destructive to plants, and prone to human error. High-Throughput Phenotyping (HTP) enables precise assessment of multiple soybean phenotypic features, including morphology, physiology, diseases, pests, and agronomic traits. Artificial Intelligence (AI) is a research field dedicated to developing algorithms for multiple tasks. This review highlights the application of HTP and AI in soybean breeding programs. We discuss the challenges of implementing HTP in soybean breeding and focus on the potential and limitations of Deep Learning (DL) to support soybean breeding goals. We demonstrate the application of HTP to key soybean traits, several HTP platforms, as well as DL applications across different datasets and strategies for developing large foundation models. While integrating AI into soybean breeding programs remains a challenge, leveraging HTP data and Large Language Models (LLMs) could reshape soybean breeding.
Why it matches plant phenotyping methods大豆育種におけるHTPとAIの応用、形質評価、プラットフォーム、データセットおよび深層学習を中心に扱うフェノタイピング手法レビューであり、方法論が中心的です。
abstractThis review highlights the application of HTP and AI in soybean breeding programs.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 15 Sept 2026
Introduction Accurate and rapid diagnosis of plant leaf disease symptoms is critical for sustainable agricultural crop production, yet traditional methods often lack efficiency and robustness under field conditions. Methods Here, we propose a deep learning framework based on an improved Vision Transformer architecture that integrates a dynamic sparse attention mechanism, termed KBTNet, for targeted feature extraction in symptom-affected regions of leaf images. The model incorporates a learnable nonlinear enhancement module to capture subtle visual disease symptom variations such as lesions, discoloration patterns, and spot distributions, and a lightweight Transformer design to reduce computational cost. Results Evaluated on a multisource dataset containing soybean and tomato leaf images representing diverse disease symptom patterns, our approach achieved 93.19% classification accuracy, outperforming current state-of-the-art models. Additional evaluations on public plant disease datasets from multiple crops further demonstrate the model's ability to recognize disease symptom patterns across diverse crop species. Discussion The proposed framework achieves stable performance across diverse crop and disease symptom categories, maintains high efficiency under reduced parameter complexity, and exhibits strong potential for realtime field diagnostics on edge devices. This work provides a scalable and efficient tool for plant disease symptom detection and classification and supports the integration of visionbased intelligence into crop disease monitoring and management systems.
Why it matches plant phenotyping methods植物葉画像から病徴（病斑、変色、斑点分布）を抽出・分類する深層学習手法を開発しており、植物病害状態の表現型取得が研究の中心である。
abstractwe propose a deep learning framework based on an improved Vision Transformer architecture that integrates a dynamic sparse attention mechanism, termed KBTNet, for targeted feature extraction in symptom-affected regions of leaf images.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 6 Sept 2026
Abstract Background High temperatures during the reproductive stage of soybean severely disrupt reproductive processes and reduce yield. Heat stress causes ultrastructural damage in pollen grains, leading to reduced pollen germination, pollen size and shortened pollen tube length, ultimately lowering seed set and yield. This experiment aimed to evaluate pollen germination as a reliable, scalable phenotyping tool for assessing male gametophytic tolerance to high temperature stress in soybean. Sixteen soybean breeding lines (genotypes) were grown under controlled environments at optimal (28/18°C; day/night) and high temperature (38/28°C; day/night) regimes during flowering. In vitro pollen germination was quantified using a deep learning–based object detection tool to reduce the manual labor and improve accuracy. Several advanced object detection models belonging to the YOLO (You Only Look Once) family, specifically, YOLOv7–YOLOv12, were evaluated to identify the most reliable model. Results Comparative evaluations of different object detection models indicated that YOLOv9 model achieved superior performance in evaluating pollen germination relative to other YOLO models, especially for detecting germinated and non-germinated pollen in complex images. High temperature significantly reduced mean pollen germination from an average of 40% under optimal conditions to an average of 21% under heat stress (P
Why it matches plant phenotyping methodsダイズの耐暑性評価のため、花粉発芽を対象とした画像ベースの深層学習測定法を開発・比較検証しており、フェノタイピング手法が研究の中心である。
abstractThis experiment aimed to evaluate pollen germination as a reliable, scalable phenotyping tool for assessing male gametophytic tolerance to high temperature stress in soybean.
Spatio-Temporal Fusion (STF) has been widely used across various remote sensing applications, including environmental monitoring, land cover change detection, and water resource management by integrating multi-sensor data with different spatial and temporal resolutions. The objective of this study was to generate spatially and temporally fine-resolution imagery and to evaluate the performance of multiple STF algorithms and Consistent Adjustment of the Climatology to Actual Observations (CACAO) post-processing for parcel-level crop monitoring. The study was conducted in a soybean field located in Anseong, South Korea, covering the entire soybean growing season from late June to early November. Near-daily Planet SuperDove imagery with 3 m resolution was used to temporally enhance UAV images, which were acquired at 0.05 m resolution but only at weekly to monthly intervals. Through the downscaling process, the UAV data were converted into a daily dataset with a target spatial resolution of 0.5 m. Relative radiometric normalization was applied, followed by the implementation and comparison of four STF algorithms- Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM), Fitting, spatial Filtering and residual Compensation (Fit-FC), Flexible Spatiotemporal Data Fusion (FSDAF), and Variation-based Spatiotemporal Data Fusion (VSDF)-within a 4-fold cross-validation framework. CACAO post-processing was then employed to reconstruct temporally continuous Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) trajectories, from which the NDVI-based Vegetation Growth Metrics (VGM)85 and the EVI-based VGMmax were derived. The validation results indicated that ESTARFM achieved the highest NDVI performance among the evaluated algorithms, with a Root Mean Square Error (RMSE) of 0.113 and the Universal Image Quality Index (UIQI) of 0.697. CACAO post-processing further improved these results, with CA-ESTARFM achieving an RMSE of 0.108 and a UIQI of 0.740, corresponding to a 4.4% reduction in RMSE and a 6.2% improvement in UIQI relative to the baseline ESTARFM. NDVI histogram and spatial analyses demonstrated that CA-ESTARFM achieved the most consistent agreement with UAV observations while preserving fine-scale spatial heterogeneity. In addition, intra-field vegetation assessment using NDVI-based VGM85 and EVI-based VGMmax showed that CA-ESTARFM remained consistent with simple linear interpolation of UAV observations while retaining finer spatial structure and reducing localized noise in the derived growth metrics. The proposed framework demonstrates strong potential for applications in comprehensive crop monitoring, precision agriculture management, and yield forecasting.
Why it matches plant phenotyping methodsUAV・衛星画像の時空間融合とCACAO処理により、NDVI/EVIおよび植生成長指標を抽出するワークフローを開発・比較検証しており、植物状態の取得手法が中心である。
abstractThe objective of this study was to generate spatially and temporally fine-resolution imagery and to evaluate the performance of multiple STF algorithms and Consistent Adjustment of the Climatology to Actual Observations (CACAO) post-processing for parcel-level crop monitoring.
Reproduction assets foundThe paper's STF/CACAO analysis code is openly available on Zenodo. The underlying Planet/UAV imagery data are only available from the corresponding author upon request, so they qualify as request_only.Code · publicThe code supporting this study is openly available at Zenodo
(https://doi.org/10.5281/zenodo.20923823).Open asset ↗Zenodo · 10.5281/zenodo.20923823pdf-page:20 lines:1-70Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 5 Sept 2026
Crop diseases pose a serious threat to agricultural yield and global food security. Accurate detection using Unmanned Aerial Vehicle (UAV) remote sensing imagery is of great significance for precision agriculture. However, this task remains challenging due to complex field backgrounds, diverse spectral-spatial characteristics of diseased leaf regions, irregular lesion boundaries, and variable texture patterns. To address these issues, this paper proposes a Task-Driven Attention VM-UNet (TDAVM-UNet), a novel deep learning model for crop disease detection from UAV imagery. The model integrates two task-driven attention modules: (1) Disease-Aware Dynamic Attention (DADA), which enhances the representation of diseased regions through disease feature enhancement, multi-scale dynamic channel attention, and texture-guided spatial attention; and (2) Channel-Spatial Visual State Space (CSVSS), which enables efficient long-range dependency modeling and local-global feature fusion while maintaining linear computational complexity. A hybrid loss strategy combining binary cross-entropy (BCE) loss, Dice loss, and cross-entropy (CE) loss with optimized coefficients is employed to address class imbalance and boundary delineation challenges. Extensive experiments are conducted on a self-constructed UAV crop disease unified mix dataset, comprising soybean disease images from Maharashtra, India, and rust disease images from wheat, corn, and other crops in Yangling, China, totaling 6,680 raw collected images, which after deduplication yields 5,000 images for experimentation. The results demonstrate that TDAVM-UNet achieves 26.87M parameters and 31.45 GFLOPs for 256×256 inputs, maintaining O(N) linear complexity (80% lower than TransUNet’s 156.78 GFLOPs), with 82.22% mIoU. This work provides a high-accuracy, robust, and computationally efficient method for UAV-based crop disease detection, offering significant technical support for precision agriculture applications.
Why it matches plant phenotyping methodsUAV画像から作物病害の病変領域・病害状態を推定する深層学習モデルを開発し、データセット上で性能評価しており、植物表現型取得・抽出手法が研究の中心である。
abstractthis paper proposes a Task-Driven Attention VM-UNet (TDAVM-UNet), a novel deep learning model for crop disease detection from UAV imagery.
Understanding root system architecture (RSA) is critical for improving crop productivity and resilience, yet phenotyping root traits such as root growth angle and rooting depth remains technically challenging, especially at high throughput. Here, we present ClearDepthIAS, a high-throughput imaging and analysis platform that enables non-destructive, automated quantification of root architecture traits in taproot system crops. By capturing and stitching 360° images of roots growing along the transparent walls of pots and applying deep learning-based segmentation (ClearDepth-WRT), we measured wall root shallowness (WRS)-a proxy for root growth angle-with high precision. We demonstrated for the tap root systems of soybean and canola that the system accurately detects root tips, quantifies their vertical distribution, and extracts biologically meaningful traits such as root area, distribution indices, and growth angles. Validation experiments in canola and soybean demonstrated that WRS can correlate with root crown architecture in mature plants, both in greenhouse and field settings. Furthermore, WRS and root distribution indices derived from ClearDepthIAS are predictors of early root architecture and can be correlated with root biomass distribution across soil depths under field conditions; however, environmental interactions may influence these relationships and weaken or even negate such correlations, as observed when comparing field to field variation in root system architecture. Our system enables efficient phenotyping of genetically diverse populations, with medium to high trait heritability, supporting its utility for genome-wide association studies and breeding. ClearDepthIAS accelerates the development of root ideotypes for improved resource acquisition and carbon sequestration, offering a scalable tool for supporting climate-resilient agriculture.
Why it matches plant phenotyping methods植物根系形態を自動取得・定量化する画像解析プラットフォームを開発し、精度と圃場での妥当性を検証しており、フェノタイピング手法が研究の中心である。
abstractwe present ClearDepthIAS, a high-throughput imaging and analysis platform that enables non-destructive, automated quantification of root architecture traits
Abstract. As key components of agricultural management, planting and harvesting schedules have strongly influenced crop production by defining the length of the crop growing season and shaping the environmental conditions crops experience. Accurate knowledge of these management data is crucial for enhancing crop yield estimates by capturing the timing of crop development relative to weather and soil conditions, assessing climate adaptation by tracking shifts in farming practices over time, and supporting agricultural carbon accounting. Yet, existing planting and harvesting date datasets are largely based on state-level statistics or rule-based calendars that overlook intra-regional variability and the influence of human decision-making. The absence of long-term, high-resolution planting and harvesting date information hinders our ability to reconstruct historical agricultural practices and assess their agronomic and environmental consequences. In this study, we introduce CropPlantHarvest, the first dataset of annual corn and soybean planting and harvesting dates across the U.S. Midwest at 500 m resolution from 2001 to 2024. Planting dates are estimated using CropSow, an integrative remotely sensed crop modeling system that aligns simulated crop growth trajectories with satellite observations to retrieve field-level planting dates. Harvesting dates are retrieved using the Normalized Harvest Phenology Index (NHPI), a novel index that integrates Normalized Difference Vegetation Index (NDVI) and near-infrared (NIR) reflectance to detect harvesting events by capturing the distinct spectral transition from senescent crops to exposed crop residues. Validation against USDA crop progress reports and field-level dataset demonstrates high accuracy of CropPlantHarvest, with a mean absolute error of approximately 5 d for both crop species. This large spatial and temporal dataset captures management-driven variability in crop season timing and duration, supporting improved modeling of crop yields, greenhouse gas emissions, and resource use. It could also serve as a benchmark for refining remote-sensing phenology products and evaluating the agro-environmental impacts of evolving crop management decisions. CropPlantHarvest is available at https://doi.org/10.5281/zenodo.16967482 (Liu and Diao, 2025).
Why it matches plant phenotyping methods衛星観測と作物モデルによる圃場レベルの作付・収穫時期推定手法を開発し、NHPIを提案して独立データで検証した大規模データセット研究であり、植物の生育・収穫状態の取得が中心的です。
abstractPlanting dates are estimated using CropSow, an integrative remotely sensed crop modeling system that aligns simulated crop growth trajectories with satellite observations to retrieve field-level planting dates.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicOur CropPlantHarvest dataset, which provides planting and harvesting dates for corn and soybean fields at 500 m spatial resolution across the U.S. Midwest from 2001 to 2024, can be accessed via Zenodo: https://doi.org/10.5281/zenodo.16967482 (Liu and Diao, 2025).Open asset ↗Zenodo · 10.5281/zenodo.16967482lines:322-333Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
Soybean leaf morphology is an important breeding trait that requires large-scale phenotyping in commercial breeding programs. Conventional leaf phenotyping still relies on manual destructive measurements, which are labor-intensive and inefficient. Low-altitude unmanned aerial vehicles (UAVs) have emerged as a high-throughput phenotyping platform, but the retrieval of leaf morphology in densely occluded canopies remains challenging due to complex canopy backgrounds, illumination heterogeneity, and leaf overlap under field conditions. To address this issue, this study developed an integrated UAV framework that couples a YOLOv10-based leaf detection module, a novel structure-adaptive segmentation network (DynamicU), and a regression-based trait prediction model for retrieving a set of leaf morphological parameters across 273 soybean genotypes under field conditions. The YOLOv10 detector reliably localized individual leaves under complex canopy conditions, achieving a mean average precision (mAP@50) of 0.84. Subsequently, the DynamicU network, whose architecture was automatically optimized via Emperor Penguin Optimization, achieved a segmentation accuracy of 97.2% and a mean Intersection over Union of 93.8%, substantially outperforming conventional models. Using random forest regression, the framework retrieved relative leaf shape traits, including length-to-width ratio and dissection index, with markedly higher accuracy (R 2 =0.97), compared to absolute morphological traits, including leaf length, width, perimeter, and area (R 2 : 0.76–0.84). Notably, relative leaf shape traits showed positive associations with oil yield per plant and protein yield per plant, supporting their potential as complementary indicators for screening soybean germplasm with differential industrial product output. This end-to-end framework establishes a reliable bridge between UAV remote sensing and leaf-level morphological quantification, advancing high-throughput phenotyping capabilities to support precision breeding in soybean.
Why it matches plant phenotyping methodsUAV画像、葉検出・セグメンテーション・回帰モデルを統合し、圃場でダイズ葉形態を自動定量するフレームワークの開発と性能評価が研究の中心であるため。
abstractthis study developed an integrated UAV framework that couples a YOLOv10-based leaf detection module, a novel structure-adaptive segmentation network (DynamicU), and a regression-based trait prediction model for retrieving a set of leaf morphological parameters across 273 soybean genotypes under field conditions.
Crop leaf diseases cause 10–40% annual yield losses, yet timely field diagnosis remains difficult. Vision-language models (VLMs) lift recognition accuracy with rich textual descriptions, but multimodal pipelines are too slow for real-time field use because they require text processing at inference. We present MTL-AWL, a framework built on a training–inference asymmetry: VLM text serves as privileged training-time supervision, and two coupled mechanisms—one retaining VLM semantics in the image encoder and one exploiting them—enable image-only deployment at multimodal accuracy. A modal-dropout strategy (p=0.6) intermittently masks the VLM text sequence during training, forcing the image encoder to retain cross-modal representations independently. An adaptive multi-task loss jointly optimizes InfoNCE contrastive alignment, attention diversity, and modality consistency under learnable softmax weights, consistently converging to a dominant contrastive weight (55% on soybean, 68% on PlantDoc)—identifying cross-modal alignment as the primary mechanism of VLM knowledge transfer. At inference, the model reaches 818 FPS (3.7× faster than multimodal methods) at only 0.41% accuracy cost, attaining 99.30%/98.89% (multimodal/image-only) on soybean and 72.65%/68.80% on PlantDoc—compact enough for real-time, offline field screening.
Why it matches plant phenotyping methods葉画像から植物病害状態を推定する画像ベース手法を開発し、複数データセットで精度・速度を評価しており、フェノタイピング手法が中心である。
abstractWe present MTL-AWL, a framework built on a training–inference asymmetry: VLM text serves as privileged training-time supervision, and two coupled mechanisms—one retaining VLM semantics in the image encoder and one exploiting them—enable image-only deployment at multimodal accuracy.
Reproduction assets foundThe paper's Data Availability Statement links a public Dryad DOI for the soybean leaf disease image dataset used in the study's phenotyping/recognition experiments. No author code or model release is stated.Dataset · publicThe datasets utilized in this study are openly accessible. The soybean dataset is available at https://doi.org/10.5061/dryad.41ns1rnj3 .Open asset ↗Dryad · 10.5061/dryad.41ns1rnj3lines:441-459Code / dataset availability confirmedEurope PMC · Crossref · checked 5 Sept 2026
Plants are geometrically and topologically complex objects, and methods and devices that produce plant point clouds often miss parts due to self occlusions, making further analysis, such as phenotypic trait extraction or 3D reconstruction, difficult. We introduce A-Occ-Plant , a novel method for point cloud completion. The first novelty of our algorithm is converting point clouds into a set of images, which are then completed using 2D amodal segmentation. The images are then converted into a complete point cloud by using view-consistent Gaussian splats. The second novelty is the use of a coarse-to-fine hierarchical Transformer with cross-scale attention. The completed soft masks are fused into a continuous 3D density field using Gaussian splatting, removing the need for external pose estimation or fixed-size inputs. We introduce a synthetic dataset using a procedural model and a real-world plant reconstruction benchmark with artificially generated occlusions. We further benchmark A-Occ-Plant against representative 3D point-cloud completion methods, demonstrate that it recovers downstream phenotypic traits (leaf count, leaf angle, plant height), and show that it generalizes to another crops (soybean). A-Occ-Plant achieves a 264.8% improvement in LPIPS and an 8.3% gain in SSIM compared to the current state of the art, while using only 2.3% of the parameters and running 39.4× faster. We release our code at https://github.com/JaeLee18/PlantPhenomics_Occlusion.
Why it matches plant phenotyping methods植物の遮蔽点群を補完し、葉数・葉角度・草丈という表現型形質を復元する手法を開発しており、データセット作成とベンチマーク検証も中心的に行っている。
abstractWe introduce A-Occ-Plant , a novel method for point cloud completion.
Reproduction assets foundThe paper explicitly releases authors' code and sample data (inference code, sample data for reproducing results) via a Google Drive project download and a GitHub repository, both with explicit availability statements and public URLs.Code · publicThe full code and data at https://github.com/JaeLee18/PlantPhenomics_Occlusion .Open asset ↗JaeLee18/PlantPhenomics_Occlusionlines:386-410Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · checked 15 Sept 2026
Global food security requires crop improvement strategies that can respond to population growth, climate variability and increasing constraints on agricultural resources. Conventional plant breeding has contributed substantially to crop productivity, yet long selection cycles and dependence on extensive field evaluation can limit the rate of genetic gain. This review synthesises advances in genomics, phenomics and machine learning for next-generation crop breeding, with emphasis on their combined contribution to selection accuracy and breeding efficiency. Key genomic approaches discussed include whole-genome sequencing, reference and pan-genome resources, genome-wide association studies, genomic selection and CRISPR-Cas-based genome editing. The review also examines high-throughput phenotyping platforms, including controlled-environment systems, ground-based robots, UAV-based remote sensing and root phenotyping tools. Machine learning approaches, ranging from random forest and support vector machines to convolutional neural networks, recurrent networks, transformers and explainable artificial intelligence, are considered in relation to genomic prediction, image analysis and breeding decision support. Multi-omics integration, data management, FAIR principles and an integrated genomics-phenomics-ML breeding pipeline are reviewed as enabling components for practical deployment. Crop-specific examples from wheat, rice, maize, soybean and legumes illustrate the potential and constraints of these technologies. The review further identifies key challenges, including phenotyping bottlenecks, genotype-environment interaction, data governance, model interpretability and regulatory uncertainty.
Why it matches plant phenotyping methods植物フェノタイピング手法を中心に、ハイスループット計測プラットフォーム、画像解析、機械学習、UAV・ロボット・根系計測などをレビューしているため。
abstractThis review synthesises advances in genomics, phenomics and machine learning for next-generation crop breeding
Accurate 3D plant models are crucial for computational phenotyping and physics-based simulation; however, current approaches face significant limitations. Learning-based reconstruction methods require extensive species-specific training data and lack editability for hypothesis-driven research. Procedural modeling offers parametric control and large model variability but demands specialized expertise in geometric modeling and an in-depth understanding of complex procedural rules, making it inaccessible to domain scientists. We present FloraForge, an LLM-assisted framework that enables domain experts to generate biologically accurate, fully parametric 3D plant models through iterative natural language Plant Refinements (PR) during template creation, minimizing the need for programming expertise. Our co-design workflow leverages LLM-assisted code generation to progressively refine Python scripts that generate parameterized complex plant geometries as Non-Uniform Rational B-Spline (NURBS) surface representations, with botanical constraints. Plant organs are represented as spline surfaces that can be easily tessellated into polygonal meshes with arbitrary precision, ensuring compatibility with functional structural plant analysis workflows such as light simulation, computational fluid dynamics, and finite element analysis. We demonstrate the framework by generating procedural models of multiple maize genotypes, soybean (which shares the procedural generator with mung bean), and mung bean, with one plant tracked across different developmental stages. We fit procedural models to empirical LiDAR and NeRF-derived point cloud data through manual refinement of the Plant Descriptor (PD), a human-readable YAML file originally templated by the LLM, obtaining consistently low mean symmetric Chamfer distances that indicate close agreement between generated models and measured plant geometry. The pipeline generates dual outputs: triangular meshes (represented as STL or OBJ files) for visualization and triangular meshes with additional parametric metadata for quantitative analysis (stored as SMESH files). We further illustrate analysis-ready use by coupling the procedurally generated models to the HELIOS framework to simulate diurnal photosynthetically active radiation interception in virtual maize and mung bean fields across growth stages. Our framework uniquely combines pre-trained LLM-assisted template creation, mathematically continuous representations that support both phenotyping and rendering, and direct parametric control through the PD. The framework makes sophisticated geometric modeling accessible to plant science researchers while maintaining mathematical rigor through biologically interpretable parameterizations; additionally, the iterative PR dialogue produces an explicit record of model properties that is typically absent in conventional procedural modeling pipelines.
Why it matches plant phenotyping methods植物の3D形状を生成・編集し、LiDAR/NeRF点群との適合で検証する、計算機フェノタイピング向けの中心的手法開発である。
abstractWe present FloraForge, an LLM-assisted framework that enables domain experts to generate biologically accurate, fully parametric 3D plant models
The spatial organization of essential, nonessential, and toxic metal(loid) elements (MEs) within plant cells underpins physiological function. Yet, comprehensive subcellular imaging of the full ME spectrum remains challenging due to trade-offs among spatial resolution, elemental coverage, and structural correlation. Here, we present an integrated scanning electron microscopy-focused ion beam-time-of-flight-secondary ion mass spectrometry platform that overcomes these limitations by achieving nanoscale coregistration of ultrastructure with ME distribution. Applying this high-fidelity workflow to Arabidopsis , soybean, and wheat, we constructed single-cell metallome maps revealing an evolutionarily conserved subcellular architecture: chloroplasts enrich essential MEs (e.g., magnesium, iron, copper), whereas vacuoles compartmentalize nonessential [e.g., lanthanum (La)] and toxic MEs [e.g., cadmium (Cd), lead, arsenic]. We demonstrate that while this architecture remains stable under homeostasis, it undergoes dynamic, stimulus-specific, and dose-dependent remodeling under stress. Low-dose La(III) enhances pairwise and higher-order colocalizations of essential MEs within chloroplasts, correlating with improved photosynthetic efficiency and growth. High-dose La(III) induces nonphysiological La-ME associations and, critically, drives aberrant Cd(II) accumulation in chloroplasts-revealing a cross-toxicity mechanism wherein La(III) disrupts native sequestration barriers. In contrast, although high-dose Cd(II) is largely excluded from chloroplasts, it triggers a widespread redistribution of essential MEs, progressively eroding spatial organization. Thus, while both ions inhibit growth, they perturb metallomic networks via distinct mechanisms: La(III)-mediated disruption of sequestration vs. Cd(II)-induced systemic compartmental collapse. Our findings establish that subcellular ME networks are dynamically regulated and orchestrate physiological outcomes.
Why it matches plant phenotyping methods植物細胞内の金属元素分布と超微細構造を取得する統合イメージング基盤とワークフローの開発が中心であり、植物の生理状態・ストレス応答に結び付けて実証している。
abstractHere, we present an integrated scanning electron microscopy-focused ion beam-time-of-flight-secondary ion mass spectrometry platform that overcomes these limitations by achieving nanoscale coregistration of ultrastructure with ME distribution.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · Crossref · checked 6 Sept 2026
Abstract Rapid and accurate quantification of crop biomass using multisource UAV imagery–derived features, such as plant height, vegetation indices, and texture indices demonstrates strong potential for soybean high-throughput phenotyping. The indeterminate growth habit of soybean, alongside extensive nodulation and intense inter-plant competition, necessitates individual plant level (IPL) monitoring to quantify plant-specific nitrogen fixation and competitive vigor. However, most studies aggregate measurements at multi-plants at the plot level, thereby masking these soybean-specific traits. This study aims to develop and evaluate a UAV imagery-based framework for estimating soybean biomass at the IPL, with the objective of characterizing high-resolution spatial variability and supporting high-throughput phenotyping. Regions of interest (ROIs) for IPL data acquisition were defined as rectangular plots based on planting density and were generated from early-stage imagery before canopy overlap occurred. Using these ROIs, structural information (SIs) including plant height (PH) and vegetation fraction (VF)), vegetation indices (VIs), and texture indices (TIs) were derived for each individual plant from RGB and multispectral imagery and organized into sensor-specific feature groups. Recursive feature elimination was applied to select optimal features, which were then used as inputs for machine learning architectures including support vector regression (SVR) with a linear kernel, Random Forest (RF), and XGBoost (XGB). Among them, the SVR model using fused multisource features (SIs + VIs + Tis) achieved the best performance, with R² = 0.88, RMSE = 55.34 g, and rRMSE = 8.50% on an independent test dataset. The results show that: (1) tree-based models, including XGB and RF may suffer from overfitting due to limited sample size and feature redundancy, whereas linear SVR showed better generalization; (2) fusing RGB and multispectral features consistently improved biomass estimation accuracy. Inparticular, near-infrared and red-edge-based indices such as RECI and NDRE, along with VF and PH, were identified as important predictors, while texture indices were not selected as significant features; and (3) the proposed framework enabled spatially explicit IPL biomass mapping and time-series analysis, revealing variability in growth conditions and distinct growth trajectories. The framework provides a reliable solution for IPL soybean biomass estimation with practical potential for UAV-based agricultural decision-making.
Why it matches plant phenotyping methodsUAV画像由来の特徴量と機械学習により個体レベルのダイズ biomass を推定・評価する枠組みが研究の中心であり、植物表現型の取得・抽出手法に該当する。
abstractThis study aims to develop and evaluate a UAV imagery-based framework for estimating soybean biomass at the IPL
Artificial intelligence applied to plant phenotyping is crucial for consistent results, as stomata classification under stress impacts physiology, water use efficiency, and productivity. Manual analysis is laborious and error-prone, limiting the efficiency and accuracy of evaluations. In this context, this study developed a soybean-specific dataset from water deficit (WD) and well-watered (WW) plants, training YOLOv8 model for automated detection and classification of open vs. closed stomata. Soybean plants were grown under water deficit and well-watered conditions, generating significant variations in stomatal structural opening and associated gas exchange traits. To capture stomata variations, epidermal printing techniques were employed, with images obtained by microscopy. The dataset was labeled using the intelligent polygon tool of the Roboflow application, with 269 images of the adaxial and abaxial surfaces of leaves annotated in two categories: open and closed stomata. The images underwent geometric transformations to facilitate model training. The results demonstrated that the YOLOV8 neural network achieved precision recall and mAP greater than 90%, highlighting its effectiveness in detecting and classifying stomata. By integrating automated classification of aperture states (open and closed) with a defined physiological stress context in soybean, this work establishes a dataset specifically designed for functional analysis. This approach extends the applicability of deep learning toward stress-oriented plant physiology studies, offering a robust tool for evaluating crop adaptation under climate change scenarios.
Why it matches plant phenotyping methodsヨロウ豆の気孔開閉状態を画像から自動検出・分類するYOLOv8手法とデータセットを開発・評価しており、植物表現型取得が研究の中心である。
abstractthis study developed a soybean-specific dataset from water deficit (WD) and well-watered (WW) plants, training YOLOv8 model for automated detection and classification of open vs. closed stomata.
Soybean production is significantly affected by crop diseases and improper pesticide use, which hinder effective disease management and reduce yield. In this study, we propose an efficient multi-task convolutional neural network (CNN) framework for the simultaneous detection of soybean seed diseases and pesticide presence from seed images. The model leverages a shared feature extraction backbone with task-specific output heads to learn complementary features for both disease classification and pesticide detection. A dataset of 429 soybean leaf images was preprocessed using normalization and augmentation techniques and split into training, validation, and testing sets. We evaluated three backbone architectures VGG19, MobileNetV3, and ConvNeXt within the multi-task framework. Experimental results demonstrate that the approach maintains computational efficiency suitable for real-world deployment while achieving high performance, with accuracies of 95%, 96%, and 97% for MobileNetV3, VGG19, and ConvNeXt, respectively. Additionally, explainable AI methods, such as Grad-CAM, highlight regions of focus for both tasks, making the model's decision-making process interpretable. This framework provides a practical tool for informed crop management and agricultural monitoring.
Why it matches plant phenotyping methods植物画像から病害状態を推定するCNN手法の開発・評価が研究の中心であり、病害表現型の画像ベース推定に該当する。
abstractwe propose an efficient multi-task convolutional neural network (CNN) framework for the simultaneous detection of soybean seed diseases and pesticide presence from seed images.
Reproduction assets foundThe paper's authors publicly released their custom analysis code (preprocessing, training, evaluation) on GitHub, matching an allowed URL. The enriched Kaggle image/annotation dataset is also public but its URL is not among the allowed URLs, so it is not listed.Code · publicThe custom code developed for this study is publicly available on GitHub at https://github.com/fikaduberie/Soybean-Disease-and-Pest (version v1.0).Open asset ↗fikaduberie/Soybean-Disease-and-Pesthtml-lines:1281-1329Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
Why it matches plant phenotyping methods大豆貯蔵タンパク質比を測定するSDS-PAGEプロトコルを開発・比較検証し、育種系統の表現型判別に適用しており、測定法が中心的である。
abstractThis study presents an improved SDS-PAGE protocol optimized for quantifying the 11S/7S ratio with improved accuracy, reproducibility, and cost-effectiveness.
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 5 Sept 2026
Drought is the major abiotic stress limiting soybean growth and yield, yet accurately identifying genotypes that sustain yield under rainfed conditions remains a major bottleneck in soybean breeding. Canopy wilting scores are widely used as a proxy for evaluating plant responses to drought stress. However, most assessments rely on leaf-level visual observations that are inherently subjective and typically based on single time-point scores, providing only a snapshot of stress expression and failing to capture their relationship with yield retention under rainfed conditions. To address these limitations, this study used Unmanned Aerial Vehicle (UAV)-based high-throughput phenotyping at a single growth stage (R4/R5) as a more quantitative and objective alternative to visual scoring, with closer relevance to yield performance under drought conditions. From 2023 to 2025, a total of 85 soybean genotypes developed by soybean breeding programs in Arkansas, Missouri, Kansas, and North Carolina, along with commercial checks, were evaluated under irrigated and rainfed conditions in Stuttgart, Arkansas. Visual canopy wilting scores were recorded at R4/R5, along with vegetation indices captured using UAV-based multispectral imagery. UAV-derived indices showed significant correlations with yield ( r = 0.22 to 0.45, p<0.05) under rainfed conditions. In contrast, visual canopy wilting scores displayed weak and inconsistent associations with yield ( r = -0.28 to 0.35, p<0.05), suggesting limited ability to capture yield retention under rainfed conditions. Unsupervised k -means clustering ( n = 2) of UAV-derived vegetation indices separated genotypes into two distinct canopy response groups that were consistent across 2023 to 2025 rainfed seasons. Significant differences were observed among clusters for several vegetation indices (ARI, CIG, CIRE, GSAVI, GNDVI, GOSAVI, OSAVI, NDVI), indicating contrasting canopy stress responses. Under rainfed conditions, these UAV-defined clusters also differed for grain yield (2023: 1,925.6 vs 1,703.1 kg/ha; 2024: 1,849.9 vs 1,229.2 kg/ha; 2025: 2,056.7 vs 1,773.8 kg/ha), whereas visual wilting scores failed to distinguish yield-retaining genotypes. Overall, UAV-based high-throughput phenotyping offers a robust and yield-relevant alternative to visual wilting scores, supporting the development of drought-tolerant soybean germplasm and cultivars.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像による植生指数抽出を、目視評価と比較・検証し、干ばつ応答および収量保持に関連する表現型測定法として中心的に評価している。
abstractthis study used Unmanned Aerial Vehicle (UAV)-based high-throughput phenotyping at a single growth stage (R4/R5) as a more quantitative and objective alternative to visual scoring
Accurate and scalable soybean crop health monitoring remains a major challenge in precision agriculture due to environment variability, inconsistent lighting conditions, and significant differences between the ground-level leaf imagery and UAV-based aerial imagery. Most existing deep learning approaches treat these two sensing modalities separately without properly exploring cross-scale feature transferability or measuring the domain gap that exists between the sensing scales. As a result, developing unified and deployment-ready crop health monitoring systems that can effectively leverage the more accessible leaf-level datasets, collected without specialized equipment or regulatory constraints, to improve UAV-scale inference remains difficult. In order to address this limitation, we propose a multi-scale soybean crop health assessment framework that integrates ground-level leaf imagery and UAV-based aerial imagery from the MH-SoyaHealthVision dataset across four health conditions, which include Healthy, Mosaic Virus, Pest attack, and Rust. CLAHE, Gray-World color constancy correction, and illumination normalization is incorporated into a structured pre-processing pipeline and further applied to reduce illumination bias and enhance cross-domain feature consistency. Six deep learning backbones were comprehensively evaluated for leaf-level classification, with MaxViT and ConvNeXt achieving the best performance. Their static weighted ensemble further improved accuracy to 87.08%. Cross-scale evaluation showed that zero-shot leap-to-UAV transfer achieved only 40% accuracy, thus highlighting the presence of a substantial domain shift. Fine-tuning improved UAV classification performance to about 97%, while a supervised contrastive learning framework specifically designed for cross-scale feature alignment further increased accuracy to approximately 98% with better convergence stability. Feature embedding analysis using PCA, t-SNE, and silhouette metrics demonstrated considerable improvements in inter-class separability (0.59 vs. 0.19) and reduced domain discrepancy (0.0336 vs. 0.114) under contrastive learning. These findings suggest that supervised alignment can generate more class-discriminative representations with lower cross-scale domain discrepancy, making them more suitable for scalable multi-scale cross-health monitoring.
Why it matches plant phenotyping methods葉およびUAV画像からダイズの健康状態・病害を推定する画像ベースの表現学習フレームワークを開発し、複数モデル、クロススケール転移、微調整、教師ありコントラスト学習を比較・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 multi-scale soybean crop health assessment framework that integrates ground-level leaf imagery and UAV-based aerial imagery
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 14 Sept 2026
Predicting canopy traits non-destructively is important for understanding crop growth and improving phenotyping efficiency. Hyperspectral reflectance provides detailed spectral information, but the role of band selection in regression-based trait prediction at the canopy scale remains unclear. In this study, we evaluated the effects of different band-selection algorithms on the prediction accuracy of aboveground biomass (AGB), leaf area index (LAI), and canopy cover (CC) in soybeans across multiple sites, years, cultivars, and irrigation treatments. We compared a full-band partial least squares regression (PLS) model with three band-selection methods (PLS-Variable Importance in Projection (VIP), Bootstrapped least absolute shrinkage and selection operator (LASSO) (BoLASSO), and an ensemble approach), and model performance was assessed using independent validation datasets. The results showed that the effectiveness of band selection depended on the target trait. Full-band PLS provided the highest accuracy for AGB, whereas BoLASSO achieved comparable accuracy to PLS for LAI and CC using a reduced number of selected bands. The selected wavelengths were located mainly in the visible, red-edge, and near-infrared regions. These results indicate that band-selection strategies should be tailored to the target trait and provide a basis for efficient band design in crop phenotyping.
Why it matches plant phenotyping methodsハイパースペクトル反射を用いた作物形質推定について、バンド選択アルゴリズムと回帰モデルを比較・独立検証しており、フェノタイピング手法の技術評価が中心である。
abstractPredicting canopy traits non-destructively is important for understanding crop growth and improving phenotyping efficiency.
Plant phenotyping relevance match · UnverifiedCrossref · OpenAlex · Europe PMC · checked 15 Sept 2026
Accurate acquisition of plant phenotypes is crucial for elucidating plant growth and development, underlying genetic mechanisms, and responses to environmental stimuli. Traditional three-dimensional (3D) phenotyping mainly captures geometric traits such as height, leaf area, and canopy volume, while overlooking physiological and biochemical information. Here, we present a hyperspectral point clouds generation method based on PlantGaussian (a 3D Gaussian Splatting technique) that integrates structural and spectral information, extending 3D phenotyping beyond geometry to include physiology. High-quality plant point clouds were first reconstructed using PlantGaussian, and hyperspectral images(HSI) were mapped onto them to produce hyperspectral point clouds. In potted soybean experiments, we built predictive models linking hyperspectral reflectance to SPAD (chlorophyll content) and EWT (equivalent water thickness), and visualized their 3D distributions. The hyperspectral point clouds achieved strong predictive performance for SPAD ( R 2 = 0.78, RMSE = 2.05) and EWT ( R 2 = 0.80, RMSE = 1.07), thereby validating the approach. It further revealed clear vertical stratification within the canopy, highlighting significant spatial heterogeneity of SPAD and EWT in individual plants. Temporal monitoring from August 6 to 21, 2025, captured a sharp increase in EWT after heavy rainfall on August 11. Overall, our results demonstrate that hyperspectral point clouds enable accurate, non-destructive trait estimation and provide a powerful tool for exploring plant function, monitoring stress responses, and advancing precision agriculture.
Why it matches plant phenotyping methods植物の3D形態とハイパースペクトル情報を統合してSPAD・EWTを推定する手法を開発し、予測性能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。
abstractHere, we present a hyperspectral point clouds generation method based on PlantGaussian (a 3D Gaussian Splatting technique) that integrates structural and spectral information, extending 3D phenotyping beyond geometry to include physiology.
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
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 15 Sept 2026
Main conclusion The progress of soybean phenotypic detection and intelligent sensing technologies has been reviewed under salt-alkali stress , and an integrated approach combining three-dimensional imaging with near-infrared spectroscopy has been proposed to construct full-spectrum three-dimensional images. The approach could provide a reference for the breeding of salt-alkali-tolerant soybean varieties and the optimization of cultivation practices. Soil saline-alkali is one of the major environmental factors limiting global agricultural development, posing a serious challenge to normal crop growth, resource use efficiency, and sustainable agricultural development. Soybeans are a vital oilseed crop and plant-based protein source, and their phenotypic traits are significantly affected by saline-alkali stress, severely limiting soybean grain yield and quality. With the rapid advancement of technologies such as intelligent sensing and big data, this progress has driven new developments in plant phenomics detection, offering fresh insights into germplasm resource evaluation, breeding, gene function, and the cultivation of salt-alkali stressed soybeans. This article introduces the impact of salinity-alkali stress on soybean "phenotype-environment-gene" information, reviews the technical progress and application fields of traditional phenotypic detection methods for obtaining various phenotypic indicators across crops, and focuses on a rapid detection method of soybean phenotype under salinity-alkali stress. This paper analyzes the current state of research on detecting phenotypic indicators of soybeans under saline-alkali stress using intelligent sensing methods, including near-infrared spectroscopy, image recognition, and three-dimensional imaging. It is anticipated that through the integration of three-dimensional imaging and near-infrared spectroscopy, forming "full-spectrum three-dimensional images" with spatial structure and spectral information, this approach will advance the breeding and cultivation of superior salt-alkali tolerant soybean varieties through "intelligent data-driven" methods.
Why it matches plant phenotyping methods植物フェノタイピング指標の迅速検出技術を中心に、画像認識、三次元 imaging、近赤外分光などをレビューし、統合的な表現型取得法を提案する方法論的レビューである。
abstractThe progress of soybean phenotypic detection and intelligent sensing technologies has been reviewed under salt-alkali stress
Fractional Vegetation Cover of Crops (CropFVC) is a critical canopy parameter for monitoring crop growth, yet the behavior of widely used global FVC products (GLASS, GEOV1, GEOV2, and GEOV3) over croplands remains insufficiently understood due to fragmented validation references and limited crop-specific assessments. This study compiled a multi-source global CropFVC reference dataset (2000–2024) by integrating five international validation networks, the literature-derived samples, and newly acquired UAV and Jilin-1 satellite-derived CropFVC samples from China in 2024. The references were organized into three complementary validation contexts (V1~V3) to examine product behavior under different temporal coverage, crop purity, and reference conditions, together with spatio-temporal observations at the KONZ site. Results show that (1) across validation contexts, the evaluated products showed consistent behavior patterns, including shared overestimation under dense canopy conditions and reduced differences at low FVC levels; (2) spatio-temporal analysis at the KONZ site confirmed that peak-season deviations reflect shared response behavior rather than site-specific reference uncertainties; (3) historical mixed references (V1~V2) showed similar bias structures, whereas crop-specific validation (V3) preliminary revealed clearer crop-dependent responses, with predictive difficulty following winter wheat > maize > rice > soybean and improved stability after integrating 2024 observations. The integration of recent high-resolution crop observations expands existing global CropFVC references and enables behavior-oriented interpretation of global FVC products beyond simple accuracy ranking, providing an updated validation perspective for future development and application of global CropFVC products in agricultural monitoring.
Why it matches plant phenotyping methods作物キャノピーのFVCという植物形質を対象に、複数の全球FVC推定プロダクトを多様な参照データで体系的に検証し、UAV・衛星観測を含むCropFVC参照データセットを構築している。形質取得・検証が研究の中心である。
titleMulti-Context Validation of Global Fractional Vegetation Cover Products in Croplands Using Multi-Source Crop FVC References
Nodule color and morphology are key readouts of legume symbiotic performance. However, long-term preservation of post-excavation nodules with intact morphology, color, and microbial cleanliness remains a major challenge. This study developed a two-stage aqueous-phase preservation method (TAPP) that enables rapid structural fixation and long-term chemical stabilization. A comprehensive evaluation was subsequently established, incorporating composite morphological score (0-5), color difference (ΔE) and its piecewise slope over time ([Formula: see text]), and visible contamination grade (0-3). Peanut and soybean nodules from multiple regions and cultivars were tracked for 24 months under five preservation methods: TAPP, FormalinCu, TAPP-Resin, Resin, and AirDry. TAPP showed the best overall preservation, with composite morphological scores of 4.65 ± 0.14 for peanut and 4.63 ± 0.22 for soybean at 24 months, and no visible mold. Color change slowed over time: [Formula: see text] decreased from 1.83 to 1.10 ΔE·month - 1 during 0-1 month to 0.16 ΔE·month - 1 during 12-24 months, yielding final ΔE values of 10.53 ± 1.88 and 10.32 ± 1.93, respectively. Notably, TAPP pretreatment markedly improved resin-embedded samples, demonstrating scalability and flexible deployment. In addition, this study further proposes a stage-wise workflow that integrates on-site pre-fixation, long-distance transport, and long-term storage to enable cross-regional circulation and collaborative phenomics of oxidation-prone, dehydration-sensitive nodules. Together, this work establishes a standardized, traceable workflow to preserve and benchmark legume root nodule phenotypes, supporting cross-laboratory comparability and longitudinal cross-source analyses.
Why it matches plant phenotyping methodsマメ科根粒の形態・色・汚染状態という植物表現型を長期保存し、定量評価・比較する手法と標準化ワークフローが研究の中心であるため。
abstractThis study developed a two-stage aqueous-phase preservation method (TAPP) that enables rapid structural fixation and long-term chemical stabilization.
Crop diseases pose a significant threat to agricultural productivity and global food security. Timely and accurate detection of such diseases is crucial for improving both crop yield and quality. While numerous deep learning approaches rely solely on image data for disease identification, they often overlook the complementary value of textual information in enhancing visual analysis. To address this limitation and effectively fuse features from different modalities, we propose a Cross-Model fusion framework based on a vision-language model that integrates cross-attention and gated fusion mechanisms for crop disease recognition. Our approach utilizes the Zhipu.ai multi-modal model to generate comprehensive textual descriptions of diseased crop leaves, including global description, local lesion description, and color-texture description. These textual descriptions are then encoded into feature embeddings, while visual features are extracted using the ShuffleNet-v2 model as the image encoder. Subsequently, a cross-attention module aligns and fuses the two modalities, and a gated fusion module enables dynamic feature selection during the fusion process. Extensive evaluations on the Soybean Disease and PlantVillage datasets demonstrate that our method outperforms existing image-based models in terms of accuracy. Specifically, our model achieves recognition accuracies of 99.04% and 99.12% on the respective datasets, surpassing the ShuffleNet-V2 model by 1.09% and 2.53%, respectively. These results highlight the effectiveness of Cross-Model learning in integrating visual and textual cues for accurate and efficient disease recognition, offering a scalable solution for crop disease diagnosis.
Why it matches plant phenotyping methods植物葉の病徴を画像と言語情報から認識する融合フレームワークを開発し、複数データセットで既存手法と比較評価しているため、植物フェノタイピング手法が中心である。
abstractwe propose a Cross-Model fusion framework based on a vision-language model that integrates cross-attention and gated fusion mechanisms for crop disease recognition.
Reproduction assets foundThe paper's crop disease recognition experiments use two openly available image datasets, both with explicit public availability statements in the Data Availability section: the Soybean Disease dataset (Dryad DOI) and the PlantVillage dataset (Kaggle). No author analysis code, trained models, or generated text-annotaitDataset · publicThe datasets utilized in this study are openly accessible. The soybean dataset is available at https://doi.org/10.5061/dryad.41ns1rnj3.Open asset ↗Dryad · 10.5061/dryad.41ns1rnj3html-lines:403-424Dataset · publicThe plantvillage dataset is available at https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset.Open asset ↗Kaggle · plantvillage-datasethtml-lines:403-424Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · checked 5 Sept 2026
Abstract Background: The stink bug complex is one of the most damaging pests of soybean, reducing yield and seed quality. Genetic resistance remains the most sustainable and effective management strategy, but its quantitative inheritance and labor-intensive field phenotyping make its implementation in breeding programs challenging. Objective: This study explored high-throughput phenotyping (HTP) using unmanned aerial vehicles (UAVs) equipped with RGB cameras to evaluate a soybean population and the potential of phenotyping to stink bug resistance by correlating image-derived features and machine learning (ML) models. Methods: A population of 304 soybean lines was evaluated in alpha-lattice design trials across two seasons under natural infestations. Five resistance-related traits, grain yield (GY), healthy seed weight (HSW), number of days to maturity (NDM), tolerance (TOL), and leaf retention (LR), were manually scored and linked to UAV-derived vegetation indices (VIs) and texture indices (TIs). Three ML models (AdaBoost, SVM, MLP) were tested to predict these traits from aerial features. Results: Results showed that VIs, particularly Visible Atmospherically Resistant Index at the 25th percentile (VARI_P25), were consistently associated with resistance-related traits, while decision tree analysis highlighted TIs at 45° and 135° as complementary sources of structural information. Prediction ability was highest for GY, HSW, and NDM, especially in flights near flowering and maturity, but remained low for TOL and LR. Integrating multiple flights modestly improved accuracy, whereas cross-season predictions were unreliable. Nonetheless, indices such as VARI_P25 provided useful cross-season correlations for HSW and TOL, enabling early screening of less promising lines. Conclusion: This pioneering study demonstrates that UAV–ML pipelines can capture genetic signals of stink bug resistance in soybean, despite environmental complexity. These findings open new avenues for resistance phenotyping, supporting more efficient breeding strategies and accelerating genetic gains in soybean improvement.
Why it matches plant phenotyping methodsUAV画像と機械学習を用いて、ダイズの抵抗性関連形質や収量を推定するHTPパイプラインを技術的に評価しており、表現型取得・推定法が研究の中心である。
abstractThis study explored high-throughput phenotyping (HTP) using unmanned aerial vehicles (UAVs) equipped with RGB cameras to evaluate a soybean population and the potential of phenotyping to stink bug resistance by correlating image-derived features and machine learning (ML) models.
Soybean protein content is a key indicator of nutritional value and quality grade, and its determination is important for quality evaluation and cultivar selection. To overcome the time-consuming and costly limitations of conventional chemical assays, this study proposed a multiple linear learner ensemble importance-score wavelength selection (MLLEISWS) method to identify informative wavelengths from soybean near-infrared spectra and establish a partial least squares (PLS) model. MLLEISWS was compared with competitive adaptive reweighted sampling, successive projections algorithm, and uninformative variable elimination. Shapley additive exPlanations (SHAP) were applied to the MLLEISWS algorithm to interpret the selected wavelengths. Results showed that the PLS model developed using MLLEISWS achieved the best performance. With only 29 selected wavelengths, the coefficients of determination for the training and test sets reached 0.941 and 0.933, respectively. Root mean square errors were 0.490% and 0.514%, relative root mean square errors were 1.32% and 1.37%, and residual predictive deviation was 3.863, indicating predictive accuracy and stability. SHAP analysis showed that the selected wavelengths were located in protein-related spectral regions and corresponded to overtone and combination bands information from functional groups. MLLEISWS effectively reduced variable dimensionality while maintaining model performance.
Why it matches plant phenotyping methods大豆種子のタンパク質含量という植物形質を対象に、近赤外分光法と波長選択・PLSモデルを開発、比較評価しており、形質取得・推定手法が研究の中心である。
abstractthis study proposed a multiple linear learner ensemble importance-score wavelength selection (MLLEISWS) method to identify informative wavelengths from soybean near-infrared spectra and establish a partial least squares (PLS) model.
Background High-throughput automated image analysis holds great promise for plant breeding by enabling faster, more accurate assessment of traits relevant to crop improvement. Imaging-based systems, such as the CropReporter, allow automated quantification of photosynthetic parameters like PSII efficiency under ambient light from a top-down 2D perspective. However, standard analysis tools average values across the 2D top view, overrepresenting upper leaves and underrepresenting those in the lower canopy. Upper leaves may occlude lower ones, and due to the pinhole projection of the camera, lower leaves of the same size appear smaller in the image. Consequently, vertical heterogeneity in PSII efficiency within the canopy cannot be resolved using a single 2D image. Results To address these issues, we integrated top-view PSII efficiency data (by CropReporter) with 3D structural data from RGB point clouds (by MaxiMarvin). Alignment accuracy between MaxiMarvin and CropReporter was high, with R² ≥ 0.98 for the x-axis and R² ≥ 0.99 for the y-axis. The method was tested using Chenopodium quinoa, Glycine max, and Solanum tuberosum, exposed to salinity, waterlogging and drought stress respectively. In Chenopodium quinoa, it allowed precise determination of when senescence began in the lower leaves. In Solanum tuberosum, the reduction in PSII efficiency by drought was the same for all leaf layers, while in Glycine max, waterlogging stress most strongly affected the middle layer of the canopy. Conclusions This framework enables the 3D mapping of PSII efficiency across the vertical plant profile by combining top-view chlorophyll fluorescence imaging (CropReporter) with 3D structural data (MaxiMarvin). It reveals vertical variation in photosynthetic activity across canopy layers. With standard 2D chlorophyll fluorescence imaging it is difficult to distinguish between non-photosynthetic tissues like flower heads and lower layers of leaves, that might have the same PSII values. Using height-based filtering, taking data from the 3D mapping, such distinction can be made with the method presented in this paper. This allows estimating the PSII efficiencies of leaves only. By capturing layer-specific responses to abiotic stress and developmental changes, the method provides physiologically relevant input for crop growth modelling and highlights the importance of accounting for canopy structure in photosynthetic analyses.
Why it matches plant phenotyping methods2Dクロロフィル蛍光によるPSII効率を3D植物構造へ投影し、群落層別の葉の生理形質を推定する手法の開発・検証が中心である。
abstractTo address these issues, we integrated top-view PSII efficiency data (by CropReporter) with 3D structural data from RGB point clouds (by MaxiMarvin).
Reproduction assets foundThe authors state that the analysis scripts (2D–3D alignment pipeline) and the phenotyping data used in the study are included with the publication as supplementary material, accessible via the article DOI. This is a paper-specific, publicly available asset containing the authors' analysis code and data.Dataset · publicThe scripts and the data that were used in the current study are available and added to this publication.Open asset ↗lines:143-180Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
Abstract Spatio-temporal fusion (STF) has been widely used across various remote sensing applications, including environmental monitoring, land cover change detection, and water resource management by integrating multi-sensor data with different spatial and temporal resolutions. The objective of this study was to generate spatially and temporally fine-resolution imagery and to evaluate the performance of multiple STF algorithms and Consistent Adjustment of the Climatology to Actual Observations (CACAO) post-processing for parcel-level crop monitoring. The study was conducted in a soybean field located in Anseong, South Korea, using Planet SuperDove satellite imagery which has 3 m spatial resolution and near-daily temporal resolution, and Phantom 4 Multispectral Unmanned Aerial Vehicle (UAV) data which has 0.05 m spatial resolution, downscaled to target resolution 0.5 m, and 1–4 week irregular temporal resolution. Relative radiometric normalization was applied, followed by the implementation and comparison of four STF algorithms— Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM), Fitting, spatial Filtering and residual Compensation (FitFC), Flexible Spatiotemporal Data Fusion (FSDAF), and Variation-based Spatiotemporal Data Fusion (VSDF)—within a 4-fold cross-validation framework. CACAO post-processing was then employed to reconstruct temporally continuous NDVI trajectories, from which growth metrics such as Vegetation Growth Metrics (VGM)85 and VGMmax were derived. The validation results indicated that ESTARFM achieved the highest Normalized Difference Vegetation Index (NDVI) performance among the evaluated algorithms, with an Root Mean Square Error (RMSE) of 0.113 and the Universal Image Quality Index (UIQI) of 0.697, and CACAO further improved the results, with CA-ESTARFM providing the highest NDVI accuracy, with an RMSE of 0.108 and a UIQI of 0.740. NDVI histogram and spatial analyses demonstrated that CA-ESTARFM achieved the most consistent agreement with UAV observations while preserving fine-scale spatial heterogeneity. In addition, intra-field vegetation assessment using VGM85 and VGMmax confirmed that CA-ESTARFM enhanced the reliability of crop growth evaluation compared to simple linear interpolation of UAV observations. The proposed framework demonstrates strong potential for applications in comprehensive crop monitoring, precision agriculture management, and yield forecasting.
Why it matches plant phenotyping methods衛星・UAV画像の時空間融合アルゴリズムを比較検証し、NDVI軌跡から作物生育指標を抽出する方法が研究の中心であるため、植物フェノタイピング手法として採択する。
abstractThe objective of this study was to generate spatially and temporally fine-resolution imagery and to evaluate the performance of multiple STF algorithms and Consistent Adjustment of the Climatology to Actual Observations (CACAO) post-processing for parcel-level crop monitoring.
Accurate segmentation of multiple phenotypic traits in early-stage soybean plants is essential for automated phenotyping and early-stage breeding analysis. However, the morphological diversity and heterogeneous visual characteristics of key traits, including hypocotyls, flowers, pubescence, and leaves, make unified segmentation challenging under complex backgrounds. To address this problem, this study proposes LASH-SegNet, a lightweight deep learning network for multi-trait segmentation of early-stage soybean plants. The network integrates dynamic snake convolution to model elongated and non-rigid structures and incorporates a SegNeXt-Attention module to enhance multi-scale feature representation and boundary awareness. In addition, the WIoUv3 loss function is adopted to improve localization accuracy and boundary alignment, particularly for slender targets. Experimental results show that LASH-SegNet achieves a precision of 88.82%, recall of 89.78%, and an F1-score of 89.30%, with an mAP50 of 91.24%, while maintaining a compact model size of 5.9 M parameters and 11.3 MB. These results demonstrate that LASH-SegNet provides an accurate and efficient solution for high-throughput multi-trait early-stage soybean plant phenotyping.
Why it matches plant phenotyping methods大豆幼苗の複数形質を自動抽出する画像セグメンテーション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。
abstractthis study proposes LASH-SegNet, a lightweight deep learning network for multi-trait segmentation of early-stage soybean plants.
Accurate plant organ segmentation is essential for high-throughput phenotyping and ideotype selection. However, current methods struggle with plants of complex morphology, particularly small organ categories with sparse point distributions. In addition, severe leaf adhesion in dense canopies often hinders reliable leaf instance segmentation using conventional clustering methods. To address these challenges, we propose a dual-path fusion network (DPFuseNet) for semantic segmentation and a hierarchical multi-scale spectral clustering algorithm (HMSC) for instance segmentation of plant point clouds. DPFuseNet introduces three innovations: a high-frequency information embedding strategy, a dual-path feature extraction module integrating CNN and Transformer branches, and a cross-attention–based dual-granularity feature fusion block. Evaluated on tomato, cabbage, and soybean datasets, DPFuseNet achieved superior performance over state-of-the-art baselines such as Stratified Transformer and Point Transformer v3, reaching average precision, recall, F1-score, and IoU of 96.51%, 96.27%, 96.38%, and 93.32%, respectively. Compared with the current leading single-branch model Point Transformer v3, DPFuseNet improves these metrics by 1.19%, 1.20%, 1.21%, and 2.05%, and by 0.89%, 1.14%, 1.02%, and 1.70% over the dual-branch model PVDST. For instance segmentation, the proposed HMSC algorithm, combined with region growing, achieved mPrec 89.65%, mRec 78.70%, mCov 76.88%, and mWCov 85.11% on multi-stage tomato, cabbage, and soybean datasets, consistently outperforming conventional spectral clustering. Overall, the proposed framework demonstrates robustness and efficiency in both semantic and instance segmentation, offering a novel pathway for advancing plant point cloud analysis and smart agriculture.
Why it matches plant phenotyping methods植物点群から器官の意味・個体分割を行う深層学習およびクラスタリング手法の開発と評価が研究の中心であり、植物表現型解析への直接的な応用を示している。
abstractAccurate plant organ segmentation is essential for high-throughput phenotyping and ideotype selection.
Accurate soybean field phenotyping is increasingly important for breeding. However, traditional measurement methods are labor-intensive and subjective, while UAV-based approaches are challenged by complex backgrounds and densely distributed small targets. This study first develops UAV-ZSAR to transform oblique UAV images into horizontal-view images and reconstruct plant geometry. A lightweight point-based model, Soy-MOPNet, is then proposed for fast and parallel detection of soybean seeds and stem nodes. The model incorporates the proposed SDConv, optimized hierarchical dilated convolution (HDC) principles, and PBOS to enhance adaptive feature fusion, receptive field design, and multi-branch training stability, respectively. Based on the detected keypoints, six phenotypic traits are extracted in parallel, providing comprehensive support for field phenotyping, breeding selection, and precision agricultural management.
Why it matches plant phenotyping methodsUAV画像の幾何再構成、軽量点検出モデル、複数器官からの6形質抽出を開発しており、植物表現型取得・抽出手法が研究の中心である。
abstractThis study first develops UAV-ZSAR to transform oblique UAV images into horizontal-view images and reconstruct plant geometry.
Accurate detection of soybean seedlings using unmanned aerial vehicles (UAVs) in complex field environments is crucial for yield estimation and agricultural planning. However, UAV images present challenges such as small, densely clustered, and partially occluded seedlings. Combined with complex field conditions marked by intricate backgrounds and resolution variations, and limited single-scene training data, these factors collectively cause significant detection model performance degradation. To overcome these limitations, we develop GAS-YOLO, an enhanced YOLOv8-based framework for accurate soybean seedling detection in complex field environments. Firstly, we integrated a Global Attention Mechanism (GAM) into the model’s neck to prioritize contextual features and suppress background noise. Secondly, the SIoU loss was employed with an angle term to mitigate bounding box drift and improve localization accuracy. Furthermore, targeted data augmentation strategies, such as defocus blur simulation and soil color variation, were applied to single-scene data to simulate diverse complex field scenarios and enhance model generalization. The experimental results indicate that GAS-YOLO shows significantly better performance than the baseline YOLOv8 model. Crucially, it shows notable improvements in high-density regions, with estimation accuracy increasing by 30.46% for densities of 80–100 seedlings and by 11.91% for densities exceeding 100 seedlings. GAS-YOLO also exhibits better performance in challenging field environments. On test sets featuring defocused images and yellow soil backgrounds, soybean seedling detection accuracy increases by 47.14%, which shows GAS-YOLO’s robustness in real-world agricultural scenarios. This study establishes GAS-YOLO as a robust and reliable solution for soybean seedling detection in complex field environments, offering advantages for practical agricultural applications.
Why it matches plant phenotyping methodsUAV画像から大豆幼苗を検出・密度推定するモデル開発が中心で、植物個体数という観測可能な作物状態を定量化しているため、植物フェノタイピング手法として含める。
abstractwe develop GAS-YOLO, an enhanced YOLOv8-based framework for accurate soybean seedling detection in complex field environments.
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 15 Sept 2026
SoybeanGrowth chamberRootMorphology / geometry measurementObject detectionRoot system architecture
). Accurate quantification of nodule traits is essential for understanding host-microbe interactions and genetic determinants of nodulation. However, traditional manual or semi-quantitative approaches are labor-intensive, subjective, and unsuitable for large-scale studies. Here, we present a high-throughput phenotyping pipeline based on the YOLO deep learning architecture for the automated detection and extraction of soybean root traits. The pipeline quantifies nodule count, dimensions, and spatial distribution, enabling measurement of 24 distinct nodulation-related traits. Using root images from 21-day-old hydroponically grown soybean plants, the model achieved a precision of 0.94, a recall of 0.95, and an F1 score of 0.94 for nodule detection, maintaining accuracy across count ranges. It processes 50 root images in 37 seconds on a single GPU (45 GB memory), representing a ~227-fold improvement in efficiency compared to manual scoring (~2 h 20 min). As proof of concept, we applied this pipeline in a genome-wide association study (GWAS) using the FarmCPU approach and identified 50 significant SNPs associated with multiple nodulation traits, including novel ones. Several candidate genes linked to these loci suggest potential new regulators of nodulation. This YOLO-based phenotyping framework provides a robust, scalable, and reproducible tool for trait discovery and genetic analysis, advancing research in legume genomics and crop improvement. To promote the adoption of this user-friendly nodulation phenotyping pipeline and to support its further development, we have made all essential resources publicly available at: https://github.com/Salk-Harnessing-Plants-Initiative/soybean-nodule-detection.
Why it matches plant phenotyping methodsYOLOを用いた根粒形質の自動取得パイプラインを開発し、検出精度・処理速度を検証した方法中心の研究である。
abstractHere, we present a high-throughput phenotyping pipeline based on the YOLO deep learning architecture for the automated detection and extraction of soybean root traits.
Introduction Rapid population growth and climate change have intensified the need for sustainable agricultural productivity. Plant leaf diseases significantly impact the crop yield, quality, and food safety, necessitating accurate and automated detection methods. Methods This study proposes a deep learning (DL)-based framework for automated detection and classification of tomato and soybean leaf diseases. The proposed framework is trained and evaluated over a large-scale datasets comprising 16,012 tomato leaf images and 6,410 soybean leaf images. Multiple convolutional neural network (CNN) models, including DenseNet121, MobileNetV2, and InceptionV3, are employed for classification. Object detection is performed using YOLOv12. To enhance interpretability, Gradient-Weighted Class Activation Mapping (Grad-CAM) is integrated. Furthermore, a novel Hybrid Attention-Based Stacking Ensemble Model is developed using ResNet152V2, VGG19, and EfficientNetB0, combined with Convolution Block Attention Module (CBAM) and spatial attention mechanisms. Results The CNN models achieved classification accuracies of 97% for DenseNet121, 98% for MobileNetV2, and 99.94% for InceptionV3. YOLOv12 attained a mean average precision (mAP) of 99.5%. The proposed hybrid ensemble model achieved an accuracy of 99.18%, demonstrating improved feature learning through combined channel and spatial attention. Grad-CAM visualizations confirmed that the model effectively identifies the disease-relevant regions. Discussion The results indicate that the proposed framework has attained a high accuracy, robustness, and interpretability for plant disease detection. The integration of attention mechanisms and explainable AI enhances model reliability and transparency. This framework shows a strong potential for the real-time agricultural monitoring, although further validation across diverse crops and real-world field conditions is required.
Why it matches plant phenotyping methods植物葉の病害状態を画像から自動検出・分類する深層学習フレームワークの開発であり、病害表現型の取得・推定が研究の中心。
abstractThis study proposes a deep learning (DL)-based framework for automated detection and classification of tomato and soybean leaf diseases.
Reproduction assets foundThe paper's leaf-disease classification/detection experiments are built on two public Kaggle image datasets cited by the authors as the study's data sources: a soybean leaf dataset (Patil 2024) and a tomato leaf disease dataset (Rex 2019). No authors' analysis code, trained model checkpoints, or paper-specific phenotypDataset · publicPatil A. ( 2024 ). Soyabean-Latest Dataset (
Kaggle ). Available online at: https://www.kaggle.com/datasets/adityapatil1205/soyabean-latestOpen asset ↗Kagglelines:1523-1646Dataset · publicRex E. ( 2019 ). Plant Disease Dataset (Tomato Leaf Diseases) (
Kaggle ). Available online at: https://www.kaggle.com/datasets/emmarex/plantdisease197Open asset ↗Kagglelines:1523-1646Plant phenotyping relevance match · UnverifiedCrossref · checked 8 Sept 2026
Effective crop monitoring during monsoon growing seasons in Central India faces challenges from persistent cloud cover that limits optical remote sensing during critical agricultural periods. This study presents the first attempt to develop a novel set of SAR-derived phenological metrics organized into five thematic categories for monsoon crop discrimination in smallholder agricultural systems. Five major monsoon crops (cotton, rice, maize, soybean, and urad) were analyzed across five different agroclimatic zones in Central India using Sentinel-1 data for the 2021 growing season. Phenological features were extracted from VV, VH polarizations, and their ratio, including seasonal extrema, threshold crossings, duration measures, curve shape descriptors, and area under the curve. Distinct crop-specific signatures were observed, with cotton showing extended phenology and cereal–legume crops displaying compressed, overlapping growth patterns. VV polarization achieved the highest statistical discrimination for intensity-based metrics, with 75% thresholds (VV_HP75V: F = 1287) providing higher separability than other thresholds by capturing near-peak biomass differences. VH performed best for duration and integration-based metrics, while VH/VV provided limited additional separability across metric types. For area-under-the-curve metrics, AUC25 outperformed AUC50 and AUC75 by capturing cumulative backscatter across the broader growing season while remaining robust to soil- and residue-dominated backscatter variability at sowing and harvest. Multiclass classification achieved 48.3% overall accuracy with systematic cereal–legume confusion, reflecting fundamental phenological convergence among monsoon-aligned crops. Cotton achieved the highest performance (F1: 0.79), with VH polarization dominating feature importance (65% of top 20 features). Binary classification revealed crop-specific discrimination patterns: cotton was best separated using VV intensity metrics, maize using the VH/VV ratio, and rice using timing-based features. Cross-district transferability showed the highest mean overall accuracy for rice (74%) and cotton (72%), while the remaining crops showed lower accuracy due to their phenological similarity. These findings highlight both the potential and limitations of SAR phenological metrics for monsoon crop discrimination, with effective results for structurally distinct crops but persistent cereal–legume confusion, requiring further investigation with multi-sensor approaches.
Why it matches plant phenotyping methodsSAR時系列から作物のフェノロジー指標を抽出・評価し、識別性能や転移性を検証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 the first attempt to develop a novel set of SAR-derived phenological metrics organized into five thematic categories for monsoon crop discrimination in smallholder agricultural systems.
Soybean fatty-acid composition is a key determinant of nutritional quality and industrial value, but conventional gas chromatography is destructive, labor-intensive, and time-consuming. This study combined hyperspectral imaging, which enables rapid and nondestructive acquisition of seed-surface spectral information, with the Tabular Prior-data Fitted Network (TabPFN) to predict the relative proportions of five major soybean fatty acids: palmitic, stearic, oleic, linoleic, and linolenic acids. Mean seed reflectance spectra extracted using three region-of-interest (ROI) strategies were subjected to preprocessing, comparison across representative models and feature-reduction strategies, and SHapley Additive exPlanations (SHAP) analysis to identify wavelength regions associated with fatty-acid variation. TabPFN achieved the best regression performance under partial least squares (PLS) reduction, with an overall R 2 of 0.9750, while all four classification metrics exceeded 0.93 under linear discriminant analysis (LDA). These results demonstrate an accurate, interpretable, and nondestructive framework for rapid prediction of soybean fatty-acid composition and quality evaluation.
Why it matches plant phenotyping methods大豆種子の脂肪酸組成という植物器官形質を、ハイパースペクトル画像と機械学習で非破壊推定する方法の開発・比較・性能評価が中心である。
abstractThis study combined hyperspectral imaging, which enables rapid and nondestructive acquisition of seed-surface spectral information, with the Tabular Prior-data Fitted Network (TabPFN) to predict the relative proportions of five major soybean fatty acids
To improve crop genetics, high-throughput, effective and comprehensive phenotyping is a critical prerequisite. While such tasks were traditionally performed manually, recent advances in multimodal foundation models, especially in vision-language models (VLMs), have enabled more automated and robust phenotypic analysis. However, plant science remains a particularly challenging domain for foundation models because it requires domain-specific knowledge, fine-grained visual interpretation, and complex biological and agronomic reasoning. To address this gap, we develop PlantXpert, an evidence-grounded multimodal reasoning benchmark for soybean and cotton phenotyping. Our benchmark provides a structured and reproducible framework for agronomic adaptation of VLMs, and enables controlled comparison between base models and their domain-adapted counterparts. We constructed a dataset comprising 385 digital images and more than 3,000 benchmark samples spanning key plant science domains including disease, pest control, weed management, and yield. The benchmark can assess diverse capabilities including visual expertise, quantitative reasoning, and multi-step agronomic reasoning. A total of 11 state-of-the-art VLMs were evaluated. The results indicate that task-specific fine-tuning leads to substantial improvement in accuracy, with models such as Qwen3-VL-4B and Qwen3-VL-30B achieving up to 78%. At the same time, gains from model scaling diminish beyond a certain capacity, generalization across soybean and cotton remains uneven, and quantitative as well as biologically grounded reasoning continue to pose substantial challenges. These findings suggest that PlantXpert can serve as a foundation for assessing evidence-grounded agronomic reasoning and for advancing multimodal model development in plant science.
Why it matches plant phenotyping methodsPlantXpertは作物フェノタイピング向けの画像ベンチマークとVLM評価基盤を構築しており、表現型解析手法・データセットの開発が研究の中心である。
abstractwe develop PlantXpert, an evidence-grounded multimodal reasoning benchmark for soybean and cotton phenotyping.
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 15 Sept 2026
Contemporary agrobiotechnology research increasingly relies on automated methods for capturing and interpreting morphophysiological and spectral plant characteristics - a field known as digital phenotyping. This approach aims to identify stable differences between genotypes cultivated under non-identical environmental conditions. We previously introduced StatFaRmer, an open-source tool that we further develop here for comprehensive analysis of temporal phenotypic datasets, with a primary focus on crops such as soybean (Glycine max). The tool implements automated data preprocessing procedures, including synchronization of timestamps across samples and removal of noise artifacts and outliers. These features are particularly relevant for multi-month experiments involving assessments of growth parameters, fluctuations in photosynthetic apparatus area, or other biometric indicators. Support for standardized data formats (XLSX, CSV) ensures compatibility with common phenotyping systems, simplifying cross-platform integration. Thus, the tool can integrate with widely used HTPP platforms (e. g., Traitmill, HyperAIxpert, Plant Accelerator), enabling data from diverse sources to be analyzed within a single pipeline. For soybean experiments, StatFaRmer provides customizable analysis of variance (ANOVA) with visualization of diagnostic parameters (normality of distribution, homogeneity of variances) and evaluation of effect significance between user-defined groups. An example application compares growth parameters across 20 soybean cultivars under controlled stress: the tool automatically aggregated data with uneven measurement frequencies (from 1 hour to 3 days), identified anomalies in hypocotyl elongation dynamics, and computed statistical significance between groups (p < 0.01).The tool has been tested on large-scale datasets (over 2,000 measurements per experiment). StatFaRmer is implemented as a Shiny-based web application, with step-by-step deployment guides for Windows and Linux. All processing stages - from raw data to final plots - are documented to ensure transparency and compliance with research reproducibility standards. Thus, StatFaRmer offers a specialized solution for statistical hypothesis testing in soybean digital phenotyping, reducing data preparation time and minimizing risks of error when handling non-stationary time series.
Why it matches plant phenotyping methods植物デジタルフェノタイピング用の時系列解析ツールを開発・拡張し、前処理、異常値除去、統計解析、再現可能なワークフローを提供しているため、フェノタイピング手法が中心である。
abstractWe previously introduced StatFaRmer, an open-source tool that we further develop here for comprehensive analysis of temporal phenotypic datasets
Abstract The global issue of water scarcity and climate change requires highly efficient and intelligent irrigation systems that are capable of optimizing water consumption with high crop productivity. The paper aims to provide a holistic machine learning framework for crop water stress prediction and efficient irrigation scheduling using multi-parametric agronomic data. The paper analyzes 55,450 soybean data with 13 physiological and biochemical parameters to implement and compare six regression models for predicting the water stress index. After eliminating tautology by removing the direct water content parameter from the prediction model, LightGBM and XGBoost ensemble tree models achieved near-perfect accuracy for predicting crop water stress using regular plant parameters alone, with R² = 1.0 and RMSE = 1.57×10⁻⁸ to 5.04×10⁻⁵. The Random Forest classifier, which was implemented without any direct stress indicators, achieved perfect discrimination between low, moderate, and high stress classes with precision/recall equal to 1.0, and 5-fold cross-validation and noise tests confirmed its robustness. SHAP analysis of the results showed protein percentage (PPE) and seed yield per unit area (SYUA) to be key drivers of water stress, providing valuable insights for precision agriculture. The model for determining irrigation requirements based on crop evapotranspiration and stress level achieved R² = 1.0 with zero error, making it possible to translate trait values directly into irrigation requirements. The framework presented in this paper brings together machine learning and agronomic knowledge to provide real-time data-driven solutions for irrigation systems, which have 30–50% water savings potential while maintaining healthy crops. It lays the ground for the development of AI-assisted irrigation systems that are applicable to different crops and climatic conditions, particularly in water-scarce countries such as Iraq.
Why it matches plant phenotyping methods作物の水ストレス状態を生理・農学データから機械学習で推定し、複数モデルの比較、交差検証、ノイズ試験、解釈分析まで行う計算的フェノタイピング手法が中心である。灌漑最適化への応用を含むが、単なる日常的測定ではない。
abstractThe paper aims to provide a holistic machine learning framework for crop water stress prediction and efficient irrigation scheduling using multi-parametric agronomic data.
Reproduction assets foundThe paper's soybean phenotyping dataset (55,450 records, 13 physiological/biochemical traits) is publicly available on Kaggle; the Data Availability statement points to it, though it ambiguously labels it as the code implementation location. No separate verified code repository is provided.Dataset · publicThe dataset used in this study (Advanced Soybean Agricultural Dataset) is available from the corresponding author upon reasonable request. The code implementation for all analyses is available at: https://www.kaggle.com/datasets/wisam1985/advanced-soybean-agricultural-dataset-2025 .Open asset ↗kaggle · wisam1985/advanced-soybean-agricultural-dataset-2025lines:372-406Plant phenotyping relevance match · UnverifiedOpenAlex · checked 14 Sept 2026
Abstract A genome‐wide association study (GWAS) using digital images was conducted to delineate regions of the genome that govern the leaf flipping quantitative trait in soybean ( Glycine max (L.) Merr). However, converting the digital data to numerical scores for downstream analyses was challenging. We have developed an algorithm that operates in the hue, saturation, and value color space in a structured image processing pipeline that includes preprocessing, binary masking for leaf region isolation, contrast enhancement, grid‐based intensity analysis, and thresholding for detecting folded leaves, a response of soybean to drought. The outputs of this image analysis reached over 90% detection accuracy for images captured under different imaging conditions. GWAS using the processed images identified the same genetic loci underlying drought tolerance as were identified earlier by GWAS of the manually curated dataset from the same photos. This approach provides a robust, scalable, and cost‐effective tool for digital image‐based high‐throughput phenotyping.
Why it matches plant phenotyping methods大豆葉の反転表現型を画像から定量化する画像処理アルゴリズムを開発し、異なる撮像条件で精度検証しているため、植物フェノタイピング手法が中心である。
abstractWe have developed an algorithm that operates in the hue, saturation, and value color space in a structured image processing pipeline that includes preprocessing, binary masking for leaf region isolation, contrast enhancement, grid‐based intensity analysis, and thresholding for detecting folded leaves
Introduction Intercropping regimes enhance the efficiency of land use and ecological sustainability but present serious problems to automated disease analysis since the overlapping canopy and the similarity of symptoms in crop species are visually indistinguishable. Methods This work presents an explainable artificial intelligence (XAI)-based hyperspectral analysis on leaf disease in intercropping systems. The framework combines the spectral-spatial feature generators that utilize transformers including vision transformer (ViT), Swin transformer, pyramid vision transformer (PVT), and detection transformer (DETR) to identify nuanced biochemical and structural changes in crop combinations for maize-soybean and pea-cucumber. In order to reduce spectral redundancy and high dimensionality, an enhanced greedy political optimization (EGPO) algorithm is used as a wrapper-based feature selection strategy. A capsule spatial shift neural network (CSSNet) is used to predict the classification of diseases. Explainable AI methods, such as Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) feature attribution analysis and gradient-weighted class activation mapping (Grad-CAM) visualization of disease-relevant regions, provide model transparency. The DETR + EGPO + CSSNet framework is tested on the conventional feature selection methods. Results and discussion The results or findings on publicly available hyperspectral datasets on intercropping show an average recall of 99.998% with high region consistency (Dice score: 99.997%) of activation maps and expert-marked disease regions. These findings affirm that the proposed framework is highly accurate, stable, and interpretable to identify subtle and overlapping disease in leaves in a complex system of intercropping.
Why it matches plant phenotyping methods植物葉の病害領域・病状をハイパースペクトル画像から推定する解析手法の開発と評価が中心であり、植物病害表現型の取得・抽出に直接関係する。
abstractThis work presents an explainable artificial intelligence (XAI)-based hyperspectral analysis on leaf disease in intercropping systems.
MaizeRiceSoybeanWheatRootMorphology / geometry measurementSegmentationGrowth / time-series analysisGrowth / development / phenologyRoot system architecture
Root phenotyping is crucial for advancing our understanding of plant development and adaptation. However, existing platforms often face challenges in balancing high-throughput capacity with long-term, high-frequency monitoring. To overcome this limitation, we present HTPRootSlides, an integrated root phenotyping platform designed for dynamic and scalable trait analysis. Its design features a circulating zone that accommodates 141 specialized root boxes for high-throughput operation synchronously. Root boxes follow a continuous S-shaped trajectory step by step, facilitating repetitive imaging for high-throughput, time-series data acquisition. To address challenges such as water vapor condensation and fine root entanglement, we developed a dedicated segmentation algorithm, achieving 89.56 % accuracy in root isolation. Combining morphological and skeleton-based feature extraction techniques, the platform ensures comprehensive and efficient phenotypic trait quantification. We validated HTPRootSlides by dynamically monitoring root development in four staple crops (soybean, maize, wheat, and rice) during early-stage germination (<14 d). The results demonstrate the capability of HTPRootSlides for high-frequency, high-precision and large-scale root phenotyping (< 1h with 141 root boxes per run), offering researchers a powerful tool to investigate root dynamics and optimize crop performance through trait selection.
Why it matches plant phenotyping methods根の動態を高スループットで撮像・分割・特徴抽出し、形態・骨格形質を定量するプラットフォームの開発と検証が中心である。
abstractwe present HTPRootSlides, an integrated root phenotyping platform designed for dynamic and scalable trait analysis
Agrophotovoltaic (APV) systems provide a unique opportunity for improving agricultural land-use efficiency by combining solar energy capture via photovoltaic panels with crop production. However, in-depth information on plant growth patterns within the spatially heterogenous microclimate created by the intermittent shading of APVs is largely missing. In the present study, we implement a customized robot-mounted 3D-multispectral imaging system to closely monitor the growth and spectral reflectance patterns of a conventional soybean cultivar “Eiko” (EK) and a chlorophyll-deficient mutant variety MinnGold (MG) under an APV system. Weekly trends in canopy morphometric features revealed significant variations in plant height, 3D leaf area, light penetration, and canopy volume across the APV field depending on the proximity with the overhead solar panels for both EK and MG, with plants receiving adequate rainfall and intermittent shade performing the best. Furthermore, although spectral indices exhibited variations between EK and MG due to intrinsic differences in pigmentation, symptoms of stress could be detected for both genotypes within rain-shaded areas of the APV plot. Hence, the present investigation depicts the potential for complementary usage of robotics and machine vision for high-precision high-throughput crop monitoring under APVs, which would enable better crop management within such non-homogenous cultivation systems.
Why it matches plant phenotyping methodsロボット搭載の3D・マルチスペクトル画像システムを構築・適用し、作物の形態形質とストレス状態を高精度・高スループットに取得しており、表現型取得手法が研究の中心である。
abstractwe implement a customized robot-mounted 3D-multispectral imaging system to closely monitor the growth and spectral reflectance patterns
Agricultural robotics-enabled crop health monitoring faces critical trade-offs: standalone on-device models sacrifice accuracy for real-time responsiveness, while cloud-dependent approaches suffer from high latency and communication overhead. Additionally, data-driven models often lack biophysical plausibility, leading to unreliable predictions for agronomic decision-making under resource constraints. We propose a hybrid LSTM-edge correction architecture that hierarchically integrates lightweight Long Short-Term Memory (LSTM) networks on field robots with physics-informed neural networks (PINNs) at the edge. On-device LSTMs process localized sensor data (soil moisture, spectral reflectance) to generate initial crop stress probability estimates with minimal latency. Edge-based PINNs refine these predictions by embedding biophysical dynamics—modeled via coupled partial differential equations (PDEs) governing the soil-plant-atmosphere continuum (SPAC)—to ensure agronomic validity, mitigate sensor noise, and account for spatial variability. The framework is deployed on NVIDIA Jetson Nano (local inference) and AMD EPYC servers (edge processing), seamlessly integrating with existing farming infrastructures to replace rule-based thresholds with adaptive, physics-grounded control commands. A Fourier Neural Operator (FNO) optimizes the edge PINN’s computational efficiency for high-dimensional PDE solving. Experimental evaluations on two real-world datasets (soybean and citrus) demonstrate that the hybrid approach improves prediction accuracy by 18% compared to standalone LSTMs (F1-score: 0.89±0.02 for soybean, 0.83±0.03 for citrus) while maintaining real-time performance (end-to-end latency: 210 ms, energy consumption: 5.1 J/prediction). Field deployment on a 50-hectare soybean farm yields tangible agronomic benefits: 22% reduction in irrigation water usage, 18% fewer pesticide applications, and 95% system uptime under field conditions. The framework exhibits robust performance against sensor noise (≥80% accuracy at 30% noise-to-signal ratio) and outperforms cloud-based PINNs (72.8% lower energy consumption) and threshold-based methods (28–33% higher F1-score). This work advances distributed agricultural robotics by bridging data-driven machine learning and domain-specific physics, delivering a scalable, interpretable, and resource-efficient solution for precision agriculture. The hierarchical prediction-correction pipeline balances real-time responsiveness with biological plausibility, making it suitable for resource-constrained field robots. By integrating legacy sensors and adaptive actuation control, the architecture offers a practical pathway to upgrade existing farming systems, enabling data-informed interventions while reducing environmental impact.
Why it matches plant phenotyping methods作物ストレス状態を推定するLSTM・PINN・FNO統合パイプラインを開発し、実データで精度・遅延・ノイズ耐性を評価しており、植物状態の取得・推定手法が中心である。
abstractWe propose a hybrid LSTM-edge correction architecture that hierarchically integrates lightweight Long Short-Term Memory (LSTM) networks on field robots with physics-informed neural networks (PINNs) at the edge.
Abstract Traditional soybean seed viability assessment methods are destructive, time-intensive, and incapable of rapid, non-destructive single-seed grading. To overcome these limitations, this study proposes a novel approach integrating Transmission Hyperspectral Imaging (THSI) with ensemble learning for rapid, non-destructive evaluation. Naturally aged soybean seeds were analyzed using full-spectrum (400–2500 nm) transmittance data to capture deep physiological information. Multiple datasets were constructed by comparing preprocessing techniques—including Smoothing, First Derivative (FD), Hilbert Transform (HT), Savitzky–Golay, Multiplicative Scatter Correction (MSC), and Standard Normal Variate (SNV)—with dimensionality reduction algorithms such as Principal Component Analysis (PCA), Successive Projections Algorithm (SPA), and Competitive Adaptive Reweighted Sampling (CARS). The proposed Stacking ensemble model integrates predictions from Random Forest (RF), Support Vector Machine (SVM), Naïve Bayes (NB), k-Nearest Neighbors (k-NN), and Logistic Regression (LR), achieving 98.33% accuracy and an F1-score of 98.12% on the CARS-HT dataset—significantly outperforming individual classifiers in both accuracy and robustness. Furthermore, the model identified 17 critical wavelengths that reveal physiological mechanisms ranging from chlorophyll degradation and antioxidant balance in the visible spectrum to water migration and lipid peroxidation in the infrared region. The method's high precision and reliability were validated, providing robust technical support for intelligent and precise soybean seed quality management.
Why it matches plant phenotyping methods大豆種子の生存性という植物状態を、透過ハイパースペクトル画像とアンサンブル学習で非破壊推定する手法を開発・検証しており、表現型取得・抽出が研究の中心である。
abstractthis study proposes a novel approach integrating Transmission Hyperspectral Imaging (THSI) with ensemble learning for rapid, non-destructive evaluation
Why it matches plant phenotyping methods発芽油種子の油含量・分布という植物器官の状態を、LF-NMR/MRIで非破壊測定する手法を開発・検証しており、表現型取得法が研究の中心である。
abstractThis study evaluated the feasibility of using low-field nuclear magnetic resonance (LF-NMR) coupled with magnetic resonance imaging (MRI) as a non-destructive approach for monitoring oil changes in germinating oilseeds.
MaizePotatoSoybeanLeafClassificationObject detectionCalibration / preprocessingSegmentationStress / disease detectionVisualization / data management
Background: Plant diseases significantly reduce global crop productivity, creating an urgent demand for intelligent, automated diagnostic systems in agriculture. Traditional manual inspection is labor-intensive, subjective and often ineffective in detecting early or latent symptoms. This study presents a multi-class classification and severity estimation framework for ten plant disease categories: Maize brown spot, maize rust, maize healthy, potato early blight (Alternaria solani), potato late blight (Phytophthora infestans), potato healthy, soybean mosaic virus (SMV), soybean pod mottle virus (SPMV), soybean sudden death syndrome (SDS/SBS) and soybean healthy. The objective is to develop a robust hybrid deep learning model capable of accurate early detection and quantitative severity assessment to support precision agriculture. Methods: A hybrid architecture combining convolutional neural networks (CNN) with LSTM and BiLSTM networks was implemented. The preprocessing pipeline included leaf segmentation, binary masking, defect localization and edge detection to enhance lesion visibility. CNN layers extracted spatial and textural features, while recurrent layers modeled contextual dependencies within feature representations. Performance was evaluated using Precision, Recall, F1-score, defect percentage estimation, convergence analysis and t-SNE visualization. Result: Results demonstrated stable convergence with decreasing loss (0.8-1.2) and improved feature clustering. Defect severity ranged from 0.00% (Soybean healthy) to 87.93% (Maize brown spot). The framework enables early detection (0.29-5% infection), reduces yield loss, minimizes chemical overuse and promotes sustainable smart agriculture systems.
Why it matches plant phenotyping methodsCNN-LSTM/BiLSTMによる葉画像からの病害検出と病徴重症度推定手法の開発が中心であり、植物状態を直接推定している。
abstractThis study presents a multi-class classification and severity estimation framework for ten plant disease categories
Seed coat pigmentation in soybean is controlled by complex genetic mechanisms involving structural and regulatory genes in the flavonoid biosynthetic pathway. Although brown seed coats are often associated with epicatechin (EC) accumulation, visual classification alone cannot reliably predict EC content. To quantitatively characterize seed coat coloration and its relationship with EC accumulation, we evaluated multivariate colorimetric traits (L*, a*, and b* values in the CIELAB color space) in 235 recombinant inbred lines (RILs) derived from Jinpung (yellow seed coat) and IT109098 (greenish-brown seed coat). Principal component analysis (PCA) of L*, a*, and b* values revealed that genotypes with detectable EC were confined to specific regions of the multivariate color space, indicating that EC accumulation is associated with coordinated color balance rather than overall pigmentation intensity. RILs with high EC content showed significantly lower L* (32.76 ± 2.49) and b* (13.18 ± 2.66) values and higher a* values (5.47 ± 1.31) than those with low EC content. Quantitative trait loci (QTL) mapping identified thirteen loci associated with L*, a*, b*, and principal component scores across chromosomes 01, 05, 06, 08, and 19 A major locus on chromosome 08 near the classical I locus explained a large proportion of phenotypic variance in pigmentation traits. In addition, loci on chromosomes 06 and 19 were associated with integrated color components, suggesting quantitative modulation of EC accumulation. Candidate genes within these regions included flavonoid 3'-hydroxylase and transcription factors such as MYB117 , MYB60 , and TCP5 , supported by sequence variation and differential expression analyses. These findings demonstrate that multivariate colorimetric traits provide a useful phenotyping framework for dissecting seed coat pigmentation and EC accumulation and for pre-selecting high-EC soybean lines. Supplementary information The online version contains supplementary material available at 10.1007/s11032-026-01655-8.
Why it matches plant phenotyping methods大豆種皮の色をCIELAB色値で定量化し、多変量解析による表現型フレームワークとして遺伝子型・EC蓄積との関連を評価しており、色表現型の取得と解析が研究の中心です。
abstractTo quantitatively characterize seed coat coloration and its relationship with EC accumulation, we evaluated multivariate colorimetric traits (L*, a*, and b* values in the CIELAB color space)
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 5 Sept 2026
Soybean ( Glycine max ) root nodules, formed through symbiosis with nitrogen-fixing rhizobia, are essential for biological nitrogen fixation. While quantifying key nodulation traits, nodule number and weight, is critical for assessing symbiotic efficiency and yield potential, current methods are destructive and labor-intensive, unsuitable for longitudinal monitoring and high-throughput phenotyping. Here, we established hyperspectral leaf reflectance as a non-destructive, high-resolution tool capable of monitoring root nodule development. Using Partial Least Squares Regression models, we connected spectral data with nodule metrics from 528 unique soybean plants across 18 genotypes, inoculated with different rhizobium strains, and under different abiotic stresses. These models achieved high accuracy for predicting nodule number (R 2 = 0.75, nRMSE = 6.02%) and moderate accuracy for nodule weight (R 2 = 0.53, nRMSE = 12.38%). Crucially, spectral analyses revealed distinct hyperspectral signatures sensitive to nodule traits. While different rhizobium strains induced comparable changes in both nodule traits, and therefore produced highly overlapped spectral domains, diagnostically distinct spectral patterns were generated under drought versus salt stress, with the former suppressing nodulation more significantly than the latter. Furthermore, we demonstrated the effectiveness of our models for real-time in-situ monitoring of nodule development for individual plants. Spectral-nodule trait covariation analyses further revealed leaf signatures correlated with nodule traits primarily through systemic physiological coupling governed by carbon-nitrogen exchange dynamics and plant water status. This study showcased hyperspectral sensing as a transformative methodology, enabling the unprecedented non-destructive quantification of nodulation dynamics, revealing novel physiological insights into plant-microbe-environment interactions, facilitating breeding and management strategies for sustainable soybean production.
Why it matches plant phenotyping methods葉のハイパースペクトル反射を用いて根粒数・重量を非破壊推定するセンシング手法を開発・評価しており、植物表現型取得が研究の中心です。
abstractcurrent methods are destructive and labor-intensive, unsuitable for longitudinal monitoring and high-throughput phenotyping.
Soybeans have become one of the most significant oilseed and food crops worldwide. However, soybean crops are susceptible to numerous factors. Damage due to pests, illnesses, and other factors exceeds 20 per cent of the world's manufacturing. The usage of Unmanned Aerial Vehicles (UAVs) in crop fields was found to be a significant tool for identifying disease patches, enabling professionals and agriculturalists to make better decisions. Furthermore, in this context, deep learning (DL) has made important developments in Artificial Intelligence (AI). As a result, several studies have employed DL to solve a wide range of diverse problems. In the agricultural sector, DL has gained significant interest in improving crop productivity. This study introduces an Unmanned Aerial Vehicle-Based Soybean Crop Health Monitoring Using Advanced Deep Learning Architectures (UAVSCHM-DLA) model. The aim is to present an intelligent system that is capable of monitoring and assessing soybean crop health using integrated UAV and leaf images. Initially, Histogram Equalisation (HE) and Bilateral Filtering (BF) methods are applied to perform image pre-processing. For effective feature extraction, a vision transformer with the Interactive Mask Self-Attention (IMViT) method is employed. Finally, multiple neural networks with an attention mechanism (MNet-Attn) method are implemented for classification. The comparison of the UAVSCHM-DLA technique illustrated superior accuracies of 98.20% and 97.01% on the leaf and UAV datasets, respectively.
Why it matches plant phenotyping methodsUAV・葉画像からダイズの健康状態／病害を分類する画像解析システムを提案し、前処理・ViT特徴抽出・ニューラルネットワーク分類を評価しているため、植物状態の取得・推定手法が中心である。
abstractThis study introduces an Unmanned Aerial Vehicle-Based Soybean Crop Health Monitoring Using Advanced Deep Learning Architectures (UAVSCHM-DLA) model.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 5 Sept 2026
max L. Merr.) pubescence color is a trait commonly recorded by breeding programs. In previous research using high-throughput phenotyping (HTP), researchers could separate gray pubescence from light tawny and tawny pubescence, but could not separate light tawny from tawny. Using the Random Forest algorithm and time series of aerial RGB (red, green, blue) and multispectral images, this study aimed to classify pubescence color by testing models in data subsets from experiments grown over three years. By incorporating the pubescence color of the parental lines and training the model with a time series of images (four drone flights before or at maturity), a higher overall accuracy was achieved compared to a single flight at maturity. The red/blue index was the most successful feature for discriminating pubescence color, and the blue normalized difference vegetation index (NDVI) and green NDVI were also helpful, mainly in discriminating gray from light tawny pubescence. The overall accuracy was 86.55% in the best scenario (Kappa = 0.7976), and the sensitivity for gray, light tawny, and tawny pubescence were 0.893, 0.788, and 0.915, respectively. When models were tested in an independent environment, they achieved a lower overall accuracy of 65.86%, but still demonstrated fair to good model reliability (Kappa = 0.4874). Applying an HTP pipeline, as used in this study, would help breeding programs save time classifying pubescence color. Since pod color interferes with this trait in the background, genotyping a proportion of the plant rows for both traits and phenotyping pod color could improve the results.
Why it matches plant phenotyping methods航空画像とRandom Forestを用いてダイズの毛茸色という植物形質を分類し、時系列画像、特徴量、独立環境での精度を検証したHTP手法研究であり、表現型取得・抽出法が中心である。
abstractUsing the Random Forest algorithm and time series of aerial RGB (red, green, blue) and multispectral images, this study aimed to classify pubescence color by testing models in data subsets from experiments grown over three years.
Traditional methods for identifying salt tolerance levels in soybean varieties are often cumbersome, time-consuming, and labor-intensive. These challenges are further exacerbated by the limited utility of chlorophyll fluorescence imaging phenotype data, which are insufficiently diverse and difficult to analyze. Additionally, the corresponding parameter text data have not been fully explored and utilized. In this study, salt stress experiments were conducted on 178 soybean varieties, and a multimodal dataset comprising chlorophyll fluorescence images and corresponding textual data was constructed using a chlorophyll fluorescence imaging instrument. A novel gated mechanism network for learnable image-text interaction (Mm-VitnNet) is proposed, which enables global cross-modal interaction between image and text data. The model introduces a gated mechanism to dynamically regulate the fusion intensity of cross-modal information and incorporates two learnable tokens that focus on feature learning for each individual modality. This approach effectively mitigates interference between modalities while preserving modality-specific features, thereby enhancing model performance. The proposed model demonstrates an accuracy rate of 98.97%, significantly outperforming typical models: it improves by 1.09 and 2.33 percentage points compared to CNN-based models such as EfficientNetV2-s (97.88%) and MobileNetV2 (96.64%), respectively, and by 3.21 and 2.60 percentage points compared to Transformer-based Swin Transformer_tiny (95.76%) and hybrid models like MobileViT_S (96.37%), respectively. The model has 10.22M parameters and a computational cost (FLOPs) of 1.84G, which is significantly lower than models like VGG and ResNet50, and only slightly higher than some lightweight CNNs, achieving an effective balance between accuracy and efficiency. The improved model demonstrates notable performance in identifying samples with varying salt tolerance levels, even under limited computational resources, ensuring reliable classification performance. Moreover, this multimodal non-destructive identification method based on chlorophyll fluorescence technology offers an efficient and feasible approach for assessing the salt tolerance levels of soybeans, while also advancing agricultural phenotyping towards greater precision and intelligence.
Why it matches plant phenotyping methodsダイズの塩耐性という植物状態をクロロフィル蛍光画像から推定するマルチモーダル画像解析手法を開発・評価しており、表現型取得と分類モデルが研究の中心である。
abstracta multimodal dataset comprising chlorophyll fluorescence images and corresponding textual data was constructed using a chlorophyll fluorescence imaging instrument.
SoybeanRootSeed / grainMorphology / geometry measurementObject detectionGrowth / development / phenologyRoot system architectureStress response / tolerance
Introduction Nanoparticle-induced treatments can promote seed germination and improve germination potential under environmental stresses such as drought and salinity. This study aimed to investigate the effects of Zinc oxide nanoparticles (ZnONPs) on soybean seed germination and to develop a precise evaluation method. Methods We developed a full-time sequence crop growth vitality monitoring system. Using germination rate and root length as primary evaluation indicators, we conducted full-time sequence germination vitality monitoring experiments on soybean seeds treated with ZnONPs. A dataset was constructed from images documenting embryonic root growth. The developed detection model was used to evaluate image detection accuracy during germination. Germination index and embryonic root length were also calculated. Further tests were performed on seeds exposed to 600 mg/L ZnONPs dispersion, followed by treatment with different concentrations of NaCl and PEG6000 solutions. Results At a concentration of 600 mg/L ZnONPs dispersion, soybean seeds showed the highest germination rate (an increase of 28%) and the longest radicle length (an increase of 42%). Compared with deionized water, the 600 mg/L ZnONPs dispersion accelerated initial germination time, increased germination rate, and enhanced radicle length under low-concentration stress. Discussion The results indicate that, at certain concentrations, ZnONPs dispersion positively influences soybean seed germination under varying salinity and drought conditions. We examined morphological and physiological changes in ZnONPs-treated seeds under stress, establishing a preliminary foundation for evaluating crop and variety vitality. These findings provide new insights that may contribute to improving soybean germination under simulated stress conditions, serving as a preliminary theoretical reference for potential applications in arid and saline environments.
Why it matches plant phenotyping methods発芽中の画像から発芽率・幼根長を抽出する連続モニタリングシステムと検出モデルを開発し、精度評価とデータセット構築を行っており、表現型取得手法が中心である。
Abstract Soybean [ Glycine max (L.) Merr.] varieties are categorized into different relative maturity groups (MGs) that correspond to the approximate region that the variety is best adapted. Maturity is an important trait that growers consider when deciding which varieties to plant and for breeders as a covariate to compare genotypes. Accurate phenotyping of maturity is an important task during line development but is labor‐intensive. High‐throughput phenotyping (HTP) of soybean maturity using unmanned aerial systems can reduce the labor and error associated with manual maturity notes. An HTP program for maturity will provide higher quality maturity data that will improve breeders’ ability to evaluate the performance of breeding lines on a large scale. The objective of this study was to develop an intuitive, accessible, and precise HTP program to determine the maturity of soybean varieties in the field that can be deployed in soybean breeding programs. In this study, “Matti,” a QGIS plugin, was developed to track the average green leaf index (GLI) of soybean research plots during the senescence period. Piecewise and local polynomial regression models monitor the senescence curve and provide maturity estimates when the GLI values are near or below a user‐specified threshold. This algorithm resulted in moderate to high correlations ( r = 0.52–0.97) between the ground truth and estimated maturity of soybean lines in both early and late maturity MGs. Similar correlations ( r = 0.43–0.72) were found for early generation materials. Results indicate that Matti can be easily implemented by soybean breeding programs to provide timely estimates of relative maturity.
Why it matches plant phenotyping methodsUAV画像から大豆の成熟度を推定するHTPプログラムとQGISプラグインを開発し、地上真値との相関で検証しており、植物表現型取得・推定手法が中心である。
abstractThe objective of this study was to develop an intuitive, accessible, and precise HTP program to determine the maturity of soybean varieties in the field that can be deployed in soybean breeding programs.
3D crop phenotyping technology provides critical support for screening morphology-related plant genes and identification of germplasm resource. Organ segmentation or recognition is the first key step in 3D crop phenotyping, where inductive deep learning currently dominates as the mainstream methodology. However, the high requirement for data annotation in inductive learning paradigm has transformed the manual data labeling into a labor-intensive task, thereby in turn restricting the progress of inductive learning. This problem has inspired us to leverage Graph Neural Networks (GNNs) as the transductive learning tool to directly segment organs on sparsely annotated crop point clouds. We propose a Dual-branch Graph Convolutional Network (DBGCN) that only requires sparse labels to perform organ instance inference directly on plant point clouds that have featureless point features. Different from existing graph-based networks, DBGCN not only carries out the static-feature-space graph convolutions that are good at mining and aggregating on local information on the point cloud, but also incorporates dynamic graph convolutions that captures the potential changes of the graph manifold in deep feature space. Extensive experiments prove that the fusion of two types of graph feature convolutions brings a high node (point) classification accuracy, outperforming mainstream GNNs and even several popular inductive deep architectures. On the PlantNet sub-dataset, DBGCN achieves an mAcc (mean accuracy of node classification) of 93.00% under 1.95% manual annotation ratio. On the Soybean-MVS sub-dataset, DBGCN achieves an mAcc of 91.05% under 4.88% manual annotation ratio. Furthermore, our DBGCN not only works well on crop 3D data but can also serve other applications such as the segmentation of point cloud data for large-scale street view. Our dataset and code can be found at https://github.com/chinazhouzhaoyi/DBGCN/tree/master/.
Why it matches plant phenotyping methods3D植物点群から器官を分割・推論する深層学習手法を開発し、植物フェノタイピングデータ上で精度検証しているため、表現型取得・抽出法が中心である。
abstractWe propose a Dual-branch Graph Convolutional Network (DBGCN) that only requires sparse labels to perform organ instance inference directly on plant point clouds
Reproduction assets foundThe authors explicitly state that their dataset (plant point clouds) and DBGCN code are publicly available on GitHub.Code · publicOur data and code are available at: https://github.com/chinazhouzhaoyi/DBGCN/tree/master/.Open asset ↗chinazhouzhaoyi/DBGCNhtml-lines:414-455Dataset · publicOur dataset and code can be found at https://github.com/chinazhouzhaoyi/DBGCN/tree/master/Open asset ↗chinazhouzhaoyi/DBGCNhtml-lines:88-94Plant phenotyping relevance match · UnverifiedEurope PMC · checked 5 Sept 2026
As the key structure connecting the vegetative and reproductive organs of soybean plants, the main stem plays a crucial role, and its morphological parameters serve as core phenotypic indicators for evaluating plant growth, lodging resistance, and yield potential. At the mature stage, the main stem exhibits high similarity to pods in color and texture, along with complex curvature and severe occlusion by pods and leaves, making accurate and continuous extraction challenging for conventional segmentation methods. To address this, this study proposes RAM-UNet, a high-precision semantic segmentation model based on an improved U-Net architecture. The model adopts ResNet50 as the backbone and replaces standard convolutions with deformable convolutions to capture curved stem morphology and improve feature extraction for low-contrast edges. In the encoder, the Convolutional Block Attention Module (CBAM) is combined with an improved atrous spatial pyramid pooling (ASPP) module (C-ASPP) with four dilation rates, enhancing multi-scale feature representation compared to the original three-rate design. A multi-scale attention aggregation (MSAA) module in the decoder improves continuity and integrity of stem boundaries. During training, a composite loss function combining Dice loss and cross-entropy loss is employed to mitigate foreground pixel sparsity. Experimental results on a self-constructed dataset show that RAM-UNet achieves a mean Intersection over Union (mIoU) of 90.58%, with Recall and Precision reaching 94.99% and 94.58%, respectively. Compared with U-Net, DeepLabv3+, PSPNet, and SegNet, RAM-UNet improves mIoU by 6.41%, 10.51%, 22.41%, and 17.37%, respectively. Automatically measured stem lengths show high agreement with manual measurements (R² = 0.9746), validating practical applicability. RAM-UNet also generalizes well on the public PASCAL VOC 2012 dataset, achieving an mIoU of 73.14%. The results indicate that the proposed model enables high-precision and continuous segmentation of main stems in mature soybean plants, providing an effective technical solution for automated and non-destructive measurement of crop phenotypic parameters.
Why it matches plant phenotyping methods大豆主茎のセグメンテーションと長さ測定を目的とする画像解析手法を開発し、手動測定との一致性で検証しており、植物表現型取得が中心である。
abstractthis study proposes RAM-UNet, a high-precision semantic segmentation model based on an improved U-Net architecture.
Intermediate omics traits, which mediate the effects of genetic variation on phenotypic traits, are increasingly recognized as valuable components of genetic evaluation. In particular, rhizosphere microbiota play a crucial role in plant health and productivity; however, their complex interactions with host genetics remain challenging to model. Although two-step modeling frameworks have been proposed to integrate intermediate omics traits into phenotype prediction, existing approaches do not incorporate nonlinear relationships between different omics layers. To address this, we have proposed a two-step phenotype prediction framework that integrates genomic, rhizosphere microbiome, and metabolome (meta-metabolome) data, while explicitly capturing omics-omics nonlinearities. The first step is to predict meta-metabolome traits from genetic and microbial features, thus effectively isolating them from the environmental noise. In this process, intermediate "proxy" omics traits are generated as general biological information to provide robust models. The second step utilizes this "proxy" to enhance the accuracy of the phenotype prediction. We compared a linear mixed model (Best Linear Unbiased Prediction, BLUP) and a nonlinear model (Random Forest, RF) at each step, as demonstrated through simulations and empirical analysis of a multi-omics soybean dataset in which nonlinear modeling captures intricate omics interactions. Notably, our approach enables phenotype prediction without requiring the original meta-metabolome data used in model training, thereby reducing reliance on costly omics measurements. This framework integrates intermediate omics traits into genomic prediction to improve prediction accuracy and provide solutions for deeper insights into plant-microbiome interactions.
Why it matches plant phenotyping methods植物の表現型予測を目的とする非線形マルチオミクス計算フレームワークが研究の中心であり、単なるオミクス測定や生物学的実験ではない。
abstractwe have proposed a two-step phenotype prediction framework that integrates genomic, rhizosphere microbiome, and metabolome (meta-metabolome) data, while explicitly capturing omics-omics nonlinearities.
Reproduction assets foundThe paper's analysis code is publicly available on GitHub, and the metabolome data are publicly available via the RIKEN DropMet website (IDs DM0071, DM0072). Phenotype and other multi-omics data are only available from the corresponding author upon request.Code · publicAll source codes are available from the repository in GitHub: https://github.com/Yoska393/Twostep .Open asset ↗Yoska393/Twosteplines:306-317Plant phenotyping relevance match · UnverifiedCrossref · checked 5 Sept 2026
Fractional vegetation cover of crops (CropFVC) is a critical indicator for remote sensing-based crop monitoring. However, existing inversion models are largely developed for general vegetation types, limiting their effectiveness for crop-specific applications. Here, we developed a gap-fraction-refined hybrid CropFVC model that integrates crop-specific PROSAIL calibration, an ALA (averages of leaf angle) -based dynamic projection function, and a Random Forest model. The model was validated with 43343 CropFVC samples of four major crops (winter wheat, rice, maize, and soybean) across China during March to August 2024, spanning key phenological stages, and further compared against SNAP (10 m) and GEOV3 (300 m) products. Results showed that (1) the proposed model achieved stable performance across diverse canopy structures, with average RMSE
Why it matches plant phenotyping methods作物の葉面積被覆率という明示的な植物キャノピー形質を推定するハイブリッドモデルを開発し、多数のサンプルと既存プロダクトで検証しており、測定・推定手法が研究の中心である。
abstractHere, we developed a gap-fraction-refined hybrid CropFVC model that integrates crop-specific PROSAIL calibration, an ALA (averages of leaf angle) -based dynamic projection function, and a Random Forest model.
MaizeRiceSoybeanField / plotMultimodalLiDAR / point cloudRGB / grayscaleMultispectral / hyperspectralThermalWhole plant / canopy / plot / field
Plant phenotyping is essential for elucidating genotype–environment interactions, yet conventional methods remain labor-intensive and low-throughput. TraitDiscover transcends these constraints by uniting multimodal sensing with tightly coupled hardware-software orchestration in a single, end-to-end phenotyping platform. Aligned with the ”Plant Phenotyping Trinity” framework, the system comprises a millimetre-accurate triaxial automation unit, a modular sensor array–RGB imaging, three-dimension laser scanner or LiDAR (3D), infrad (IR) thermal imaging, hyperspectral imaging (HSI), and photosynthesis (PS) imaging–and the dedicated software TraitNavigator suite into one cohesive system. A unified spatiotemporal synchronization mechanism enables robust time-series analysis and fusion of multisource phenotypic data across the entire crop growth period, while the DepthCropSeg algorithm and a night-time imaging module enhance trait extraction under complex conditions, providing G × E × P-ready, multimodal phenotypic datasets. Validation across soybean, maize, and rice trials demonstrated high sensitivity—detecting drought stress four days before visible symptoms, identifying glyphosate injury 24 hours ahead of manual scoring, and quantifying local adaption patterns across ecological gradients. While challenges remain in scaling to complex open-field conditions, TraitDiscover offers a scalable, data-driven approach to accelerate stress phenotyping and breeding decisions and is readily poised for deeper integration with AI to advance sustainable agriculture.
Why it matches plant phenotyping methodsマルチモーダルセンシング、画像解析、同期機構、形質抽出アルゴリズムを統合した植物フェノタイピング基盤の開発と検証が中心であり、ストレス検出や形質定量も実証している。
abstractTraitDiscover transcends these constraints by uniting multimodal sensing with tightly coupled hardware-software orchestration in a single, end-to-end phenotyping platform.
The strip intercropping of soybean and maize, characterized by planting the two crops alternately in adjacent rows, has been widely promoted in several regions of China due to its potential to enhance resource utilization efficiency and overall yield. Accurate detection of crop rows and missing seedlings is essential for enabling precision field operations such as variable fertilization and targeted spraying. Monocular vision has emerged as a core sensing modality owing to its low cost and high resolution. However, the significant differences in row and plant spacing between maize and soybean, coupled with complex field conditions such as weed interference and uneven emergence, severely limit the effectiveness of traditional image processing techniques based on thresholding and geometric fitting. These methods struggle to accommodate the morphological variability of multiple crops, resulting in poor row detection precision and unreliable identification of missing seedlings. In recent years, deep learning has shown strong performance in crop row detection and object recognition tasks, particularly through multi-task networks that integrate segmentation and localization-related features. Nevertheless, most existing studies focus on single-crop scenarios and often neglect the integration of agronomic knowledge, thereby limiting their robustness and interpretability in real-world field environments. To address these issues, this study proposes a Multi-Task Geometric Regression Network for row extraction and missing seedling detection in maize-soybean intercropping systems, built upon an improved U-Net++ architecture and guided by agronomic priors. The proposed method simultaneously performs crop segmentation, row direction prediction, and generation of a missing seedling heatmap. The geometric features output by the network are subsequently processed using agronomic prior-informed post-processing and geometric fitting to finally achieve row extraction and missing seedling localization. Agronomic constraints, such as row spacing regularity, are embedded in the loss function as prior-informed regularization terms, which further enhance detection accuracy and robustness in intercropped fields. Experimental results demonstrate that the semantic segmentation achieves an average Intersection over Union (IoU) of 0.82, an F1-score of 0.86, and a pixel accuracy of 0.91. Row centerline detection attains an F1-score of 0.86 and a mean offset (MO) of 3.9 pixels. For missing seedling detection, the crop classification accuracy reaches 0.91, the average localization error (ALE) is only 2.5 pixels, and the composite detection score (CD-F1) is 0.89. Compared with single-task methods without agronomic priors, the proposed multi-task framework exhibits significant improvements in both stability and accuracy for row detection and missing seedling localization in intercropping scenarios. These results provide practical guidance for deploying intelligent visual systems in precision agriculture and intercropping management.
Why it matches plant phenotyping methods作物画像から畝構造と欠株状態を抽出するマルチタスク手法を開発し、精度評価も行っており、植物状態の取得・推定が研究の中心である。
abstractthis study proposes a Multi-Task Geometric Regression Network for row extraction and missing seedling detection in maize-soybean intercropping systems
Nodule formation and their involvement in biological nitrogen fixation are critical features of leguminous plants, with phenotypic characteristics closely linked to plant growth and nitrogen fixation efficiency. However, the phenotypic analysis of root nodules remains technically challenging due to their small size, weak texture, dense clustering, and occlusion. To address these challenges, this study constructed a scanner-based imaging platform and optimized data acquisition conditions for high-resolution, high-consistency root nodule images under field conditions. In addition, A hybrid small-object detection method, SCO-YOLOv8s, was proposed, integrating Swin Transformer and CBAM attention mechanisms into the YOLOv8s framework to enhance global and local feature representation. Furthermore, an Otsu segmentation-based post-processing module was incorporated to validate and refine detection results based on geometric features, boundary sharpness, and image entropy, effectively reducing false positives and enhancing robustness in complex scenes. Using this integrated approach, over 3375 nodules were identified from a single plant sample in under 1 min, with extracted phenotypic features such as diameter, color, and texture. A total of 10,879 high-quality annotated images were collected from 39 peanut varieties across 14 provinces and 31 soybean varieties across 12 provinces in China, addressing the current lack of large-scale datasets for legume root nodules. The SCO-YOLOv8s model achieved a precision of 97.29 %, a mAP of 98.23 %, and an overall identification accuracy of 95.83 %. This integrated approach provides a practical and scalable solution for high-throughput nodule phenotyping, and may contribute to a deeper understanding of nitrogen fixation mechanisms.
Why it matches plant phenotyping methods根粒の画像取得・検出・セグメンテーション・形質抽出を統合した高スループット表現型解析手法を開発し、精度評価と大規模データセット構築も行っているため、方法が研究の中心である。
abstractthis study constructed a scanner-based imaging platform and optimized data acquisition conditions for high-resolution, high-consistency root nodule images under field conditions.
Accurate in-season prediction of seed yield and seed composition traits such as oil and protein are useful for gaining accuracy and efficiency in soybean breeding. These predictions can also inform farmers, enabling them to improve their field management practices, and guide their market decisions. We report a Transformer-based deep learning framework built on 30 years of multi-environment performance data from the Northern and Southern Uniform Soybean Tests (UST) across North America. Unlike earlier studies on seed yield, oil and protein prediction that focus on limited years, regions, single modalities, we utilized a comprehensive dataset that includes weather, genotype, and management factors, ensuring a more holistic approach to soybean yield, oil, and protein prediction. Our model integrates multivariate time-series weather data with genotypic relationship information, maturity group, and geographic location, to predict variety performance in diverse environments. Our model captures complex temporal patterns associated with trait variability; showing high predictive accuracy (R2) of 77.6 ± 0.2%, 63.9 ± 4.7%, and 79.3 ± 2.3% for seed yield, oil, and protein, respectively. Additionally, for seed yield, we also evaluated multiple interpretability methods to assess feature importance for predictor variables and critical growing timepoints, and solar radiation and temperature were noted as the key predictors. Overall, these results demonstrate the usefulness of a Transformer-based model in trait predictions, and the utility of large cooperative datasets from breeding programs.
Why it matches plant phenotyping methodsTransformerによるダイズ収量・油・タンパク質形質の予測フレームワークが研究の中心であり、植物形質の計算的推定手法を開発・評価している。
abstractWe report a Transformer-based deep learning framework built on 30 years of multi-environment performance data
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
Abstract Soybean growth is determined by the interaction of genetic, environmental, and management factors. In the context of future climate and climate extremes, understanding genotype by environment interaction (GxE) will be crucial for selecting resilient breeding lines and optimizing management practices to minimize stress. This requires an in depth elucidation of stressful weather conditions and differing temporal responses of genotypes to those conditions. In field studies, however, the environment is often treated as a static factor, and the specific effects of weather variability on crop growth remain poorly understood. Here, we present a longitudinal dataset comprising 17,247 high-resolution RGB images of soybean breeding lines collected throughout eight years in Eschikon, Switzerland. Top-of-canopy images were acquired throughout the entire growing seasons and complemented by hourly weather data, enabling a comprehensive analysis of soybean growth dynamics under varying field conditions. High spatio-temporal image resolution allows detailed analysis of growth dynamics and GxE, supporting identification of stress-tolerant genotypes to improve yield prediction and yield stability.
Why it matches plant phenotyping methods8年間の高スループット画像フェノタイピングによる大規模データセットを提示し、画像取得基盤と作物成長動態の解析を中心に扱っているため、方法論文として適格です。
titleFIP 1.0 soybean data: Insights on soybean growth from eight years of high-throughput image field phenotyping
Reproduction assets foundThe paper's canopy cover analysis code is publicly available on the authors' ETH GitLab repository. The FIP 1.0 soybean image/trait dataset itself is deposited in the ETH Research Collection and Hugging Face, but those URLs are not among the allowed URLs, so only the code asset qualifies.Code · publicCode availability
The code is available on: https://gitlab.ethz.ch/crop_phenotyping/fip-soybean-canopycover. Users with similar data can use the implemented workflow to get canopy cover from their experiments.Open asset ↗gitlab.ethz.ch/crop_phenotyping/fip-soybean-canopycoverhtml-lines:207-226Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Published15 Feb 2026New Zealand Journal of Crop and Horticultural ScienceCited by 0 · OpenAlex ↗
Early plant disease detection is crucial to mitigate the disease progression and avoid the negative impacts on crops. Consequently, this research proposes the Ceta‐Coyote calibrated deep convolutional neural network (CtCODCN) for early plant leaf disease detection. Specifically, the proposed approach exploits the inherent feature extraction capability of CtCODCN and effectively learns the complex relationship between features, leading to improved disease detection. Besides, the Ceta‐Coyote optimization (CtCO) adaptively fine‐tunes the hyperparameters of the CtCODCN and improves the overall detection accuracy. In addition, the Local Binary and Ternary Residual Wavelet (LBTRW) approach extracts the textural and color data via capturing the smaller intensity variations by utilizing the wavelet features, local binary pattern (LBP), local ternary pattern (LTP), and Residual network‐101 (ResNet‐101). Moreover, the proposed approach utilizes advanced mechanisms to minimize the computational complexity and the error rate of the plant disease detection. The extensive experiments demonstrate the proposed CtCODCN model's effectiveness, evaluated in terms of metrics, achieving 98.270% of accuracy, 98.680% of sensitivity, and 97.04% of specificity, respectively, while using real‐time tomato images. Further, the proposed approach obtains the metric values of 96.39%, 99.64%, and 96.14% while using real‐time soya bean images, demonstrating the optimized performance of the proposed model.
Why it matches plant phenotyping methods植物葉の病害状態を画像から推定する深層学習手法を提案・評価しており、病害フェノタイピング手法が研究の中心です。
abstractthis research proposes the Ceta‐Coyote calibrated deep convolutional neural network (CtCODCN) for early plant leaf disease detection.
Lipid droplets (LDs) play pivotal roles in crop physiology, stress adaptation, and product quality by serving as dynamic reservoirs of energy and essential nutrients. However, the lack of rapid and accurate methods for on-site quantitative detection of LDs has hindered their comprehensive analysis in agricultural systems. Herein, we report the rational design and synthesis of three near-infrared (NIR) fluorescent probes, YD-1 , YD-2 , and YD-3 , for the sensitive and specific detection of LDs in crops. These probes feature a hydrophobic donor-π-acceptor (D-π-A) framework comprising a 7-diethylaminoquinoline electron donor and distinct electron-accepting groups, including (5,5-dimethylcyclohex-2-en-1-ylidene)malononitrile (DCM), 4-fluorophenyl, and phenyl moieties. Among them, YD-1 exhibited the most pronounced fluorescence response to LDs, displaying intense NIR emission within LDs and remarkable quenching in non-LD regions, enabling high-fidelity LD imaging. TD-DFT calculations revealed that the superior sensitivity of YD-1 originates from its efficient intramolecular charge transfer (ICT) process, which is highly responsive to the polarity differences between LDs and their surroundings. YD-1 was successfully applied to monitor LDs dynamics in living cells, zebrafish under a high-fat diet, and soybeans at different growth stages. Furthermore, a custom-built mobile fluorescence analysis device was developed and coupled to probe YD-1 to achieve rapid, on-site quantitative detection of LDs in crop samples. This work provides a powerful analytical platform for on-site monitoring of LDs dynamics, offering insights into crop lipid metabolism and quality control.
Why it matches plant phenotyping methods作物の脂質滴を対象に、蛍光プローブと携帯型蛍光解析装置を開発し、生体内画像化および現場での定量検出を実現しているため、植物表現型取得法が研究の中心である。
abstractwe report the rational design and synthesis of three near-infrared (NIR) fluorescent probes, YD-1 , YD-2 , and YD-3 , for the sensitive and specific detection of LDs in crops.
Green canopy cover dynamics extracted from high-throughput RGB imagery via automated labeled segmentation enabled nonlinear modeling that accounted for short-term weather variation, explaining up to 56% of protein yield variation in 150 soybean genotypes across ten seasons and enabling prediction in new environments as well as identification of tolerant genotypes.
Why it matches plant phenotyping methods高スループットRGB画像から自動セグメンテーションでキャノピー被覆率を抽出する手法が研究の中心であり、気象変動を考慮したモデル化と新環境での予測まで行っているため、植物表現型手法として採用。
abstractGreen canopy cover dynamics extracted from high-throughput RGB imagery via automated labeled segmentation
Soybean ( Glycine max L. ) performs an important position as a main resource of protein in Indonesia. Its quality and productivity can be assessed based on the characteristics of its seed. Accordingly, the identification process through the observation of soybean seed traits is a crucial step in plant breeding and quality assurance. Manual approaches rely on manual observation, which is subjective, prone to human error and time-consuming. With the improvement of artificial intelligence, automated seed identification has appeared as a potential solution. However, progress is constrained by the lack of open and standardized image datasets, especially for locally bred varieties in developing countries. To address this gap, we propose an open image dataset of Indonesian soybean seeds from three widely cultivated and plant-bred varieties: Anjasmoro, Grobogan, and DEGA-1. The dataset consists of high-resolution seed images captured with an Epson L360 flatbed scanner, with the optical resolution fixed at 800 dots per inch, yielding images of 6800 × 9359 pixels. All raw images are saved in JPG format. No manually segmentation masks are released in this version, instead of using Deeplab V3+ with MobileNet as backbone to enable the automated seed image segmentation. The curated dataset is intended to support a broad range of applications, including computer vision tasks such as image classification and segmentation, as well as research in plant breeding, seed quality assessment, and agricultural informatics. By providing a standardized and publicly accessible resource, this dataset contributes to the advancement of interdisciplinary studies at the intersection of agriculture and artificial intelligence.
Why it matches plant phenotyping methods大豆種子画像を標準化して公開するデータセット研究であり、種子形質の自動画像解析・セグメンテーションを支援する方法論的資源が中心です。
titleAn open image dataset of Indonesian soybean seed varieties (Anjasmoro, Grobogan, DEGA-1) for agricultural research and machine learning applications.
Reproduction assets foundThe paper is a data descriptor for a public Mendeley Data repository containing the authors' own soybean seed image dataset (raw scans and segmented seed images) used for seed phenotyping, with an explicit direct URL and DOI.Dataset · publicData accessibility
Repository name: Mendeley Data
Data identification number: DOI: 10.17632/c733bjz4m3.3
Direct URL to data: https://data.mendeley.com/datasets/c733bjz4m3/3Open asset ↗Mendeley Data · 10.17632/c733bjz4m3.3html-lines:115-142Plant phenotyping relevance match · UnverifiedCrossref · checked 15 Sept 2026
Published1 Feb 2026Engineering Applications of Artificial IntelligenceCited by 2 · OpenAlex ↗
Improving crop yield prediction accuracy is crucial for precision agriculture, particularly for irrigation management. Unmanned aerial vehicle (UAV)-based multispectral imaging has become a key tool for crop phenotyping due to its high spatiotemporal resolution and cost-effectiveness. In this study, field experiments were conducted in northwestern China over two consecutive growing seasons (2021–2022), incorporating different mulching practices and supplemental irrigation treatments, to systematically analyze the sensitivity of soybean seed yield to various physiological and growth indices measured at different phenological stages. The results indicated that the full pod stage (R4) was the most sensitive window for yield prediction. At this stage, canopy cover (CC) and chlorophyll content reached their peak values. Most vegetation indices (VIs), texture features (TFs), and texture indices (TIs) extracted from the UAV imagery showed significant correlations (P < 0.05) with final seed yield. Among these, the ratio texture index (RTI, defined as DIS1/HOM3) exhibited the strongest correlation with yield (R = 0.69). A three-source data fusion framework combining VIs, TFs, and TIs was constructed, and an extreme gradient boosting (XGBoost) algorithm was applied to optimize feature weighting. This integrated model achieved optimal performance at the R4 stage, with coefficient of determination R² = 0.83 on the validation set, root mean square error (RMSE) = 280.80 kg ha⁻¹ , and mean relative error (MRE) = 6.32 %. Compared to a model based solely on spectral VIs (R² = 0.63), the multi-source XGBoost model improved R² by 31.7 % and reduced the error metrics (RMSE) by up to 17.5 %. These findings provides a theoretical basis for precise field management in arid areas and a technical framework for remote sensing monitoring of crop yield.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像からキャノピー形態・生理指標を抽出し、特徴量融合とXGBoostで大豆収量を推定する手法が中心であり、検証性能も報告している。
abstractUnmanned aerial vehicle (UAV)-based multispectral imaging has become a key tool for crop phenotyping due to its high spatiotemporal resolution and cost-effectiveness.
Introduction High-throughput and accurate phenotyping is critical for enhancing crop breeding efficiency by enabling rapid identification of superior cultivars within large populations. For soybean [ Glycine max (L.) Merr. ], maturity group is a key determinant of geographic adaptation and influences yield potential. Consequently, accurate assessment of physiological maturity dates is essential for selecting lines suited to specific environments. This study evaluated the feasibility of three transfer learning techniques in improving the generalizability of models developed using historical data to predict the maturity dates of soybean breeding lines across new environments. Methods Our dataset included five breeding trials conducted in two sites from 2018 to 2021. Maturity dates were visually assessed at the R8 stage, and multispectral imagery from an unmanned aerial vehicle (UAV) was collected within each trial. Seven image features served as predictors in the models. Transfer learning techniques, namely pre-training and fine-tuning, single-source and multiple-source domain adaptation, were evaluated using the multiple-year datasets. Results When models were trained on data from three prior years and tested on two independent trials, the pre-training and fine-tuning technique demonstrated the best performance, with the highest agreement with visual ratings (coefficient of determination R 2 = 0.74 and 0.79) and root mean square errors of 1.70 and 1.96 days, respectively. The quantity for fine-tuning samples had minimal influence on the prediction accuracy for previously unseen data. Discussion These findings provide a reference for leveraging accumulated knowledge to generalize deep learning models for future practical utilization.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像からダイズ成熟日を推定するモデルについて、転移学習とドメイン適応の性能・汎化性を評価しており、植物形質取得手法が中心です。
abstractThis study evaluated the feasibility of three transfer learning techniques in improving the generalizability of models developed using historical data to predict the maturity dates of soybean breeding lines across new environments.
Accurate, near real-time soybean phenology information is critical for crop management and breeding. Previous approaches relying on satellite remote sensing time-series data suffer from temporal delays, limiting their usefulness for in-season decision-making. To overcome this limitation, this study reframes phenology identification as a near real-time classification task using single-timepoint Unmanned Aerial Vehicle (UAV) imagery collected from 420 soybean germplasm resources across three experimental sites, and proposes an innovative multi-modal dynamic Gating Fusion Model that integrates two optimized pathways. one based on machine learning (ML) and the other on deep learning (DP). In the ML branch, systematic benchmarking of tabular-feature models identified the Soft Voting ensemble as the best classifier. In the DL branch, an enhanced BC-ConvNeXt model equipped with BiFPN and CBAM modules was developed to strengthen visual feature extraction. Building on these two optimal classifiers, the dynamic gating fusion model achieved the highest F1-score of 94.3% across seven key growth stages (V1, V2, R1, R2, R6, R7, R8). This result represents a significant improvement of 1.5% and 10.6% over the best performing ML and DL models, respectively. The superior performance arises from the intelligent arbitration of complementary strengths, with gating-weight analysis revealing a strategy that prioritizes ML predictions while leveraging DL for error correction. This work establishes a complete framework for near real-time crop phenology detection and demonstrates the strong potential of intelligent multi-modal fusion in high-throughput phenotyping.
Why it matches plant phenotyping methodsUAV画像からダイズの生育段階を推定するマルチモーダル分類モデルを開発・ベンチマークしており、植物表現型取得手法が研究の中心である。
abstractproposes an innovative multi-modal dynamic Gating Fusion Model
Point clouds and digital surface models (DSMs) derived from unmanned aircraft system (UAS) imagery are widely used for plant height estimation in plant phenotyping and precision agriculture. However, comprehensive evaluations across multiple crops, flight altitudes, and image overlaps are limited, restricting guidance for optimizing flight strategies. This study evaluated the effects of flight altitude, side and front overlap, and image processing parameters on point cloud generation and plant height estimation. UAS imagery was collected at four altitudes (30–120 m, corresponding to 0.5–2.0 cm ground sampling distance, GSD) with multiple side and front overlaps (67–94%) over a 2–ha field planted with corn, cotton, sorghum, and soybean on three dates across two growing seasons, producing 90 datasets. Orthomosaics, point clouds, and DSMs were generated using Pix4Dmapper, and plant height estimates were extracted from both DSMs and point clouds. Results showed that point clouds consistently outperformed DSMs across altitudes, overlaps, and crop types. Highest accuracy occurred at 60–90 m (1.0–1.5 cm GSD) with RMSE values of 0.06–0.10 m (R2 = 0.92–0.95) in 2019 and 0.07–0.08 m (R2 = 0.80–0.89) in 2022. Across multiple side and front overlap combinations at 60–120 m, reduced overlaps produced RMSE values comparable to full overlaps, indicating that optimized flight settings, particularly reduced side overlap with high front overlap, can shorten flight and processing time without compromising point cloud quality or height estimation accuracy. Pix4Dmapper processing parameters strongly affected 3D point cloud density (2–600 million points), processing time (1–16 h), and plant height accuracy (R2 = 0.67–0.95). These findings provide practical guidance for selecting UAS flight and processing parameters to achieve accurate, efficient 3D modeling and plant height estimation. By balancing flight altitude, image side and front overlap, and photogrammetric processing settings, users can improve operational efficiency while maintaining high-accuracy plant height measurements, supporting faster and more cost-effective phenotyping and precision agriculture applications.
Why it matches plant phenotyping methodsUAS画像からの点群・DSM生成と草丈推定について、飛行条件および処理パラメータの影響を体系的に評価・検証しており、植物表現型取得法が研究の中心である。
abstractThis study evaluated the effects of flight altitude, side and front overlap, and image processing parameters on point cloud generation and plant height estimation.
In order to realize the rapid detection of soybean seed germination potential, this study designed a fusion model to solve the problem that the single model was insufficient in spectral feature analysis and the prediction performance was limited. The model combines the advantages of the Partial Least Squares Regression (PLSR) and the Multilayer Perceptron (MLP), and utilizing principal components extracted by PLSR as the input features for MLP to construct a soybean seed germination potential prediction model with both linear and nonlinear modeling capabilities. The PLSR module accurately extracts the linear features of the spectrum, and the MLP network further captures the nonlinear relationship between the spectral data and the target variable, which significantly improves the generalization ability of the model. The experimental results show that the prediction performance of the proposed PLSR-MLP fusion model (R p 2 = 0.9534, RMSEP = 7.3821) is significantly improved compared with the single PLSR model (R p 2 = 0.7284, RMSEP = 17.8154) and the single MLP model (R p 2 = 0.7935, RMSEP = 15.5335). In the prediction of soybean germination potential, the PLSR-MLP model also outperforms other single models (Support Vector Machine, SVM; Random Forest, RF) and other fusion models such as PLSR-SVM and PLSR-RF. The PLSR-MLP fusion model effectively addresses the limitations of a single model's performance enhancement potential and the susceptibility to overfitting. It provides a new method for the efficient evaluation of seed germination potential. It also has practical application value for precision seed selection in agriculture and offers a new idea for near-infrared spectrum modeling.
Why it matches plant phenotyping methods大豆種子の発芽能力という植物状態を近赤外スペクトルとPLSR-MLP融合モデルで推定する手法の開発・比較検証が研究の中心である。
abstractthis study designed a fusion model to solve the problem that the single model was insufficient in spectral feature analysis and the prediction performance was limited.
While essential for precision agriculture, the accurate and dynamic monitoring of crop phenotypic parameters faces challenges, including the constraints of single-data sources and insufficient model generalization across growth stages. This research introduced an integrated framework that leverages multi-source data fusion and the XGBoost algorithm to estimate key soybean parameters, including Leaf Area Index (LAI) and Above-Ground Biomass (AGB). Field experiments incorporated different irrigation methods (drip/micro-sprinkler) and planting densities (210,000/270,000 plants ha −1 ), multispectral images and corresponding ground truth data were acquired across five critical growth stages.We extracted 11 vegetation indices (V) and 8 texture features (T) and constructed inversion models using Support Vector Regression (SVR), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) based on single and multi-source (V+T) features. The results indicated that: the multi-source feature fusion model outperformed single-feature models. The XGBoost algorithm outperformed all other models, achieving average R 2 values of 0.673, and 0.671, and RMSE values of 0.117, and 79.751 kg ha −1 for LAI, and AGB inversion, respectively. The full pod stage (R4) was identified as the optimal remote sensing observation window, where the best models achieved R 2 values of 0.846 (LAI) and 0.731 (AGB), with RMSE values of 0.131 and 81.01 kg ha −1 , respectively. Drip irrigation combined with high planting density significantly ( P < 0.05) increased soybean LAI and AGB. This study provides a robust, high-throughput technical solution for dynamic crop phenotyping, it highlights the value of fusing multi-source UAV features with machine learning for advancing data-driven smart agriculture. • Achieved dynamic soybean phenotyping by fusing unmanned aerial vehicle (UAV) multi-source features with machine learning. • Multi-source feature fusion outperformed single-feature models in accuracy and robustness. • Full pod stage identified as the optimal UAV remote sensing observation window. • Drip irrigation with high planting density significantly enhanced soybean Leaf Area Index and Above-Ground Biomass.
Why it matches plant phenotyping methodsUAVマルチソース画像と機械学習により、LAIおよび地上部バイオマスを推定する方法を開発・比較検証しており、植物表現型取得が研究の中心である。
abstractThis research introduced an integrated framework that leverages multi-source data fusion and the XGBoost algorithm to estimate key soybean parameters, including Leaf Area Index (LAI) and Above-Ground Biomass (AGB).
Introduction Soybean diseases pose a significant threat to global crop yield and food security, necessitating rapid and accurate identification for effective management. While deep learning offers promising solutions for plant disease recognition, existing models often struggle with the complexities of in-field soybean disease identification, particularly due to high intra-class variations and subtle inter-class differences. Methods To address these challenges, we propose a novel region-specific feature decoupling and adaptive fusion network (RFDAF-Net) designed for robust and precise soybean disease recognition under real-world field conditions. The core of RFDAF-Net consists of two key components: a region-specific feature decoupling (RFD) module that enhances discriminative patterns and suppresses redundant information through a dual-pathway design, explicitly separating shallow, intermediate, and deep features; and a region-specific feature adaptive fusion (RFAF) module that dynamically integrates these multi-scale features via learned spatial attention. This hierarchical feature decomposition effectively isolates discriminative disease signatures while suppressing irrelevant variations. The architecture is flexible, enabling seamless integration with various backbone networks including both convolutional neural networks and Transformers. Results We evaluate RFDAF-Net extensively on a comprehensive soybean disease dataset containing images captured in diverse field environments. Experimental results show that our method significantly outperforms current state-of-the-art models across multiple architectures, achieving a top accuracy of 99.43% when implemented with a Swin-B backbone. Discussion The proposed framework offers an interpretable and field-ready solution for precision crop protection, demonstrating strong generalization ability and practical utility for real-world agricultural applications.
Why it matches plant phenotyping methods圃場画像からダイズ病害を識別する深層学習ネットワークを新規開発し、データセット上で広範に評価しているため、植物病害状態の取得・推定手法が研究の中心である。
abstractwe propose a novel region-specific feature decoupling and adaptive fusion network (RFDAF-Net) designed for robust and precise soybean disease recognition under real-world field conditions.
Maize-soybean intercropping is a sustainable intensive agroecosystem, though the productivity is constrained by interspecific competition for water and light resources. To enhance the water use efficiency in this intercropping system and understand canopy structure dynamics under the water-limited conditions of arid northwest China, this study proposes a novel optimization strategy that synchronizes deficit irrigation scheduling with crop-specific water requirements during critical phenological phases. Four irrigation regimes were implemented: W1 (full irrigation for both maize and soybean crops), W2 (maize-full and soybean-deficit), W3 (maize-deficit and soybean-full), and W4 (dual deficit). Through UAV-based high-resolution 3D canopy reconstruction (R = 0.98 for plant height validation), 14 spatial-geometric descriptors were quantified. The W2 strategy demonstrated superior competitive coordination, enhancing aggressivity of maize (Ams) by 85.9 % through strategic canopy reconfiguration: 11.8 % reduction in maize maximum leaf layer width position (MLLWP), 28.3 % decrease in inter-specific canopy overlap area (COA), and 40.0 % compression of shading convex hull volume (SCHV). These optimized structural adaptations synergistically enhanced photosynthetically active radiation interception (+13.4 %) while achieving concurrent reductions in crop evapotranspiration (ET, -19.7 %) without yield penalty, thereby elevating irrigation water use efficiency (IWUE) by 14.4 % and water equivalent ratio (WER) by 15.9 %. This work provides mechanistic insights into canopy architecture-mediated resource competition mitigation and establishes a technological framework for sustainable intensification in water-limited environments.
Why it matches plant phenotyping methodsUAVによる3Dキャノピー再構成を用いた植物構造形質の取得と検証が、灌漑試験の主要な解析基盤として明示されているため、実質的なフェノタイピング手法の応用に該当する。
abstractThrough UAV-based high-resolution 3D canopy reconstruction (R = 0.98 for plant height validation), 14 spatial-geometric descriptors were quantified.
Reproduction assets foundThe paper's data availability statement explicitly deposits the authors' analysis source code on a public GitHub repository, which qualifies as a paper-specific public code asset. The study's phenotype data (UAV-derived 3D canopy point clouds, geometric trait measurements, yield/biomass data) are only available upon请求,Code · publicThe source code used in this study is available for noncommercial use and the code can be downloaded from https://github.com/Pepe-oss/3D-Reconstruction-analysis-of-maize-soybean-intercropping-competition-under-water-stress . The data of this study are available from the corresponding author upon request.Open asset ↗Pepe-oss/3D-Reconstruction-analysis-of-maize-soybean-intercropping-competition-under-water-stresslines:320-407Code / dataset availability confirmedEurope PMC · OpenAlex · checked 5 Sept 2026
Three-dimensional (3D) reconstruction technologies for crops are of significant importance in the context of smart breeding and precision agriculture, as they enable accurate characterization of crop spatial architecture and developmental dynamics. Such capabilities provide essential phenotypic information for the rapid selection of breeding materials and informed agronomic decision-making. A critical requirement for the practical application of crop 3D models is high-accuracy organ-level segmentation. However, the absence of a stage-universal segmentation framework capable of operating across complete soybean growth cycle remains a major bottleneck hindering progress in this field. To address this issue, we propose SOY3DSEG-a high-precision framework based on an improved Point Transformer, designed to support the full developmental spectrum of soybean (V1-R7). The framework incorporates a novel down sampling strategy termed Dynamic Multi-Stage Sampling Strategy (DMSS), alongside multi-scale feature enhancement and a local geometry-aware attention mechanism, enhancing segmentation accuracy and efficiency. Performance evaluations across 12 consecutive soybean growth stages (V1 to R7) indicate that SOY3DSEG achieved an average mean Intersection-over-Union (mIoU) of 93.34 % for stem-leaf segmentation-surpassing RandLA-Net, BAAF-Net, PointNet++, and PointConv by over 30 %, and outperforming the baseline Point Transformer by 14.18 %. A moderate accuracy decline appears at R6-R7 due to dense canopies and strong occlusion, yet SOY3DSEG retains clear superiority over the baseline Point Transformer, demonstrating robustness under complex morphology. In cross-crop transfer tests limited to early seedling stages of maize and tomato, the model achieves an mIoU of approximately 99 %, indicating strong early-stage transferability while mature-stage generalization across species remains open for future study. SOY3DSEG thus provides a stage-robust and scalable solution for full-cycle soybean phenotyping and growth monitoring, contributing to precision agricultural practice.
Why it matches plant phenotyping methods大豆の3D点群から器官レベル形態を抽出する分割フレームワークを開発・評価しており、植物表現型取得手法が研究の中心である。
abstractA critical requirement for the practical application of crop 3D models is high-accuracy organ-level segmentation.
Reproduction assets foundThe authors state that the dataset (Soybean-MVS point clouds) and program code used in this study are publicly available at their GitHub repository, which is an allowed URL.Code · publicThe dataset and program code used in this study can be found at the link below: https://github.com/NiuJiarui718/SOY3DSEG .Open asset ↗NiuJiarui718/SOY3DSEGlines:306-323Plant phenotyping relevance match · UnverifiedCrossref · checked 14 Sept 2026
Published10 Jan 2026International Journal of Scientific Research in Engineering and ManagementCited by 0 · OpenAlex ↗
Abstract - The rapid advancement of unmanned aerial vehicles (UAVs) and deep learning techniques has significantly transformed crop monitoring and precision agriculture. Among various crops, soybean plays a crucial role in global food and oilseed production, making timely and accurate crop health assessment essential. This review presents a comprehensive analysis of UAV-based soybean crop health evaluation methods, with a particular focus on image-based disease detection and stress monitoring using deep learning models. Recent progress in convolutional neural networks, patch-level image analysis, attention mechanisms, and lightweight architectures is systematically examined. The paper discusses commonly used UAV imaging modalities, preprocessing strategies, model architectures, and evaluation practices reported in the literature. Furthermore, existing challenges such as environmental variability, computational complexity, data imbalance, and real-world deployment constraints are critically analyzed. Based on the reviewed studies, potential research directions are identified, emphasizing efficient patch-level learning, interpretable health mapping, and scalable field-level assessment. This review aims to provide researchers and practitioners with a clear understanding of current trends, limitations, and future opportunities in UAV-assisted soybean crop health monitoring. Key Words: - Unmanned Aerial Vehicles (UAVs), Soybean Crop Health Monitoring, Precision Agriculture, Deep Learning
Why it matches plant phenotyping methodsUAV画像と深層学習によるダイズの病害・ストレス評価手法を体系的にレビューしており、植物状態の画像ベース取得・推定方法が中心である。
abstractThis review presents a comprehensive analysis of UAV-based soybean crop health evaluation methods, with a particular focus on image-based disease detection and stress monitoring using deep learning models.
Early and precise diagnosis of crop diseases is crucial for global food security, particularly in developing countries where agriculture still plays a dominant role. This study presents a deep learning approach for labelling ten different plant disease conditions across three principal crops—maize, potato, and soybean. The Convolutional Neural Network (CNN) model incorporates multiple convolutional and batch normalization layers, achieving an overall classification accuracy of 95 %. Class-wise F1-scores range from 0.84 to 0.96, with notably strong performance for the Potato-Healthy and Soybean-Healthy categories. The model demonstrates robust generalization to variations in background, lighting, and leaf orientation, highlighting its suitability for real-world agricultural environments. This work supports the development of automated, scalable, and accurate multi-crop disease detection systems. The study also examines challenges such as class imbalance and overfitting, and proposes improvements including the integration of attention mechanisms and transfer learning. However, the model’s performance is still limited by the relatively small dataset size and restricted environmental diversity, suggesting future scope for expansion through larger field-based datasets, multimodal sensing, and advanced hybrid architectures.
Why it matches plant phenotyping methods葉画像から植物病害状態を分類するCNN手法が研究の中心であり、精度やF1スコアによる性能評価も行っているため、植物フェノタイピング手法として収録する。
titleA Convolutional Neural Network (CNN) Based Classification Framework for Multi-Crop Disease Detection using Leaf Images
Introduction Soybean mosaic virus (SMV) is one of the major pathogens affecting global soybean yield and quality, and its early and accurate detection is essential for disease warning and precision management. This study proposes a non-invasive early detection method by integrating portable Raman spectroscopy with artificial intelligence algorithms. Methods Raman spectra of leaves from both resistant and susceptible soybean cultivars were collected at different infection stages (0, 2, 4, and 6 days post-inoculation), and preprocessed using Savitzky-Golay (S-G) smoothing and adaptive iteratively reweighted penalized least squares (Air-PLS) baseline correction. Four classification models-1D-CNN, SVM, KNN, and BP-ANN-were developed to classify samples from different infection stages. Results Spectral feature analysis revealed significant changes in carotenoid levels caused by viral infection, and distinct spectral responses between resistant and susceptible cultivars during disease progression. Among the four classification models, the 1D-CNN model achieved the highest prediction accuracy of 90%. In addition, principal component analysis (PCA) indicated that the Raman spectroscopy-based method significantly advanced the early detection of SMV (SC3) to 4 days post-inoculation, compared to 7-10 days required by conventional methods. Discussion This evidences the superior capability of Raman spectroscopy for monitoring the dynamics of SMV infection and its potential to considerably reduce the duration of diagnosis. This study confirms the feasibility and efficiency of Raman spectroscopy combined with deep learning for in situ early detection of plant viral diseases and provides a promising reference for non-destructive diagnosis of early-stage foliar infections.
Why it matches plant phenotyping methods携帯型ラマン分光と機械学習を統合し、感染葉の病態を非破壊・早期検出する方法を開発および評価しており、植物病害状態の表現型取得が中心である。
abstractThis study proposes a non-invasive early detection method by integrating portable Raman spectroscopy with artificial intelligence algorithms.
Accurate estimation of above-ground biomass (AGB) and plant height is essential for precision crop management. However, traditional methods like synthetic aperture radar (SAR) data and optical vegetation indices (VIs) often face signal saturation at medium to high AGB levels. To address this, we proposed two polarization texture indices, i.e., Ratio SAR Texture Index (RSTI) and Normalized Difference SAR Texture Index (NDSTI), derived from Sentinel-1 (S-1) data to estimate crop AGB and height. We further investigated their integration with S-1 polarizations and Sentinel-2 (S-2) VIs using four machine learning algorithms to enhance retrieval performance. Results revealed that both RSTI and NDSTI outperformed individual polarizations, polarization texture features, and most of VIs in estimating crop AGB and height. Furthermore, the combination of these indices with S-2 VIs significantly improved the retrieval accuracy. The optimal models achieved R2 values up to 0.75 and 0.80 for maize and soybean AGB, 0.89 and 0.94 for maize and soybean height, respectively. Validation with an independent dataset confirmed the robustness and transferability of the proposed models for estimating maize AGB and height. Overall, RSTI and NDSTI, along with their integration with optical VIs, provide an effective approach for improving crop AGB and height estimation for agricultural monitoring.
Why it matches plant phenotyping methodsSentinel-1/2リモートセンシングから作物バイオマスと草丈を推定する新規テクスチャ指数を開発し、独立データで検証しており、植物形質取得手法が中心である。
abstractwe proposed two polarization texture indices, i.e., Ratio SAR Texture Index (RSTI) and Normalized Difference SAR Texture Index (NDSTI), derived from Sentinel-1 (S-1) data to estimate crop AGB and height.
ABSTRACT High-throughput phenotyping using unmanned aerial vehicles (UAVs) and spectral vegetation indices has been proposed to overcome the cost and logistical constraints of manual measurements in multi-environment breeding trials. However, the reliability of models trained on spectral data to predict structural traits across genotypes and environments remains unclear. This study aimed to develop an approach for predicting soybean plant height (PH) and first pod insertion height (FPIH) using UAV-based vegetation indices acquired at the flowering stage, as well as to compare extreme gradient boosting (XGBoost), multilayer perceptron (MLP), random forest (RF), and multiple linear regression (MLR) models under realistic cross-validation scenarios. Trials were conducted across multiple seasons using UAV multispectral imagery, with PH and FPIH manually measured. The models were evaluated under five phenotyping scenarios: baseline calibration using all data; prediction in a completely unmeasured future season; estimation of missing genotypes within a partially sampled season; calibration using a small fraction of data from a new season; and prediction under absence of field records for specific genotypes across environments. When all data were used for calibration, non-linear models showed a high apparent accuracy. However, prediction in unseen seasons failed for all models, reflecting strong genotype × environment interactions. Under reduced phenotyping within the same environment network, the models maintained a robust accuracy for PH, whereas FPIH predictions declined to moderate levels. UAV-based models are reliable for interpolation, but limited for extrapolation without local calibration, enabling reductions of up to 80 % in manual measurements for PH and 20-30 % for FPIH.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像と機械学習によるダイズ形質推定を開発・比較検証し、異なる環境での予測性能と手測定削減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。
abstractThis study aimed to develop an approach for predicting soybean plant height (PH) and first pod insertion height (FPIH) using UAV-based vegetation indices acquired at the flowering stage, as well as to compare extreme gradient boosting (XGBoost), multilayer perceptron (MLP), random forest (RF), and multiple linear regression (MLR) models under realistic cross-validation scenarios.
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
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
Satellite-based crop phenology provides critical information for agricultural management; however, accurately detecting specific crop development stages remains challenging. During early stages, such as sowing and emergence, satellite imagery captures a spectral signal that is a mixture of soil and vegetation. This study developed an operational framework for estimating field-scale sowing and emergence dates using daily synthetic Harmonized Landsat Sentinel-2 (HLS) data. The assumption is that sowing and emergence dates can be estimated by using later growth stages since crop development follows a consistent pattern driven by physiological processes and environmental conditions. We first evaluated 4 gap-filling techniques to generate a daily synthetic HLS-based enhanced vegetation index (EVI) time series over 15 tiles across the USA. Then, 6 phenological metrics were retrieved using the asymmetric double sigmoid function, and the results were validated over 20 PhenoCam sites with corn and soybeans. Different predictive models were evaluated for sowing and emergence date estimation, and the optimal approach was used to predict these dates in crop fields in Iowa and Missouri. The polynomial gap-filling technique performed best in reconstructing the original EVI images, and phenological stages derived from daily EVI images and PhenoCam data showed strong agreement, with an R 2 of 0.94 and a bias of 12 d. Elastic net regression performed better in estimating sowing and emergence dates, with a root mean square error of ±10 d. The proposed framework offers a consistent pipeline to reconstruct gap-free HLS data, extract phenological stages, and estimate sowing and emergence dates for agricultural monitoring.
Why it matches plant phenotyping methods衛星時系列から作物の播種・出芽日や生育段階を抽出する手法を開発し、PhenoCamで検証しており、植物表現型取得が研究の中心である。
abstractThis study developed an operational framework for estimating field-scale sowing and emergence dates using daily synthetic Harmonized Landsat Sentinel-2 (HLS) data.
SoybeanNeRF / 3D Gaussian SplattingLiDAR / point cloudSegmentationTracking
Abstract has not been obtained from indexed metadata or an accessible article page.
Why it matches plant phenotyping methods成熟ダイズ個体を対象とする点群インスタンスセグメンテーション・パイプラインの開発であり、植物画像から個体を分離・再構成する手法が中心です。
titleSoybeanInsGS: A high-precision, data-efficient point cloud instance segmentation pipeline for mature soybean plants via cross-view instance tracking and instance-aware 3DGS
Maize-soybean intercropping is a sustainable intensive agroecosystem, though the productivity is constrained by interspecific competition for water and light resources. To enhance the water use efficiency in this intercropping system and understand canopy structure dynamics under the water-limited conditions of arid northwest China, this study proposes a novel optimization strategy that synchronizes deficit irrigation scheduling with cropspecific water requirements during critical phenological phases. Four irrigation regimes were implemented: W1 (full irrigation for both maize and soybean crops), W2 (maize-full and soybean-deficit), W3 (maize-deficit and soybean-full), and W4 (dual deficit). Through UAV-based high-resolution 3D canopy reconstruction (R = 0.98 for plant height validation), 14 spatial-geometric descriptors were quantified. The W2 strategy demonstrated superior competitive coordination, enhancing aggressivity of maize (Ams) by 85.9 % through strategic canopy reconfiguration: 11.8 % reduction in maize maximum leaf layer width position (MLLWP), 28.3 % decrease in inter-specific canopy overlap area (COA), and 40.0 % compression of shading convex hull volume (SCHV). These optimized structural adaptations synergistically enhanced photosynthetically active radiation interception (+13.4 %) while achieving concurrent reductions in crop evapotranspiration (ET, -19.7 %) without yield penalty, thereby elevating irrigation water use efficiency (IWUE) by 14.4 % and water equivalent ratio (WER) by 15.9 %. This work provides mechanistic insights into canopy architecture-mediated resource competition mitigation and establishes a technological framework for sustainable intensification in water-limited environments.
Why it matches plant phenotyping methodsUAVベースの3Dキャノピー再構成を用いて植物構造形質を抽出し、草丈検証と14個の空間・幾何記述子の定量を行っており、表現型取得・解析が実質的に記述されている。灌漑試験への応用ではあるが、方法の検証と再利用可能なワークフローが明示されているため採用。
abstractThrough UAV-based high-resolution 3D canopy reconstruction (R = 0.98 for plant height validation), 14 spatial-geometric descriptors were quantified.
Soybean-rhizobia symbiotic nitrogen fixation, a process in which rhizobia mediate biological nitrogen fixation by converting inert atmospheric nitrogen (N 2 ) into biologically available forms (e.g., ammonium, NH 4 + ), has been extensively investigated. However, non-invasive, in situ monitoring methods for this process remain lacking. Herein, we report a solid-state membrane potentiometric ammonium ion-selective microelectrode (NH 4 + -ISμE) for the in situ detection of NH 4 + in soybean root nodules. A Prussian blue analogue with ion channels, which enables the specific insertion/extraction of NH 4 + ions while excluding interfering cations, was electrodeposited on a carbon fiber to fabricate the microelectrode. The cation sorption capability and ion selectivity of the thin film were explored by modulating the intercalation/deintercalation process and reducing the interfering cations within the framework. The NH 4 + -ISμE exhibits a Nernstian response to NH 4 + over the concentration range of 1.0 × 10 -6 to 1.0 × 10 -3 M, with a detection limit of 6.2 × 10 -7 M. This sensor enables in situ, real-time detection of NH 4 + -the direct product of biological nitrogen fixation in the legume plant-rhizobium symbiotic system. The release of NH 4 + ions in soybean root nodules during nitrogen fixation was successfully monitored. Overall, this work provides a simple and versatile tool for studying and monitoring biological symbiotic nitrogen fixation processes.
Why it matches plant phenotyping methodsダイズ根粒内のアンモニウムと窒素固定状態を非侵襲・リアルタイムに測定する新規マイクロセンサーを開発し、性能評価と植物体内での実証を行っており、植物生理状態の取得法が中心である。
abstractHerein, we report a solid-state membrane potentiometric ammonium ion-selective microelectrode (NH 4 + -ISμE) for the in situ detection of NH 4 + in soybean root nodules.
The non-destructive estimation of grain yield could increase the efficiency of soybean breeding through early genotype testing, allowing for more precise selection of superior varieties. High-throughput phenotyping (HTPP) data can be combined with machine learning (ML) to develop accurate prediction models. In this study, an unmanned aerial vehicle (UAV) equipped with a multispectral camera was utilized to collect data on plant density (PD), plant height (PH), canopy cover (CC), biomass (BM), and various vegetation indices (VIs) from different stages of soybean development. These traits were used within random forest (RF) and partial least squares regression (PLSR) algorithms to develop models for soybean yield estimation. The initial RF model produced more accurate results, as it had a smaller error between actual and predicted yield compared with the PLSR model. To increase the efficiency of the RF model and optimize the data collection process, the number of predictors was gradually decreased by eliminating highly correlated VIs and selecting the most important variables. The final prediction was based only on several VIs calculated from a few mid-soybean stages. Although the reduction in the number of predictors increased the yield estimation error to some extent, the R2 in the final model remained high (R2 = 0.79). Therefore, the proposed ML model based on specific HTPP variables represents an optimal balance between efficiency and prediction accuracy for in-season soybean yield estimation.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像から植物形質を取得し、機械学習でダイズ収量を推定するワークフローが研究の中心であり、形質抽出・予測モデルの開発と効率化を扱っている。
abstractHigh-throughput phenotyping (HTPP) data can be combined with machine learning (ML) to develop accurate prediction models.
Sustainable improvement of crop performance requires integrative approaches that link genomic variation to phenotypic expression through intermediate molecular pathways. Here, we present Reciprocal Best Linear Unbiased Prediction (Reciprocal BLUP), a predictability-guided multi-omics framework that quantifies the cross-layer relationships among the genome, metabolome, and microbiome to enhance phenotype prediction. Using a panel of 198 soybean accessions grown under well-watered and drought conditions, we first evaluated four direction-specific prediction models (genome → microbiome, genome → metabolome, metabolome → microbiome, and microbiome → metabolome) to estimate the predictability of individual omics features. We evaluated whether subsets of features with high cross-omics predictability improved phenotype prediction. These cross-layer models identify features that play physiologically meaningful roles within multi-omics systems, enabling the prioritization of variables that capture coherent biological signals enriched with phenotype-relevant information. Consequently, metabolome features were highly predictable from microbiome data, whereas microbiome predictability from metabolomic data was weaker and more environmentally dependent, revealing an asymmetric relationship between these layers. In the subsequent phenotype prediction analysis, the model incorporating predictability-based feature selection substantially outperformed models using randomly selected features and achieved prediction accuracies comparable to those of the full-feature model. Under drought conditions, the phenotype prediction models based on metabolomic or microbiomic kernels (MetBLUP or MicroBLUP) outperformed the genomic baseline (GBLUP) for several biomass-related traits, indicating that the environment-responsive omics layers captured phenotypic variations that were not explained by additive genetic effects. Our results highlight the hierarchical interactions among genomic, metabolic, and microbial systems, with the metabolome functioning as an integrative mediator linking the genotype, environment, and microbiome composition. The Reciprocal BLUP framework provides a biologically interpretable and practical approach for integrating multi-omics data, improving phenotype prediction, and guiding omics-based feature selection in plant breeding.
Why it matches plant phenotyping methods植物形質予測のための新しい多層オミクス統合フレームワークを提案し、予測モデル比較と性能評価を行っているため、計算的フェノタイピング手法が中心である。
abstractwe present Reciprocal Best Linear Unbiased Prediction (Reciprocal BLUP), a predictability-guided multi-omics framework that quantifies the cross-layer relationships among the genome, metabolome, and microbiome to enhance phenotype prediction.
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits all source codes and data in a public GitHub repository (Yoska393/ReciprocalBLUP), which contains the authors' analysis code and data for the soybean multi-omics phenotype prediction study. The NARO Genebank URL is only the source of plant accessions, not a phCode · publicAll source codes and data are available from the repository in GitHub: https://github.com/Yoska393/ReciprocalBLUP (accessed on 20 November 2025).Open asset ↗Yoska393/ReciprocalBLUPlines:285-308Plant phenotyping relevance match · UnverifiedOpenAlex · checked 15 Sept 2026
Published9 Dec 2025Engineering Applications of Artificial IntelligenceCited by 0 · OpenAlex ↗
Abstract has not been obtained from indexed metadata or an accessible article page.
Why it matches plant phenotyping methodsマルチスペクトル画像からダイズの高温耐性等級を予測する特徴量・学習フレームワークが題名上の中心であり、植物のストレス耐性状態を推定するフェノタイピング手法に該当する。
titleA multispectral feature framework for predicting soybean high temperature resistance grades based on masked autoencoding and supervised contrastive learning with dual-branch pretraining
SoybeanLaboratory / benchtopRootMorphology / geometry measurement2D/3D reconstructionSegmentationVisualization / data managementRoot system architecture
Root system analysis remains methodologically challenging in plant research: traditional soil cultivation obstructs comprehensive root observation, whereas hydroponic visualization lacks ecological relevance due to soil environment exclusion—a critical limitation for crops like soybean. This manuscript developed a cost-effective hybrid imaging system integrating transparent acrylic plates, semi-permeable membranes, and natural soil substrates with high-resolution imaging and controlled illumination, enabling non-destructive root monitoring in quasi-natural soil conditions. Complementing this hardware innovation, this manuscript proposed an unsupervised semantic segmentation algorithm that synergizes path planning with an enhanced DBSCAN framework, achieving the precise extraction of primary and lateral root architectures. Experimental validation demonstrated superior performance in soybean root analysis, with segmentation metrics reaching 0.8444 accuracy, 0.9203 recall, 0.8743 F1-score, and 0.7921 mIoU—significantly outperforming existing unsupervised methods (p 0.94) with WinRHIZO in quantifying root length, projected area, dimensional parameters, and lateral root counts confirmed system reliability. This soil-compatible phenotyping platform establishes new opportunities for root research, with future developments targeting multi-crop adaptability and complex soil condition applications through modular hardware redesign and 3D reconstruction algorithm integration.
Why it matches plant phenotyping methods根系観察用ハードウェアと画像セグメンテーション手法を開発し、根形質抽出性能を検証した、中心的な植物フェノタイピング研究である。
abstractThis manuscript developed a cost-effective hybrid imaging system
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits the study's soybean root image data (the time-series NRMS dataset and scanner validation dataset used for phenotyping) in a public GitHub repository under the authors' account, matching an allowed URL. No separate analysis code availability is stated, so the资产Dataset · publicData Availability Statement: The data presented in this study are openly available in [GitHub] at
[https://github.com/xusiyue/RootPO_DBSCAN/tree/master/project_rootSystem/data (accessed
on 31 October 2025)].Open asset ↗GitHub · xusiyue/RootPO_DBSCANpdf-page:18 lines:1-58Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 6 Sept 2026
Abstract Protein-rich leguminous plants, such as faba bean and white clover are prospectively interesting crops in the North-European countries for reducing dependence on soybean import. Significant expansion of the production area of leguminous crops is challenged by the sub-optimal climatic conditions in this region, especially by the increasing probability of year-to-year fluctuation of extreme weather conditions due to global climate change. To overcome these challenges, development of new climate-resilient varieties suitable for growing under Northern-European conditions are needed. Root architecture and early root development, as well as the availability of efficient root phenotyping technologies are crucial factors of advancing in breeding of adequate varieties. We report a study of a simple and affordable screening technology of early root development using rhizoboxes in connection with semi-automated image analysis and provide a conceptual pipeline for estimation of Genomic Estimated Breeding Values (GEBVs) and correlating greenhouse and field phenotype data. Based on bivariate models, high genetic correlation (r=0.83) could be detected between total root length values recorded in greenhouse rhizobox experiments and field grain yield in faba bean. In white clover, moderately positive genetic correlation (r=0.17) between estimated breeding values of rhizobox-detected total root length and field yield could be identified. Our results suggest that phenotyping and selection of early root development components could potentially be useful in breeding programs to increase the genetic gain for field yield.
Why it matches plant phenotyping methods根系形態を対象に、rhizoboxと半自動画像解析による早期根発達の表現型取得技術を提示し、育種価推定へのパイプラインも示しているため、フェノタイピング手法が中心的である。
abstractWe report a study of a simple and affordable screening technology of early root development using rhizoboxes in connection with semi-automated image analysis and provide a conceptual pipeline for estimation of Genomic Estimated Breeding Values (GEBVs) and correlating greenhouse and field phenotype data.
The accurate prediction of plant height is crucial for optimizing soybean cultivar selection and improving yield estimations. In this study, we investigate the potential of machine learning (ML) algorithms to predict soybean plant height (PH) based on a diverse set of agronomic parameters analyzed from forty soybean cultivars evaluated across sequential harvests. Using a comprehensive dataset, the models Elastic Net (EN), Extra Trees (ET), Gaussian Process Regressor (GPR), K-Nearest Neighbors, and XGBoost (XGB) were compared in terms of predictive accuracy, uncertainty, and robustness. Our results demonstrate that ET outperformed other models with an average correlation coefficient of 0.674, R2 of 0.426 and the lowest RMSE of 6.859 cm and MAE of 5.361 cm, while also showing the lowest uncertainty (5.07%). The proposed ML framework includes an extensive model evaluation pipeline that incorporates the Performance Index (PI), ANOVA, and feature importance analysis, providing a multidimensional perspective on model behavior. The most influential features for PH prediction were the number of stems (NS) and insertion of the first pod (IFP). This research highlights the viability of integrating explainable ML techniques into agricultural decision support systems, enabling data-driven strategies for cultivar evaluation and phenotypic trait forecasting.
Why it matches plant phenotyping methods大豆の草丈という植物形質を予測する機械学習フレームワークを提案し、複数モデルの精度・不確実性・頑健性を比較評価しており、形質推定手法が研究の中心である。
abstractThe proposed ML framework includes an extensive model evaluation pipeline that incorporates the Performance Index (PI), ANOVA, and feature importance analysis, providing a multidimensional perspective on model behavior.
SoybeanLiDAR / point cloudRootMorphology / geometry measurement2D/3D reconstructionRoot system architecture
Abstract Accurate phenotyping of root traits is essential for understanding how plants respond to varying soil water treatment conditions, yet traditional phenotyping methods are often destructive and limited in capturing the full three‐dimensional (3D) complexity of root systems. Existing two‐dimensional imaging techniques and advanced 3D methods for performing root phenotyping, like magnetic resonance imaging or computed tomography, either compromise on resolution, are cost‐prohibitive, or lack scalability. To address these limitations, this study proposes fringe projection profilometry (FPP), a rapid, nondestructive 3D imaging method, for root phenotyping. Using FPP, two architectural root traits were extracted: the number of root tips and the volumetric occupancy of the root system. These traits, difficult to obtain through manual phenotyping or conventional imaging, were automatically derived from the FPP 3D point clouds and validated against expert‐assigned fibrosity scores serving as the biological reference. The study involved 36 soybean ( Glycine max (L.) Merr.) plants from six genotypes, pre‐classified as either stress‐treated or grown under rain‐fed conditions. Results showed strong alignment between FPP‐derived traits and expert evaluations. Stress‐ treated plants consistently exhibited more root tips and greater volumetric occupancy, confirming the biological relevance of these metrics. While this study does not attempt to classify drought tolerance directly, the structural variations observed under drought stress may serve as a foundation for identifying stress‐responsive phenotypes in future work. Overall, the findings demonstrate that FPP provides a fast, scalable, and accurate tool for 3D root phenotyping under variable water conditions.
Why it matches plant phenotyping methodsFPPによる根系の3次元形質取得・自動抽出を開発し、専門家評価と検証した研究であり、フェノタイピング手法が中心です。
abstractthis study proposes fringe projection profilometry (FPP), a rapid, nondestructive 3D imaging method, for root phenotyping.
Reproduction assets foundThe paper's Data Availability Statement provides a public Google Drive link to the datasets generated and/or analyzed in this soybean root FPP phenotyping study, which is an allowed URL. No author analysis code is explicitly deposited.Dataset · publicying and Overcoming Weaknesses via Breed-
ing, Genomics, Phenomics and Physiology).
C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare no conflicts of interest.
DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated and/or analyzed dur-
ing the current research are available at Google
Drive link: https://drive.google.com/file/d/1BJ4yq8QEWY3E5qQIQmYcOXHhEYn1zTE-
/view?usp=sharing
O RC I D
JiaqiongLi https://orcid.org/0009-0006-2247-425X
ZengluLi https://orcid.org/0000-0003-4114-9509
BeiwenLi https://orcid.org/0000-0001-8130-7730
R E F E R E N C E S
Balasubramaniam, B., Li, J., Liu, L., & Li, B. (2023). 3D imaging with
fringe projection for food and agriculturalOpen asset ↗pdf-raw-page:17 lines:1-91Plant phenotyping relevance match · UnverifiedCrossref · checked 14 Sept 2026
• A UAV-based novel phenotyping pipeline using multispectral imaging and LASSO regression accurately predicts soybean traits and yield across growth stages by selecting key vegetation indices. • Red-edge and NIR indices best predict plant height, stomatal conductance, and yield. • Chlorophyll-related indices were effective for estimating LAI and leaf chlorophyll. • Best aerial phenotyping time is between pod development and the full seed stage. Morphophysiological parameters, such as plant height, leaf chlorophyll content, stomatal conductance, and leaf area index, are key indicators of soybean ( Glycine max (L.) Merril) yield potential. Traditional in situ methods for assessing these traits, while accurate in small areas, are slow, labor-intensive, and impractical for large-scale monitoring. Similarly, extrapolating yield from manual counts of plant stands, pods, and seeds per pod may provide unreliable results. Therefore, high throughput sensor-based approaches are becoming increasingly popular to efficiently quantify these traits and predict yield. Among various remote sensing sensors, multispectral provides information in the red, green, red-edge, and near-infrared bands, which are critical for studying plant growth and vegetation health by combining multiple spectral bands. While many studies have used vegetation indices (VIs) to estimate individual traits, fewer have predicted multiple traits and yield at the same time using multispectral data. Thus, a study was conducted to identify the most effective VIs and determine the optimal timing for aerial phenotyping using multispectral sensors and LASSO regression. The study suggested Red-edge and NIR-based indices were best for predicting plant height, stomatal conductance, and yield, while chlorophyll-related indices were more effective for LAI and chlorophyll content. The pod development to full seed stages was the best time for aerial phenotyping. Overall, UAV-derived MS images combined with LASSO regression proved to be a practical and efficient approach for large-scale soybean phenotyping and yield monitoring. This study supports precision agriculture by providing a remote sensing-based, rapid, and non-destructive method for assessing crop status.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像とLASSO回帰による複数の植物形質・収量の推定手法を開発・評価しており、フェノタイピング手法が研究の中心である。
abstractA UAV-based novel phenotyping pipeline using multispectral imaging and LASSO regression accurately predicts soybean traits and yield across growth stages by selecting key vegetation indices.
SoybeanLiDAR / point cloudRoot2D/3D reconstructionSegmentationSkeletonization / topologyRoot system architecture
Characterizing root system architecture (RSA) is essential for understanding plant acclimatization and guiding breeding strategies to enhance stress tolerance and optimize resource uptake. Although 3D root analysis provides significantly more detailed and structurally informative insights than conventional 2D methods, the development of robust and quantitative tools for 3D root phenotyping has been hindered by challenges such as data complexity, noise, and root overlap. In this study, we present a biologically inspired skeletonization framework that segments root architectures by tracing root growth trajectories. The primary objective is to enable anatomically accurate extraction of RSA traits from 3D point clouds. Our method begins by segmenting the primary root through shortest-path extraction and tangent-plane-based clustering. Lateral root initiation points are then detected, and candidate paths are grown using a bionic pathfinding strategy with adaptive parameters; an optimal, non-overlapping skeleton is selected through clustering and combination sorting, and finally refined via an inward back-tracing procedure to improve junction connectivity. To support downstream phenotyping, we compute root length and angle from the segmented skeletons, and reconstruct anatomically faithful tubular meshes for each lateral root to analytically estimate surface area and volume. Our method achieved high accuracy across multiple traits, including an F1 score of 0.88 for lateral root numeration, R2 values of 0.992 and 0.987 for primary and lateral root length estimation, respectively, and strong agreement in surface area (R2=0.953) and volume (R2=0.912) validation against reference methods. Overall, our method offers a robust and biologically meaningful solution for 3D root phenotyping. The extracted traits provide plant breeders with critical insights for genotype selection and offer plant scientists a powerful tool to evaluate the effects of agronomic treatments and environmental interventions.
Why it matches plant phenotyping methods3D根系骨架化と形態形質抽出法の開発・検証が研究の中心であり、根長・角度・表面積・体積などの表現型を定量化している。
abstractwe present a biologically inspired skeletonization framework that segments root architectures by tracing root growth trajectories
The accurate identification of soybean growth stages is critical for optimizing agricultural interventions, where mistimed treatments can result in yield losses ranging from 2.5% to 40%. Existing deep learning approaches remain limited in scope, targeting isolated developmental phases rather than providing comprehensive phenological coverage. This paper presents a novel object detection architecture DELTA-SoyStage, combining an EfficientNet backbone with a lightweight ChannelMapper neck and a newly proposed DELTA (Denoising Enhanced Lightweight Task Alignment) detection head for soybean growth stage classification. We introduce a dataset of 17,204 labeled RGB images spanning nine growth stages from emergence (VE) through full maturity (R8), collected under controlled greenhouse conditions with diverse imaging angles and lighting variations. DELTA-SoyStage achieves 73.9% average precision with only 24.4 GFLOPs computational cost, demonstrating 4.2× fewer FLOPs than the best-performing baseline (DINO-Swin: 74.7% AP, 102.5 GFLOPs) with only 0.8% accuracy difference. The lightweight DELTA head combined with the efficient ChannelMapper neck requires only 8.3 M parameters-a 43.5% reduction compared to standard architectures-while maintaining competitive accuracy. Extensive ablation studies validate key design choices including task alignment mechanisms, multi-scale feature extraction strategies, and encoder-decoder depth configurations. The proposed model's computational efficiency makes it suitable for deployment on resource-constrained edge devices in precision agriculture applications, enabling timely decision-making without reliance on cloud infrastructure.
Why it matches plant phenotyping methods大豆の生育ステージという植物状態をRGB画像から推定する検出アーキテクチャを開発し、データセット、比較評価、アブレーション検証まで行っており、植物フェノタイピング手法が中心である。
abstractThis paper presents a novel object detection architecture DELTA-SoyStage, combining an EfficientNet backbone with a lightweight ChannelMapper neck and a newly proposed DELTA (Denoising Enhanced Lightweight Task Alignment) detection head for soybean growth stage classification.
Herbicides play a crucial role in cropping systems by providing effective weed control strategies that help farmers eliminate yield-reducing weeds. However, crop injury may result from herbicides applied in current or previous cropping systems, and in some instances, this injury may reduce crop yield. Currently, herbicide related crop injury is commonly determined by subjective visual assessments. Spectral imaging provides an alternative solution, which is high-throughput and non-invasive. In this study, a novel machine vision method utilizing hyperspectral imaging (HSI) and multispectral imaging (MSI) was developed and integrated into Colby’s method—a traditional approach in weed science for analyzing the interaction effects of herbicide mixtures. Mesotrione and diflufenican, both herbicides that cause bleaching symptomology, were applied in this study. Two rounds of field experiments were conducted in the summer of 2024, where hyperspectral and multispectral images were collected 26 DAT in each trial. Partial Least Squares Discriminant Analysis (PLS-DA) models were built to identify soybean injury from mesotrione, diflufenican, and the mixture. For Colby’s method to study the interaction effect, spatial-spectral features were generated from MSI. The HSI models achieved an accuracy exceeding 90 %. Thirteen distinct features were identified and selected to illustrate the synergistic effects of the herbicides, showing consistency across two experimental rounds and aligning with findings from traditional methods.
Why it matches plant phenotyping methods除草剤によるダイズ傷害という植物状態を、HSI/MSIと機械学習で客観的・高スループットに推定する手法を開発し、実験間で検証しているため、植物フェノタイピング手法が中心です。
abstractSpectral imaging provides an alternative solution, which is high-throughput and non-invasive.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 14 Sept 2026
Abstract Sustainable improvement of crop performance requires integrative approaches that link genomic variation to phenotypic expression through intermediate molecular layers. Here, we present Reciprocal Best Linear Unbiased Prediction (Reciprocal BLUP), a predictability-guided multi-omics framework that quantifies cross-layer relationships among the genome, metabolome, and microbiome to enhance phenotype prediction. Using a panel of 198 soybean accessions grown under well-watered and drought conditions, we first evaluated four direction-specific prediction models (genome→microbiome, genome→metabolome, metabolome→microbiome, and microbiome→metabolome) to estimate the predictability of individual omics features. Then, we evaluated whether subsets of features with high cross-omics predictability improve phenotype prediction. These cross-layer models identify features that play physiologically meaningful roles within multi-omics systems, enabling us to prioritize variables that capture coherent biological signals enriched for phenotype-relevant information. As a result, metabolome features were highly predictable from microbiome data, whereas microbiome predictability from metabolomic data was weaker and more environment-dependent, revealing an asymmetric relationship between these layers. In the subsequent phenotype prediction analysis, the model incorporating predictability-based feature selection substantially outperformed models using randomly selected features and also achieved prediction accuracies comparable to those of the full-feature model. Under drought, phenotype prediction models based on metabolomic or microbiomic kernels (MetBLUP or MicroBLUP) outperformed the genomic baseline (GBLUP) for several biomass-related traits, indicating that environment-responsive omics layers captured phenotypic variation not explained by additive genetic effects. Our results highlight hierarchical interactions among genomic, metabolic, and microbial systems, with the metabolome functioning as an integrative mediator linking genotype, environment, and microbiome composition. The Reciprocal BLUP framework provides a biologically interpretable and practical approach for integrating multi-omics data, improving phenotype prediction, and guiding omics-based feature selection in plant breeding.
Why it matches plant phenotyping methods植物表現型予測のためのマルチオミクス計算フレームワークを提案・評価しており、表現型推定手法が研究の中心である。
abstractwe present Reciprocal Best Linear Unbiased Prediction (Reciprocal BLUP), a predictability-guided multi-omics framework that quantifies cross-layer relationships among the genome, metabolome, and microbiome to enhance phenotype prediction.
Feeding a growing global population requires that we must try to increase production and to reliably predict yields before a harvest. Since one anticipates farm-level yields, one enables better resource management, market planning, and sustainable agricultural decisions. The objective of this study is to develop also evaluate a deep learning regression framework because it predicts the yields of wheat, corn, as well as soybean farms using multi-temporal Landsat-5 and Landsat-7 imagery together with annual ground-truth yield records from the Kellogg Biological Station Long-Term Ecological Research (KBS LTER) site in Michigan, United States. The dataset spans across 11 cropping years (2001, 2012, with 2002 excluded) and the dataset covers 24 farms. There were two models: one had training directly on seven spectral bands, and one trained on vegetation indices (VIs) (NDVI, SAVI, EVI2, GRNDVI). A 2×2-pixel window data augmentation strategy was used to address the limited sample size and farm-level yields were then reconstructed via weighted aggregation of window-level predictions. The band-based model did achieve a higher accuracy of about 89.44%. That figure is superior to the result of 87.22% for the VI model. Wheat yields were most accurately predicted (88.3%) when crops were assessed. Soybean (87.01%) then corn (85.08%) followed this result. This study provides an effective and reproducible framework for farm-level yield prediction under limited data conditions with a fully connected deep learning model, combined with systematic window-based augmentation and weighted yield reconstruction. Landsat imagery with its single-site scope and its 30 m resolution did constrain the framework yet it highlights the potential of combining deep learning with optimisation principles that are regression-based. The framework offers too a scalable basis for integration with multi-source datasets as well as decision-support systems in precision agriculture.
Why it matches plant phenotyping methods衛星画像から農場レベルの作物収量を推定する深層学習フレームワークを開発・評価しており、収量という植物形質の取得・推定手法が研究の中心である。
abstractThe objective of this study is to develop also evaluate a deep learning regression framework because it predicts the yields of wheat, corn, as well as soybean farms using multi-temporal Landsat-5 and Landsat-7 imagery
Introduction The Leaf Area Index (LAI) is a critical biophysical parameter for assessing crop canopy structure and health. Unmanned Aerial Vehicles (UAVs) equipped with multispectral sensors offer a high-throughput solution for LAI estimation, but flight altitude compromises between efficiency and image resolution, ultimately impacting accuracy. This study investigates the integration of super-resolution (SR) image reconstruction with multi-sensor data to enhance LAI estimation for soybeans across varying UAV flight altitudes. Methods RGB and multispectral images were captured at four flight altitudes: 15 m, 30 m, 45 m, and 60 m. The acquired images were processed using several SR algorithms (SwinIR, Real-ESRGAN, SRCNN, and EDSR). Texture features were extracted from the RGB images, and LAI estimation models were developed using the XGBoost algorithm, testing data fusion strategies that included RGB-only, multispectral-only, and a combined RGB-multispectral approach. Results (1) SR performance declined with increasing altitude, with SwinIR achieving superior image reconstruction quality (PSNR and SSIM) over other methods. (2) Texture features from RGB images showed strong sensitivity to LAI. The XGBoost model leveraging fused RGB and multispectral data achieved the highest accuracy (relative error: 4.16%), outperforming models using only RGB (5.25%) or only multispectral data (9.17%). (3) The application of SR techniques significantly improved model accuracy at 30 m and 45 m altitudes. At 30 m, models incorporating Real-ESRGAN and SwinIR achieved an average R 2 of 0.86, while at 45 m, these methods yielded models with an average R 2 of 0.77. Discussion The results demonstrate that the fusion of SR-reconstructed imagery with multi-sensor data can effectively mitigate the negative impact of higher flight altitudes on LAI estimation accuracy. This approach provides a robust and efficient framework for UAV-based crop monitoring, enhancing data-driven decision-making in precision agriculture.
Why it matches plant phenotyping methodsUAV画像の超解像・マルチセンサー融合・機械学習を用いて、植物キャノピー形質であるダイズLAIの推定手法を開発・評価しており、フェノタイピング手法が中心です。
abstractThis study investigates the integration of super-resolution (SR) image reconstruction with multi-sensor data to enhance LAI estimation for soybeans across varying UAV flight altitudes.
Background: The purpose of this project is to evaluate the prospective use of drone-based remote sensing for assessing legume crop development and predicting their yields. Traditional agricultural procedures often fall short in delivering timely and accurate monitoring, necessitating the adoption of innovative techniques. Methods: The study considers vegetative indicators such as NDVI, GNDVI and canopy cover to track the growth of three legume crops-peanut, soybean and common bean. Machine learning models, including random forest, support vector machines and multiple linear regression, were developed to predict agricultural production using remote sensing data. Statistical analysis was performed to verify the trustworthiness of vegetation indicators against ground-truth measurements. Result: The models achieved high accuracy, with R² values reaching up to 0.92. Statistical analysis confirmed strong relationships between vegetation indicators and ground-truth data. Among the studied crops, soybeans exhibited the highest growth vigor and yield. The study demonstrates that integrating machine learning with drone photography can enhance precision agriculture, making it more scalable and sustainable. Future research is recommended to explore different crop varieties and environmental conditions to further optimize the application of these technologies.
Why it matches plant phenotyping methodsドローンリモートセンシングと機械学習を用いて作物生育指標および収量を推定し、地上実測値で検証することが研究の中心である。
abstractThe purpose of this project is to evaluate the prospective use of drone-based remote sensing for assessing legume crop development and predicting their yields.
The widespread use of rare earth elements (REEs) has led to their accumulation in crops, threatening plant health and agricultural productivity. Early detection of hidden stress and subvisible plant damage is therefore crucial for sustainable agriculture. In this study, we systematically investigated the impact of lanthanum [La(III)], a representative REE, on plant health using soybean, a major economic crop, as a model system. A dual-component living-cell biosensor based on soybean leaves was developed to monitor in real-time the changes in electron transfer impedance (R et ) associated with two La(III)-binding proteins: extracellular vitronectin-like protein (VN), and plasma membrane-anchored arabinogalactan proteins (AGPs). When soybean leaves were exposed to 40-100 μM La(III), significant subvisible cellular damage occurred, despite the absence of visible symptoms, indicating a state of hidden stress. This was evidenced by a 115.9 %-259.68 % increase in malondialdehyde levels and a 2.39 %-26.27 % decrease in chlorophyll content. Concurrently, the dual biosensor detected a 74.34 %-151.98 % increase in R et for VN and a 47.32 %-104.68 % increase for AGPs, demonstrating its ability to sensitively capture early physiological alterations. At La(III) concentrations exceeding 100 μM, visible leaf damage emerged, covering 31 % of the leaf surface, accompanied by a 306.57 % increase in malondialdehyde content and a 34.75 % reduction in chlorophyll levels. Under these conditions, R et increased by 155.43 % for VN and 108.77 % for AGPs. These results indicate that the biosensor enables early, sensitive detection of REE-induced hidden stress in crops before visible symptoms occur, offering a promising tool for proactive agricultural monitoring and sustainable food security.
Why it matches plant phenotyping methodsダイズ葉を用いたデュアルバイオセンサーを開発し、可視症状前のストレス状態・生理変化をリアルタイム検出する方法が研究の中心であるため、植物フェノタイピング手法に該当する。
abstractA dual-component living-cell biosensor based on soybean leaves was developed to monitor in real-time the changes in electron transfer impedance (R et ) associated with two La(III)-binding proteins
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 6 Sept 2026
Abstract The stink bug complex is one of the most damaging pests of soybean, reducing yield and seed quality. Genetic resistance remains the most sustainable and effective management strategy, but its quantitative inheritance and labor-intensive field phenotyping make its implementation in breeding programs challenging. This study explored high-throughput phenotyping (HTP) using unmanned aerial vehicles (UAVs) equipped with RGB cameras to evaluate a soybean population and the potential of phenotyping to stink bug resistance by correlating image-derived features and machine learning (ML) models. A population of 304 soybean lines was evaluated in alpha-lattice design trials across two seasons under natural infestations. Five resistance-related traits, grain yield (GY), hundred-seed weight (HSW), number of days to maturity (NDM), tolerance (TOL), and leaf retention (LR), were manually scored and linked to UAV-derived vegetation indices (VIs) and texture indices (TIs). Three ML models (AdaBoost, SVM, MLP) were tested to predict these traits from aerial features. Results showed that VIs, particularly Visible Atmospherically Resistant Index at the first percentile (VARI_P25), were consistently associated with resistance-related traits, while decision tree analysis highlighted TIs at 45° and 135° as complementary sources of structural information. Prediction ability was highest for GY, HSW, and NDM, especially in flights near flowering and maturity, but remained low for TOL and LR. Integrating multiple flights modestly improved accuracy, whereas cross-season predictions were unreliable. Nonetheless, indices such as VARI_P25 provided useful cross-season correlations for HSW and TOL, enabling early screening of less promising lines. This pioneering study demonstrates that UAV–ML pipelines can capture genetic signals of stink bug resistance in soybean, despite environmental complexity. These findings open new avenues for resistance phenotyping, supporting more efficient breeding strategies and accelerating genetic gains in soybean improvement.
Why it matches plant phenotyping methodsUAV画像と機械学習による形質推定パイプラインを開発・評価し、抵抗性関連形質や収量を対象とするフェノタイピングが研究の中心である。
abstractThis study explored high-throughput phenotyping (HTP) using unmanned aerial vehicles (UAVs) equipped with RGB cameras to evaluate a soybean population and the potential of phenotyping to stink bug resistance by correlating image-derived features and machine learning (ML) models.
Background: Legumes play an important role in improving soil quality and nutrition in humans. Soybean is one of the legumes which are rich in protein and oil content. Nutrients deficiency in soybean plant could affect the growth of the plants and might lead to loss in its yield. The developments in modern technologies such as computer vision and deep learning are being leveraged in identifying nutrient deficiencies in soybean plants. Methods: Convolutional neural network with six feature extraction blocks and one classification block is developed to identify macro nutrient deficiency in soybean plants. Images are first collected, pre-processed and labeled. They are then split into training images and testing images in the ratio of 80:20. The proposed convolutional neural network model is trained over training images and the final trained model is tested with testing images. Result: In detecting the deficiency of macro nutrients such as nitrogen, phosphorus and potassium in soybean plants, the testing results of the proposed convolutional neural network architecture achieved an accuracy of 97.43%. Accuracy comparison against existing models such as VGG16, ResNet50 and MobileNetV3 demonstrate that the proposed model effectively identifies nutrient deficiencies in soybean plants. Thus, the proposed system is designed to support farmers in making timely decisions and to contribute to food security by leveraging deep learning techniques.
Why it matches plant phenotyping methods大豆葉画像から栄養欠乏状態を推定するCNNを開発・評価しており、植物状態の取得・分類手法が研究の中心であるため含める。
abstractConvolutional neural network with six feature extraction blocks and one classification block is developed to identify macro nutrient deficiency in soybean plants.
Precision agriculture technologies based on satellite remote sensing remain largely inaccessible to smallholder farmers in developing countries due to technical complexity, cost barriers, and infrastructure demands. This study presents the design and implementation of an open-source, web-based platform for processing Sentinel-2 Level-2A imagery tailored to the specific needs of family farming systems. The platform integrates a FastAPI backend for geospatial data processing with a Next.js frontend providing simplified tools for spectral index computation (NDVI, EVI, SAVI, NDWI, NDBI), crop classification using supervised and unsupervised machine learning, and interactive 2D/3D visualization. A laboratory module implements thirteen digital image processing techniques—including Gaussian filtering, edge detection, morphological operations, and thresholding—for educational and comparative analysis. The browser-based system eliminates installation requirements and automates key workflows such as coordinate reprojection, JP2 band extraction, and statistical evaluation. Validation using ground-truth data from coffee and soybean fields in the Brazilian Cerrado achieved classification accuracies above 85% and correlation coefficients exceeding 0.90 for biomass estimation based on NDVI-derived metrics. The platform contributes to the democratization of remote sensing technologies and enhances accessibility of precision agriculture tools for smallholder farmers.
Why it matches plant phenotyping methods植物圃場の衛星画像を処理し、NDVI等からバイオマスを推定するオープンソース基盤の設計・実装・検証が中心であり、植物形質推定ワークフローとして収録対象。
titleAn Open-Source Web Platform for Sentinel-2 Multispectral Analysis in Smallholder Agriculture: Design, Implementation and Validation
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits the authors' complete source code, documentation, and example datasets for the Sentinel-2 phenotyping/analysis platform on a public GitHub repository under MIT license. Sentinel-2 imagery is from the public Copernicus browser, but that is a generic data sourceCode · publicresearch received no external funding
Institutional Review Board Statement: Not applicable. This study did not involve humans or animals.
Informed Consent Statement: Not applicable. This study did not involve humans.
Data Availability Statement: Complete source code, documentation, and example datasets are publicly available
at https://github.com/rexionmars/icev-remote-sensing under MIT license. The platform can be deployed locally
or accessed via hosted instance for testing purposes. Sentinel-2 satellite imagery used in this study was obtained
from the Copernicus Open Access Hub (https://browser.dataspace.copernicus.eu/) and is freely available.
Acknowledgments: The authors thank the iCEV IOpen asset ↗https://github.com/rexionmars/icev-remote-sensing · icev-remote-sensingpdf-layout-page:11 lines:1-70Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Abstract Background and Aims: Portable X-ray fluorescence spectrometry (pXRF) has emerged as a robust analytical approach for elemental determination in plant tissues, enabling rapid, non-destructive, and reagent-free measurements. This study developed and validated an empirical calibration of pXRF for quantifying silicon (Si) in plants, using autoclave-induced digestion (AID) as the reference method. Methods A total of 374 samples from seven plant species (rice, maize, soybean, cowpea, sorghum, lettuce, and beet) were analyzed. Silicon concentrations obtained via AID ranged from 1.07 to 19.23 g kg − ¹ (mean = 4.48 g kg − ¹; coefficient of variation = 67%), reflecting substantial interspecific variability. Each sample was also analyzed by pXRF under optimized instrumental conditions, and a calibration model was constructed using 75% of the dataset to predict Si concentrations relative to AID values. Results The pXRF calibration exhibited a strong linear relationship with AID results (R² = 0.94; R = 0.97; p
Why it matches plant phenotyping methods植物組織中のケイ素濃度を測定するpXRF法の開発と、基準法との校正・検証が研究の中心であり、植物形質の測定法に該当する。
abstractThis study developed and validated an empirical calibration of pXRF for quantifying silicon (Si) in plants, using autoclave-induced digestion (AID) as the reference method.
• High-throughput phenotyping of 202 soybean accessions across two growth stages identified key shade tolerance indicators. • A two-stage framework assessed symbiotic shade tolerance, recovery ability, and overall performance. • Machine learning–based trait mining was validated across multiple environments. Intercropping is a promising cultivation strategy that enhances the sustainable use of water and land resources while contributing to national food and oil security. To improve the yield stability of soybeans in intercropping systems, there is an urgent need to develop a scientific and efficient framework for evaluating shade tolerance. In this study, we propose an integrated shade tolerance assessment method based on high-throughput phenotyping, multienvironment trials, and machine learning (ML) approaches. Utilizing multivariate analysis, we evaluated 202 soybean accessions and partitioned their performance under intercropping into two distinct capacities, namely, shade tolerance during the cogrowth stage and recovery ability during the independent growth stage, each of which was classified into five levels from weak to strong. Preliminary trait selection was performed through correlation analysis and broad-sense heritability estimation, followed by the application of six ML models to identify the key shade tolerance traits across different growth stages. The robustness and generalizability of the selected traits were validated in three environments—a field pot, an open field, and a greenhouse—using soybean varieties with known shade tolerance levels. The results revealed that three traits—the side canopy area (SCA), top canopy area at stage 3 (TCA3), and top-view mixed entropy (TME)—were strongly associated with shade-tolerant varieties. These traits presented two distinguishing features: significantly higher values under shaded conditions and greater increases during the recovery phase. The prediction models constructed with these three traits achieved strong performance, with coefficients of determination of R ²=0.776 for shade tolerance and R ²=0.959 for recovery ability. In summary, this study demonstrates the potential for integrating high-throughput phenotyping with ML to efficiently identify the key indicators of shade tolerance. By measuring only three indicators—SCA, TCA3, and TME—soybean shade tolerance at the seedling stage, recovery ability during later growth, and overall shade tolerance across the full growth period can be rapidly and accurately evaluated. This method offers a powerful and practical tool for implementing shade tolerance evaluations, gene discovery, and targeted breeding of soybean cultivars that are suitable for intercropping systems.
Why it matches plant phenotyping methods高スループット表現型解析と機械学習を統合し、画像由来の植物形質からダイズの遮陰耐性・回復能を評価する方法を提案・検証しており、表現型取得と抽出が研究の中心である。
abstractwe propose an integrated shade tolerance assessment method based on high-throughput phenotyping, multienvironment trials, and machine learning (ML) approaches.
We address the challenge that occlusions in on-branch soybean images impede accurate pod-level phenotyping. We propose a lab on-branch pipeline that couples a prior-guided synthetic data generator (producing synchronized visible and amodal labels) with an amodal instance segmentation framework based on an improved Swin Transformer backbone with a Simple Attention Module (SimAM) and dual heads, trained via three-stage transfer (synthetic excised → synthetic on-branch → few-shot real). Guided by complete (amodal) masks, a morphology-driven module performs pose normalization, axial geometric modeling, multi-scale fused density mapping, marker-controlled watershed, and topological consistency refinement to extract seed per pod (SPP) and geometric traits. On real on-branch data, the model attains Visible Average Precision (AP) 50/75 of 91.6/77.6 and amodal AP50/75 of 90.1/74.7, and incorporating synthetic data yields consistent gains across models, indicating effective occlusion reasoning. On excised pod tests, SPP achieves a mean absolute error (MAE) of 0.07 and a root mean square error (RMSE) of 0.26; pod length/width achieves an MAE of 2.87/3.18 px with high agreement (R 2 up to 0.94). Overall, the co-designed data-model-task pipeline recovers complete pod geometry under heavy occlusion and enables non-destructive, high-precision, and low-annotation-cost extraction of key traits, providing a practical basis for standardized laboratory phenotyping and downstream breeding applications.
Why it matches plant phenotyping methods大豆莢の遮蔽下形状を画像から復元し、種子数や幾何形質を抽出する画像解析パイプラインを開発・評価しており、植物表現型取得が中心である。
abstractWe propose a lab on-branch pipeline that couples a prior-guided synthetic data generator (producing synchronized visible and amodal labels) with an amodal instance segmentation framework
Learning 3D parametric shape models of objects has gained popularity in vision and graphics and has showed broad utility in 3D reconstruction, generation, understanding, and simulation. While powerful models exist for humans and animals, equally expressive approaches for modeling plants are lacking. In this work, we present Demeter, a data-driven parametric model that encodes key factors of a plant morphology, including topology, shape, articulation, and deformation into a compact learned representation. Unlike previous parametric models, Demeter handles varying shape topology across various species and models three sources of shape variation: articulation, subcomponent shape variation, and non-rigid deformation. To advance crop plant modeling, we collected a large-scale, ground-truthed dataset from a soybean farm as a testbed. Experiments show that Demeter effectively synthesizes shapes, reconstructs structures, and simulates biophysical processes. Code and data is available at https://tianhang-cheng.github.io/Demeter/.
Why it matches plant phenotyping methods植物形態のトポロジー・形状・関節・変形を表現する3Dパラメトリックモデルを開発し、実世界の作物データセットで再構成・シミュレーションを評価しているため、形態フェノタイピング手法が中心である。
abstractwe present Demeter, a data-driven parametric model that encodes key factors of a plant morphology, including topology, shape, articulation, and deformation into a compact learned representation.
Charcoal rot of soybean caused by Macrophomina phaseolina is a major disease of economic significance around the world. The objective of this test is to evaluate and identify new method(s) that can identify resistant and susceptible (S) genotypes when inoculated with the fungus that causes charcoal rot in nonfield environments. Four independent experiments were performed to determine the variability in disease severity when soybean genotypes are inoculated with isolates up to a total of 100 different variants of the charcoal rot fungus in laboratory, greenhouse, and growth chamber tests. Linear mixed models were fit to area under the disease progress curve values from the four experiments and model predictions of disease progress were used to determine the best method to classify moderately resistant (MR) and S genotypes. In a growth chamber study using a modified cut-tip inoculation method, 28 of the 32 M. phaseolina isolates tested differentiated MR and S with >87% accuracy. In a study where 16 field-grown soybean genotypes were stem-wound inoculated with one isolate, the MR genotypes were correctly classified, but not all S genotypes were. Correct classification of MR genotypes dramatically increased with plant age, approaching 100% accuracy at 120 days after planting. In a study of stem-wound inoculation of field-grown soybean genotypes with 20 M. phaseolina isolates, MR were identified with more than half the isolates having >75% accuracy in detecting MR genotypes. In a study of greenhouse-grown soybeans stem-wound-inoculated with 100 isolates, classification was less accurate than samples grown in the field, with median correct classification P = 0.0006), and the rank correlations with the growth chamber study were weak. The results showed that, except for the growth chamber study, the nonfield environments did not consistently identify the same soybean lines as being S or MR to charcoal rot as were identified as in naturally infested field testing because of differences in isolates, environment, soybean varieties, and methods (or all the above). Stakeholders will benefit more from the use of the field assessment method in naturally infested soil to identify reliable sources of resistance than from the nonfield methods.
Why it matches plant phenotyping methodsダイズの炭腐病重症度を用いて、接種法・栽培環境・分離株による抵抗性判別法を比較検証しており、植物病害表現型の取得と分類性能が研究の中心である。
abstractThe objective of this test is to evaluate and identify new method(s) that can identify resistant and susceptible (S) genotypes when inoculated with the fungus that causes charcoal rot in nonfield environments.
Introduction The rapid growth of the global population and intensive agricultural activities has posed serious environmental challenges. In response, there is an increasing demand for sustainable agricultural solutions that ensure efficient resource utilization while maintaining ecological balance. Among these, intercropping has gained prominence as a viable method, promoting enhanced land use efficiency and fostering environment for crop development. However, disease management in intercropping systems remains complex due to the potential for cross-infection and overlapping disease symptoms among crops. Early and precise illness recognition is, therefore, critical for sustaining crop condition and efficiency. Methods This study introduces an intelligent intercropping framework for early leaf disease detection, utilizing hyperspectral imaging and hybrid deep learning models for precision agriculture. Hyperspectral imaging captures intricate biochemical and structural variations in crops like maize, soybean, pea, and cucumber-subtle markers of disease that are otherwise imperceptible. These images enable accurate identification of diseases such as rust, leaf spot, and complex co-infections. To refine disease region segmentation and improve detection accuracy, the proposed model employs the synergistic swarm optimization (SSO) algorithm. A phase attention fusion network (PANet) is utilized for deep feature extraction, minimizing false detection rates. Furthermore, a dual-stage Kepler optimization (DSKO) algorithm addresses the challenge of high-dimensional data by choosing the most applicable landscapes. The disease classification is performed using a random deep convolutional neural network (R-DCNN). Results and discussion Experimental evaluations were conducted using publicly available hyperspectral datasets for maize-soybean and pea-cucumber intercropping systems. The suggested ideal attained remarkable organization accuracies of 99.676% and 99.538% for the respective intercropping systems, demonstrating its potential as a robust, non-invasive tool for smart, sustainable agriculture.
Why it matches plant phenotyping methods植物の葉の病徴をハイパースペクトル画像から検出・分類する画像解析手法が研究の中心であり、病害状態という植物表現型を直接推定しているため含める。
abstractThis study introduces an intelligent intercropping framework for early leaf disease detection, utilizing hyperspectral imaging and hybrid deep learning models for precision agriculture.
Crop canopy height is a key structural indicator that is strongly associated with crop development, biomass accumulation, and crop health. To overcome the limitations of time-consuming and labor-intensive traditional field measurements, Unmanned Aerial Vehicle (UAV)-based Light Detection and Ranging (LiDAR) offers an efficient alternative by capturing three-dimensional point cloud data (PCD). In this study, UAV-LiDAR data were acquired using a DJI Matrice 600 Pro equipped with a 16-channel LiDAR system. Three canopy height estimation methodological approaches were evaluated across three crop types: corn, soybean, and winter wheat. Specifically, this study assessed machine learning regression modeling, ground point classification techniques, percentile-based method and a newly proposed Dual-Range Averaging (DRA) method to identify the most effective method while ensuring practicality and reproducibility. The best-performing method for corn was Support Vector Regression (SVR) with a linear kernel (R2 = 0.95, RMSE = 0.137 m). For soybean, the DRA method yielded the highest accuracy (R2 = 0.93, RMSE = 0.032 m). For winter wheat, the PointCNN deep learning model demonstrated the best performance (R2 = 0.93, RMSE = 0.046 m). These results highlight the effectiveness of integrating UAV-LiDAR data with optimized processing methods for accurate and widely applicable crop height estimation in support of precision agriculture practices.
Why it matches plant phenotyping methodsUAV-LiDAR点群から作物群落高を推定する複数手法を比較・評価し、新規DRA法も提案しており、植物形質取得手法が研究の中心です。
abstractThree canopy height estimation methodological approaches were evaluated across three crop types: corn, soybean, and winter wheat.
Methods based on upward canopy gap fractions are widely employed to measure in-situ effective LAI (Le) as an alternative to destructive sampling. However, these measurements are limited to point-level and are not practical for scaling up to larger areas. To address the point-to-landscape gap, this study introduces an innovative approach, named NeRF-LAI, for corn and soybean Le estimation that combines gap-fraction theory with the neural radiance field (NeRF) technology, an emerging neural network-based method for implicitly representing 3D scenes using multi-angle 2D images. The trained NeRF-LAI can render downward photorealistic hemispherical depth images from an arbitrary viewpoint in the 3D scene, and then calculate gap fractions to estimate Le. To investigate the intrinsic difference between upward and downward gaps estimations, initial tests on virtual corn fields demonstrated that the downward Le matches well with the upward Le, and the viewpoint height is insensitive to Le estimation for a homogeneous field. Furthermore, we conducted intensive real-world experiments at controlled plots and farmer-managed fields to test the effectiveness and transferability of NeRF-LAI in real-world scenarios, where multi-angle UAV oblique images from different phenological stages were collected for corn and soybeans. Results showed the NeRF-LAI is able to render photorealistic synthetic images with an average peak signal-to-noise ratio (PSNR) of 18.94 for the controlled corn plots and 19.10 for the controlled soybean plots. We further explored three methods to estimate Le from calculated gap fractions: the 57.5° method, the five-ring-based method, and the cell-based method. Among these, the cell-based method achieved the best performance, with the r² ranging from 0.674 to 0.780 and RRMSE ranging from 1.95 % to 5.58 %. The Le estimates are sensitive to viewpoint height in heterogeneous fields due to the difference in the observable foliage volume, but they exhibit less sensitivity to relatively homogeneous fields. Additionally, the cross-site testing for pixel-level LAI mapping showed the NeRF-LAI significantly outperforms the VI-based models, with a small variation of RMSE (0.71 to 0.95 m²/m²) for spatial resolution from 0.5 m to 2.0 m. This study extends the application of gap fraction-based Le estimation from a discrete point scale to a continuous field scale by leveraging implicit 3D neural representations learned by NeRF. The NeRF-LAI method can map Le from raw multi-angle 2D images without prior information, offering a potential alternative to the traditional in-situ plant canopy analyzer with a more flexible and efficient solution.
Why it matches plant phenotyping methodsNeRFとUAV多視点画像を組み合わせ、トウモロコシ・ダイズの葉面積指数を推定・マッピングする手法を開発し、実圃場で性能検証しているため、植物表現型取得が中心である。
abstractthis study introduces an innovative approach, named NeRF-LAI, for corn and soybean Le estimation that combines gap-fraction theory with the neural radiance field (NeRF) technology
Accurate and efficient crop height information retrieval is crucial for applications such as farmland management, growth monitoring, yield estimation, and pest monitoring. Polarimetric Synthetic Aperture Radar (PolSAR) is known for its high sensitivity to the shape, structure, and dielectric constant of vegetation, presenting great potential for crop height retrieval. In this study, we compare the performance of three machine learning algorithms, Random Forest Regression (RFR), Bagging Decision Tree (BAGTREE), and Extreme Gradient Boosting (XGBoost), in the retrieval of crop height from PolSAR data. Using a comprehensive approach, we constructed a set of 32 polarimetric features as the initial input for the model. Subsequently, feature selection is employed to generate a subset aimed at reducing redundancy and improving the final estimation accuracy. Multi-temporal C-band PolSAR RADARSAT-2 data collected over three distinct agricultural types (corn, wheat, and soybean) in Canada are chosen for this study. The results indicate that the optimal average Root Mean Square Error (RMSE) for height retrieval in corn, wheat, and soybean throughout their growth cycles is 43.69 cm, 10.78 cm, and 20.92 cm, respectively. Among the three algorithms, RFR consistently demonstrates stable retrieval performance, and the polarimetric decomposition parameters exhibit the highest sensitivity to crop height. This study offers a valuable technical reference for SAR-based crop height retrieval and remote sensing-based crop growth monitoring without interferometry.
Why it matches plant phenotyping methodsPolSARセンサーデータと機械学習により作物高を推定し、複数アルゴリズムを比較・検証することが研究の中心であるため、植物形質計測手法として含める。
titleCrop height retrieval from polarimetric SAR data using machine learning: A comparative and validation study
Increasing combined heat and drought extremes due to climate change heighten the risk of crop failure, underscoring the need for improved stress diagnosis for effective management strategies. However, current plant physiology indicators struggle to differentiate crop stresses in hot-dry environments. This study proposes using specific leaf metabolites, detectable by leaf reflectance spectra, for more precise identification of heat and drought stress compared to traditional methods. We conducted two rounds of one-week drought treatments under heat stress on soybean seedlings. Throughout the experiment, we monitored stomatal conductance, reflectance spectra, and metabolites, including Abscisic Acid (ABA), Jasmonic Acid (JA), Salicylic Acid (SA), and proline (Pro), on a daily basis. Our findings revealed that ABA and JA exhibited differential sensitivities to drought and heat stress, respectively. In contrast, stomatal conductance was unable to differentiate between the two stressors. Using partial least-squares regression (PLSR), we determined that both ABA and JA could be detected via leaf spectroscopy with moderate predictive performance (R² = 0.53, relative RMSE = 14.28 %; R² = 0.53, relative RMSE = 14.96 %) and exhibited distinct sensitive spectral signatures. The metabolite-derived, stress-specific spectral models enable more precise and earlier diagnosis and differentiation of stress in a hot-dry environment than traditional physiological indicators (e.g., relying on stomatal conductance). This study provides an example of using metabolites as novel stress indicators, which could contribute to precision agriculture, offering the potential for accurate, stress-specific, and pre-physiological detection of crop health.
Why it matches plant phenotyping methods葉の反射スペクトルとPLSRにより、熱・ drought stressに関連する代謝物を推定し、植物ストレス状態を識別する手法の開発・検証が研究の中心である。
abstractThis study proposes using specific leaf metabolites, detectable by leaf reflectance spectra, for more precise identification of heat and drought stress compared to traditional methods.
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.
Soybean is one of the world's major oil-bearing crops and occupies an important role in the daily diet of human beings. However, the frequent occurrence of soybean leaf diseases caused serious threats to its yield and quality during soybean cultivation. Rapid identification of soybean leaf diseases could provide a better solution for efficient control and subsequent precision application. In this study, a lightweight deep convolutional neural network (CNN) based on multiscale feature extraction fusion (MFEF) and combined with a dense connectivity (DC) network (MFEF-DCNet) was proposed for soybean leaf disease identification. In MFEF-DCNet, a multiscale feature extraction fusion (MFEF) module for soybean leaves was constructed by utilizing a convolutional attention module and depth-separable convolution to improve the model feature extraction capability. Multiscale features are fused by using dense connections (DC) in the backbone network to improve the model generalization capability. Experiments were implemented on eight distinct disease and deficiency classes of soybean images (including bacterial blight, cercospora leaf blight, downy mildew, frogeye leaf spot, healthy, potassium deficiency, soybean rust, and target spot) using the proposed network. The results showed that the MFEF-DCNet had an accuracy of 0.9470, an average precision of 0.9510, an average recall of 0.9480, and an F1-score of 0.9490 for soybean leaf disease identification. And MFEF-DCNet had certain performance advantages in terms of classification accuracy, convergence speed and other effects compared with VGG16, ResNet50, DenseNet201, EfficientNetB0, Xception and MobileNetV3_small models. In addition, the accuracy of the MFEF-DCNet model in recognizing soybean diseases in local data was 0.9024, which indicated that the MFEF-DCNet model had favorable application in practical applications. The proposed model and experience in this study could provide useful inspiration for automated disease identification in soybean and other crops.
Why it matches plant phenotyping methods大豆葉の病害状態を画像から認識するCNNを開発・評価しており、植物病害フェノタイピング手法が研究の中心である。
abstracta lightweight deep convolutional neural network (CNN) based on multiscale feature extraction fusion (MFEF) and combined with a dense connectivity (DC) network (MFEF-DCNet) was proposed for soybean leaf disease identification.
Reproduction assets foundThe paper's phenotyping analysis is built on a public plant-image dataset: the Auburn Soybean Disease Image Dataset (ASDID), explicitly described as publicly available and cited with a Dryad DOI. No author analysis code, trained model checkpoints, or supplementary code/model deposit is stated in the supplied blocks. NoDataset · publicThe dataset used for the study was the Auburn Soybean Disease Image Dataset (ASDID), which is publicly available and has been extensively studied and validated ( Bevers et al., 2022 ).Open asset ↗lines:324-376Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Stomata play a crucial role in plant immune responses, with their morphological characteristics closely linked to disease resistance. Accurate detection and analysis of stomatal phenotypic parameters are essential for soybean disease resistance research and variety breeding. However, traditional stoma detection methods are challenged by complex backgrounds and leaf vein structures in soybean images. To address these issues, we proposed a Soybean Stoma-YOLO (You Only Look Once) model (SS-YOLO) by incorporating large separable kernel attention (LSKA) in the Spatial Pyramid Pooling-Fast (SPPF) module of YOLOv8 and Deformable Large Kernel Attention (DLKA) in the Neck part. These architectural modifications enhanced YOLOV8's ability to extract multi-scale and irregular stomatal features, thus improving detection accuracy. Experimental results showed that SS-YOLO achieved a detection accuracy of 98.7%. SS-YOLO can effectively extract the stomatal features (e.g., length, width, area, and orientation) and calculate related indices (e.g., density, area ratio, variance, and distribution). Across different soybean rust disease stages, the variety Dandou21 (DD21) exhibited less variation in length, width, area, and orientation compared with Fudou9 (FD9) and Huaixian5 (HX5). Furthermore, DD21 demonstrated greater uniformity in stomatal distribution (SEve: 1.02-1.08) and a stable stomatal area ratio (0.06-0.09). The analysis results indicate that DD21 maintained stable stomatal morphology with rust disease resistance. This study demonstrates that SS-YOLO significantly improved stoma detection and provided valuable insights into the relationship between stomatal characteristics and soybean disease resistance, offering a novel approach for breeding and plant disease resistance research.
Why it matches plant phenotyping methods大豆葉画像から気孔を検出し、形態・分布などの植物形質を抽出するYOLO手法を開発しており、表現型取得が研究の中心である。
abstractwe proposed a Soybean Stoma-YOLO (You Only Look Once) model (SS-YOLO)
Selecting soybean lines that thrive under increasingly extreme weather conditions requires a deep understanding of genotype by environmental interaction (GxE) interactions. While controlled experiments offer valuable insights, they rarely capture the complexity of real field conditions. Here, we combine high-resolution, 8-year field imagery with weather data to model soybean canopy growth using automatically labeled deep learning for soybean-weed segmentation and one-stage nonlinear mixed-effects modeling. The models revealed distinct genotypic growth patterns, allowing to pinpoint environmental covariates, specifically photothermal product and precipitation, as drivers of canopy development. Two breeding lines, ‘Everest’ and ‘Kalinka’ recently released as varieties, show stable performance across contrasting weather scenarios with minimal GxE interaction. Genotypic model coefficients predicted key agronomic traits such as maturity timing and protein yield. This demonstrates that growth models can reveal genotypic resilience to environmental variation, even from single-environment measurements, offering valuable insights for breeding and climate adaptation.
Why it matches plant phenotyping methods画像からの雑草分割とキャノピー成長モデルにより、ダイズのキャノピー成長という植物形質を抽出・予測する手法が研究の中心である。
abstractwe combine high-resolution, 8-year field imagery with weather data to model soybean canopy growth using automatically labeled deep learning for soybean-weed segmentation and one-stage nonlinear mixed-effects modeling.
Deep understanding of slow-wilting is essential for developing drought-tolerant crops. Existing approaches to measure transpiration rates are difficult to apply to large populations due to their high cost and low throughput. To overcome these challenges, we developed a high-throughput phenotyping system that integrates a load cell sensor and an Arduino-based microcontroller device. The system tracked the transpiration rate in real time by measuring changes in the pot weight in 224 recombinant inbred lines of Taekwangkong (fast-wilting) x SS2-2 (slow-wilting) under water-restricted conditions. Among five transpiration features we determined, stress recognition time point (SRTP) and decrease in transpiration rate by stress (DTrs) are informative parameters, that are interconnected and independently affect slow-wilting as well. Quantitative trait loci (QTL) for SRTP and DTrs were identified at the same location as the major QTL for slow wilting, qSW_Gm10 , identified in the previous study. Notably, we found a novel major QTL for DTrs, qDTrs_Gm04 , with a LOD value of 42 and PVE of 47 %. As a candidate gene for qDTrs_Gm04 , GmWRKY58 was selected with differential expression between the parental lines under drought conditions as well as upstream sequence variation. Our high-throughput system is of help not only to biological research but breeding programs of drought-tolerant lines.
Why it matches plant phenotyping methods高スループットなセンサー基盤を開発し、ポット重量変化からダイズの蒸散率・乾燥ストレス応答をリアルタイム抽出することが研究の中心であるため。
abstractwe developed a high-throughput phenotyping system that integrates a load cell sensor and an Arduino-based microcontroller device.
Reproduction assets foundThe paper's data availability statement explicitly deposits the processed phenotypic data (transpiration features from the RIL drought experiment) and trained Random Forest/XGBoost model objects on Figshare, which is a paper-specific, publicly accessible asset.Dataset · publicThe processed phenotypic data, along with the trained Random Forest and XGBoost machine learning model objects (.rds files), are publicly available on Figshare at https://doi.org/10.6084/m9.figshare.c.7951601.v1.Open asset ↗Figshare · 10.6084/m9.figshare.c.7951601.v1html-lines:276-299Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Sept 2025Computers and Electronics in Agriculture.
Traditional methods for measuring pre-harvest loss, such as using quadrats, are labor-intensive and provide sparse data coverage. This study proposes an automated approach that leverages computer vision to replace and enhance the current method, using advanced imaging technologies and deep learning methodologies to detect and quantify pre-harvest losses in grain crops. Specifically, the methodology employs a camera mounted on the front snout of a ground vehicle, allowing continuous image capture along the crop rows. By automating image collection and analysis, this approach provides denser spatial coverage across the field, reduces errors associated with manual sampling and human judgment, and significantly accelerates the process compared to traditional quadrat sampling. In addition, this approach offers the potential to segregate different types of pre-harvest loss, such as natural shattering versus losses caused by mechanical disturbance, providing a level of granularity not achievable with conventional quadrat methods. By leveraging state-of-the-art object detection architectures the system is designed to handle the complex visual environment of the field floor, where grains may be obscured by crop residue, shadows, and similarly-colored objects such as stones. This capability represents a significant advancement over traditional methods, which cannot distinguish between these different loss sources. The images were annotated using the Segment Anything Model (SAM) to ensure consistency and accuracy across the dataset. Several state-of-the-art models were trained and evaluated on the collected data, including Mask RCNN, YOLOX, DETR, and a modified YOLOv8-p2. The modified YOLOv8-p2 model, which incorporated a p2 head to improve the detection of smaller objects, outperformed the others, yielding the highest Precision, Recall, and F1 scores on both the soybean (Precision = 0.727, Recall = 0.694, F1 = 0.710) and wheat (Precision = 0.709, Recall = 0.688, F1 = 0.698) datasets. Integrating the 850 nm NIR image channel did not produce a meaningful boost in performance, as evidenced by the soybean (Precision = 0.741, Recall = 0.689, F1 = 0.715) and wheat (Precision = 0.729, Recall = 0.690, F1 = 0.709) results. This research demonstrates that it is possible to integrate a vision system on the header of a combine and identify the initial shedding loss in the field. Future work will focus on refining the models further, exploring their applicability to other crop types, and integrating real-time processing and automation in data collection and analysis.
Why it matches plant phenotyping methods穀粒の収穫前損失という植物・作物状態を、車載カメラとコンピュータビジョンで自動検出・定量する手法を開発し、複数モデルで性能評価しており、表現型取得が研究の中心である。
abstractThis study proposes an automated approach that leverages computer vision to replace and enhance the current method, using advanced imaging technologies and deep learning methodologies to detect and quantify pre-harvest losses in grain crops.
Accurate knowledge of vegetation water content (VWC) and crop height is crucial for agricultural management, environmental monitoring, and for satellite-based retrieval algorithms for geophysical variables. Traditional methods to estimate VWC, primarily rely on optical indices, which has limitations of biomass saturation, and sensitivity to atmospheric conditions. This study introduces a novel application of geospatial foundation models (GFMs), leveraging extensive, unlabeled datasets through self-supervised learning to enhance the skill of VWC and crop height estimation. We developed a comprehensive model integrating Sentinel-1 A C-band SAR and Sentinel-2 A/B indices with weather parameters to estimate soybean and corn VWC and crop height. Our research study area spans a variety of climatic zones and management practices, from the humid continental climate of Iowa and Michigan to the subtropical environment of Florida, encompassing both irrigated and non-irrigated fields as well as diverse tillage practices. We compared the performance of Single-Task Learning GFM (STL-GFM), Multi-Task Learning GFM (MTL-GFM), and machine learning techniques including Random Forest (RF), and XGBoost (XGB) to evaluate their effectiveness in estimating VWC and crop height. Results demonstrated that STL-GFM outperforms other methods in accuracy and generalizability. For VWC estimation, STL-GFM achieved R² values of 0.90 and 0.89 for soybean and corn, respectively. For crop height, R² values reached 0.95 for soybean and 0.98 for corn. The integration of SAR, optical, and climate data provided more reliable estimations than using individual data sources. Feature importance analysis identified NDVI, NDWI, VH backscatter, and precipitation as key drivers for accurate VWC and height estimations. The red-edge band emerged as significant for VWC estimation but showed limited importance for height prediction. Notably, surface roughness demonstrated a noticeable impact on corn VWC and height estimations, while soil moisture exhibited less influence than initially anticipated. Notably, without directly incorporating soil moisture and surface roughness data, but by including diverse field conditions in training and validation, the STL-GFM model demonstrated strong generalization capabilities. This study highlights the potential of GFMs in advancing crop monitoring techniques, offering more reliable data for precision agriculture, and supporting sustainable farming practices across diverse agricultural landscapes.
Why it matches plant phenotyping methodsSAR・光学衛星データと自己教師あり地理空間基盤モデルを統合し、作物のVWCと草丈を推定する手法を開発・比較・評価しており、植物表現型の取得・抽出が中心である。
abstractThis study introduces a novel application of geospatial foundation models (GFMs), leveraging extensive, unlabeled datasets through self-supervised learning to enhance the skill of VWC and crop height estimation.
Plant phenotyping relevance match · UnverifiedCrossref · OpenAlex · Europe PMC · checked 15 Sept 2026
With the increasing global demand for food, breeding soybean varieties resistant to dense planting is crucial for achieving high and stable yields. Traditional phenotyping methods are limited by insufficient temporal resolution and challenges in dynamic modeling continuity, making it difficult to elucidate the intrinsic relationship between canopy development rate and yield stability. Moreover, existing machine learning models often neglect temporal dependencies in time series predictions, leading to insufficient biological interpretability. This study proposes an innovative approach integrating spatiotemporal deep learning and dynamic modeling to quantify the dynamic changes in canopy parameters using UAV high-throughput phenotyping technology, revealing the key regulatory mechanisms of traits associated with resistance to dense planting. Based on a two-year field experiment (2022-2023) in northeast China (Qiqihaer, black soil region), this study set high (50w plants/ha) and low density (30w plants/ha) treatments across 208 soybean varieties, combined with multispectral UAV imagery (15-18 times per season) and ground-truth data, to develop a time series prediction model for leaf area index (LAI). Comparing the performance of spatiotemporal residual networks (ST-ResNet), long short-term memory networks (LSTM), and traditional random forests (RF), the ST-ResNet model demonstrated significantly superior prediction accuracy (R 2 = 0.90, RMSE = 0.23 m 2 /m 2 ), effectively capturing the continuous dynamics of canopy growth through its spatiotemporal feature fusion ability. By fitting the time series curves of LAI, canopy cover (CC), and plant height (PH) with P-spline, 15 intermediate traits (e.g., ΔMean LAI-mid ) were extracted. Mixed models and SHAP interpretability analysis showed that ΔMean LAI-mid was most correlated with the dense planting yield index (ΔYield, r = 0.51). Furthermore, the high-frequency data acquisition and automated analysis framework using UAVs enabled high-throughput phenotypic screening for 208 varieties per year, significantly improving efficiency compared to traditional methods that rely on manual sampling. This study pioneers the integration of spatiotemporal deep learning with dynamic trait modeling, markedly improving the temporal continuity and stability of LAI estimation compared to traditional single-time-point prediction methods. This advancement allows for more precise quantification of canopy development rates across various growth stages, enabling a systematic analysis of how these dynamic patterns influence resistance to dense planting. By elucidating the dynamic relationship between intermediate traits and yield, this approach offers a high-precision, interpretable phenotypic analysis framework for effectively screening soybean varieties resilient to dense planting.
Why it matches plant phenotyping methodsUAV画像と時系列深層学習・動的モデリングを用いてLAI、群落被覆、草丈などの植物形質を推定し、高スループット表現型スクリーニング基盤を開発・検証しているため、方法が研究の中心である。
abstractThis study proposes an innovative approach integrating spatiotemporal deep learning and dynamic modeling to quantify the dynamic changes in canopy parameters using UAV high-throughput phenotyping technology
Soybean production in Japan is increasingly affected by climate change, with rising temperatures and changing soil moisture conditions contributing to green stem disorder (GSD), seed coat cracking (SCC), and seed coat wrinkling (SCW). These disorders reduce seed yield, lower seed quality, and complicate harvesting. To better understand and predict their occurrence (score), we developed random forest (RF) regression models using historical cultivar data and environmental factors from four major soybean breeding sites in Japan. The RF models outperformed traditional regression methods, achieving moderate prediction accuracy for GSD, SCC, and SCW scores (R² > 0.5). Analysis of the partial dependence plot suggested that increased GSD and SCC scores were associated with high temperatures during reproductive stages, while the SCW score showed a stronger link to cultivar traits. Future projections, derived from predictive models and future climate scenarios, suggested that GSD and SCC scores could increase at all sites, whereas the SCW score might rise at specific sites. Adaptation strategies such as late sowing and use of late-maturing cultivars showed potential for reducing risks, but their effectiveness varied by site and disorder type. These findings underscore the importance of considering region-specific strategies to address climate-related challenges in soybean production. By integrating machine learning with historical cultivar data, this study offers insights into developing targeted adaptation measures that could support sustainable soybean cultivation in a changing climate.
Why it matches plant phenotyping methods大豆の障害スコアという植物状態を対象に、RF回帰モデルを開発・比較評価し、予測性能も検証しているため、計算的な表現型推定が中心的です。
abstractwe developed random forest (RF) regression models using historical cultivar data and environmental factors
To address the challenges of high model complexity, substantial computational resource consumption, and insufficient classification accuracy in existing soybean seed identification research, we first perform soybean seed segmentation based on polygon features, constructing a dataset comprising five categories: whole seeds, broken seeds, seeds with epidermal damage, immature seeds, and spotted seeds. The MobileViT module is then optimized by employing Depthwise Separable Convolution (DSC) in place of standard convolutions, applying Transformer Half-Dimension (THD) for dimensional reconstruction, and integrating Dynamic Channel Recalibration (DCR) to reduce model parameters and enhance inter-channel interactions. Furthermore, by incorporating the CBAM attention mechanism into the MV2 module and replacing the ReLU6 activation function with the Mish activation function, the model’s feature extraction capability and generalization performance are further improved. These enhancements culminate in a novel soybean seed detection model, MobileViT-SD (MobileViT for Soybean Detection). Experimental results demonstrate that the proposed MobileViT-SD model contains only 2.09 million parameters while achieving a classification accuracy of 98.39% and an F1 score of 98.38%, representing improvements of 2.86% and 2.88%, respectively, over the original MobileViT model. Comparative experiments further show that MobileViT-SD not only outperforms several representative lightweight models in both detection accuracy and efficiency but also surpasses a number of mainstream heavyweight models. Its highly optimized, lightweight architecture combines efficient inference performance with low resource consumption, making it well-suited for deployment in computing-constrained environments, such as edge devices.
Why it matches plant phenotyping methods大豆種子の状態・損傷を画像セグメンテーションと軽量分類モデルで推定する手法開発が中心であり、植物器官の形態・品質状態を取得するフェノタイピング手法に該当する。
abstractwe first perform soybean seed segmentation based on polygon features, constructing a dataset comprising five categories: whole seeds, broken seeds, seeds with epidermal damage, immature seeds, and spotted seeds.
Introduction Atmospheric CO 2 elevation significantly impacts plant carbon metabolism, yet accurate quantification of respiratory parameters-photorespiration rate (R p ) and mitochondrial respiration rate in the light (R d )-under varying CO 2 remains challenging. Current CO 2 -response models exhibit limitations in estimating these parameters, hindering predictions of crop responses under future climate scenarios. Methods Low-oxygen treatments and gas exchange measurements, calculating CO 2 recovery/inhibition ratio in of wheat ( Triticum aestivum L. ) and bean ( Glycine max L. ) were employed to elucidate the biological significance and interrelationships of R p and R d . Model-derived estimates of R p and R d were compared with measured values to assess the accuracy of three CO 2 -response models (biochemical, rectangular hyperbola, modified rectangular hyperbola). Furthermore, the effects of ambient CO 2 concentration (0~1200 μmol·mol -1 ) on the measured R p and R d were quantified through polynomial regression. Results The A/C a model achieved superior fitting performance over the A/Ci model. However, significant disparities persisted between A/Ca-derived R p /R d estimates and measurements ( p 2 concentration exhibited dose-dependent regulation of respiratory fluxes: R p-measured ranged from 4.923 ± 0.171 to 12.307 ± 1.033 μmol (CO 2 ) m -2 s -1 (wheat) and 4.686 ± 0.274 to 11.673 ± 2.054 μmol (CO 2 ) m -2 s ⁻ ¹ (bean), while R d-measured varied from 0.618 ± 0.131 to 3.021 ± 0.063 μmol (CO 2 ) m -2 s -1 (wheat) and 0.492 ± 0.069 to 2.323 ± 0.312 μmol (CO 2 ) m -2 s -1 (bean). Polynomial regression revealed strong non-linear correlations between CO 2 concentrations and respiratory parameters (R ² > 0.891, p p- C a : R ² = 0.797). Species-specific CO 2 thresholds governed peak R p (600 μmol·mol -1 for wheat vs. 1,000 μmol·mol -1 for bean) and R d (400 μmol·mol -1 for wheat vs. 200 μmol·mol -1 for bean). Discussion These findings expose critical limitations in current respiratory parameter quantification methods and challenge linear assumptions of CO 2 -respiration relationships. They establish a critical framework for refining photosynthetic models by incorporating CO 2 -responsive respiratory mechanisms. The identified non-linear regulatory patterns and model limitations provide actionable insights for advancing carbon metabolism theory and optimizing crop carbon assimilation strategies under rising atmospheric CO 2 , with implications for climate-resilient agricultural practices.
Why it matches plant phenotyping methodsCO₂応答モデルによる光呼吸・明所ミトコンドリア呼吸の推定精度を実測ガス交換値と比較検証しており、植物生理状態の取得・定量法が研究の中心です。
abstractModel-derived estimates of R p and R d were compared with measured values to assess the accuracy of three CO 2 -response models
Plant breeding programs know the advantages of high-throughput phenotyping (HTP) in increasing efficiency over classical phenotyping and screening methods, which is achieved by saving time and improving selection accuracy. Even so, most programs have not yet systematically implemented this technology into their breeding pipelines. This review aims to indicate the restrictions of implementing HTP at a large scale and to summarize studies according to the used devices, data classes collected, and artificial intelligence (AI) methods applied to predict and classify agronomic traits in plant breeding programs with a focus on soybean [Glycine max (L.) Merr.]. Excluding HTP platforms in laboratories and greenhouses, satellite remote sensing, and autonomous mobile robots, this review focuses on field-based HTP platforms that take aerial images from drones and apply AI methods to associate those images with the traits of interest. Field-based HTP research is also conducted using hand-held devices that record individual vegetation indices (e.g., NDVI), a few spectral bands (multispectral radiometers), or the continuous range of the electromagnetic light spectrum (spectroradiometers). However, plant breeders must evaluate thousands of experimental lines each year, so using these devices instead of drones implies a trade-off between acquisition accuracy and the time it takes to collect the data. A challenge in the coming years is fine-tuning scalable, reliable models and optimizing data input, processing, and output pipelines to provide breeders with helpful information before they make selections.
Why it matches plant phenotyping methods植物表現型取得を中心に、ドローン空撮とAIによる農業形質の推定を扱うHTPレビューであり、対象・方法・実装上の課題を体系的に整理しているため。
abstractThis review aims to indicate the restrictions of implementing HTP at a large scale and to summarize studies according to the used devices, data classes collected, and artificial intelligence (AI) methods applied to predict and classify agronomic traits in plant breeding programs
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 6 Sept 2026
BACKGROUND: In major soybean-growing regions worldwide, vertical (three-dimensional) planting systems are widely adopted. Achieving precise phenotyping of individual soybean plants is crucial for breeding shade-tolerant cultivars and optimizing high yields. However, canopy shading from taller crops severely restricts the acquisition of phenotypic information from the lower-growing soybeans, and conventional phenotyping platforms struggle to meet the demands of such complex planting structures. To address this challenge, this study developed a field-based high-throughput phenotyping platform specifically designed to accommodate the structural characteristics of vertical planting systems. RESULTS: The platform integrates the characteristics of vertical planting systems and consists of an imaging system and a rail-based transportation system.The imaging system balances the growth requirements of soybeans under natural conditions with the stability of indoor imaging, and is equipped with adjustable sensors, an automated rotating stage for image capture, and modules for image classification and storage. The transportation system includes X and Y dual-directional tracks and programmable rail carts, enabling automated movement of potted soybean plants in the field. Platform performance was validated through correlation analysis and predictive modeling. The extracted plant height and width showed high agreement with manual measurements, with coefficients of determination (R²) of 0.99 and 0.95, respectively. During the vegetative stage, the predictive accuracy (R²) for canopy fresh weight and leaf area reached 0.965 and 0.972, demonstrating strong predictive performance and robustness. In addition, the platform supports modular sensor integration and features an open-source control architecture, allowing seamless incorporation of additional sensors such as infrared cameras, LiDAR, and fluorescence imaging. This expands trait detection capacity while reducing costs for reuse and secondary development. CONCLUSION: This study demonstrated the feasibility of combining natural field conditions with standardized indoor imaging for phenotypic research on soybeans under vertical planting systems. The platform provides a flexible and scalable technical solution for analyzing plant architecture and screening germplasm in complex planting environments, opening up new technological pathways for precision agriculture and crop breeding research.
Why it matches plant phenotyping methods大豆の高スループット表現型計測プラットフォームを開発し、画像取得・搬送・形質抽出を検証した研究であり、フェノタイピング手法が中心である。
abstractthis study developed a field-based high-throughput phenotyping platform specifically designed to accommodate the structural characteristics of vertical planting systems.
Disease incidence is a key factor contributing to reduced crop yield. Thus, early identification of crop diseases is crucial for minimizing the effects of disease incidence and maximizing crop yield. Therefore, this study aims to identify soybean yellow mottle mosaic virus (SYMMV) using the hyperspectral imaging (HSI) method combined with the machine learning (ML) technique. The soybeans were cultivated under two different environmental conditions, namely, EN I and EN II. In EN I, soybean plants were infected with SYMMV at the third vegetative growth stage, whereas in EN II, infected seeds were used. A reverse transcription polymerase chain reaction was conducted to distinguish the infected from noninfected plants. Mean spectrum values obtained from regions of interest in the Environmental Visualizing Images software served as data, while their respective wavelengths were used as features for ML models. The information gain method was used for the selection of characteristic wavelengths associated with disease identification. Continuous wavelengths ranging from 653 nm to 682 nm showed more information gain in both environments, indicating their significant role in SYMMV classification. Two classification models, random forest and k-nearest neighbor, classified the infected and noninfected plants at an early stage with over 90% accuracy. The support vector machine classified the disease with an average accuracy of > 95% across both environments, showing the best performance among the selected models. The logistic regression model showed lower accuracy, exceeding 82% in EN I, but improved to > 90% in EN II. These findings suggest that HSI combined with ML is the best alternative to the traditional method of disease identification in plants.
Why it matches plant phenotyping methodsHSIと機械学習により感染植物を非感染植物から識別し、植物病害状態を推定する手法が研究の中心であるため。
abstractthis study aims to identify soybean yellow mottle mosaic virus (SYMMV) using the hyperspectral imaging (HSI) method combined with the machine learning (ML) technique.
Seed-level disease detection in soybeans presents significant challenges, including small-sample limitations, spectral interference, and dense occlusions, which are less pronounced in leaf-level analysis. To overcome these obstacles, we propose YOLOv8-ECCI, an enhanced algorithm based on YOLOv8 for high-precision identification of purple spot disease directly on soybean seeds. Experimental results demonstrate that YOLOv8-ECCI substantially outperforms the baseline YOLOv8n model, achieving significant gains of +5.7% precision, +6.5% recall, +8.0% mAP@0.5, and +7.1% mAP@0.5:0.95. Crucially, the model exhibits superior generalization capability, validated through rigorous cross-dataset testing on the African Wildlife dataset, where it surpasses conventional methods by +6.0% precision and +2.9% mAP@0.5. These results confirm that YOLOv8-ECCI effectively addresses the critical challenges in seed-level pathology, providing a robust and accurate solution for practical in-field agricultural disease detection and quality control.
Why it matches plant phenotyping methods大豆種子の病徴を画像から検出するYOLOv8改良手法の開発と、ベースラインおよびクロスデータセットによる技術検証が中心である。
abstractwe propose YOLOv8-ECCI, an enhanced algorithm based on YOLOv8 for high-precision identification of purple spot disease directly on soybean seeds.
Abstract Soybean growth is determined by the interaction of genetic, environmental, and management factors. In the context of future climate and climate extremes, understanding genotype by environment interaction (GxE) will be crucial for selecting resilient breeding lines and optimizing management practices to minimize stress. As stress periods occur periodically in a season, in depth knowledge, about causing weather variables and differing responses of genotypes over time is required. In field studies, however, the environment is often treated as a static factor, and the specific effects of weather variability on growth remain poorly understood. Here, we present a longitudinal dataset comprising 17,247 high-resolution RGB images of soybean breeding line collected over eight years in Eschikon, Switzerland. Top of canopy images were acquired throughout the entire growing seasons and complemented by hourly weather data, enabling a comprehensive analysis of soybean growth dynamics under varying field conditions. High spatio-temporal image resolution enables detailed analysis of growth dynamics and GxE, supporting identification of stress-tolerant genotypes to improve yield prediction and yield stability.
Why it matches plant phenotyping methods8年間の高解像度RGB画像による圃場フェノタイピングデータセットを提示し、作物生育動態とG×E解析を可能にする方法・データ基盤が中心である。
titleFIP 1.0 Soybean data: Insights on soybean growth from eight years of high-throughput image field phenotyping
Reproduction assets foundThe paper is a data note whose core contribution is a public soybean phenotyping dataset (raw FIP images, segmentation masks, canopy cover data, BLUEs, weather, reference traits) deposited at ETH Research Collection, plus the authors' canopy cover extraction workflow code on GitLab. Both are paper-specific, public, andDataset · publicason, therefore,
from 2020 to 2022, photosynthetic photon fluence rate (PPFR) was taken from a LI-COR sensor placed next to the
field. The factor to convert radiation in MJ m−
2 to PPFR was 2.04 according to [26].
4.1 Data Files and Structure
The dataset presented in this study is available at ETH Research Collection under DOI: https://doi.org/10.3929/ethz-b-000742401. The dataset is structured into directories that align with the described data processing
pipeline used for extracting and analyzing canopy cover traits from field images. All files are provided in interoperable
and widely-used ‘.csv‘ and ‘.png‘ format.
• data/Design 2015 2022 Eschikon.csv: Experimental design file, including pOpen asset ↗ETH Research Collection · 10.3929/ethz-b-000742401pdf-raw-page:6 lines:1-47Code · publicCollection (https://doi.org/10.3929/ethz-b-000742401) and as Hugging Face data set card (doi.org/10.57967/hf/6052) allowing
interoperability and standardization with other datasets.
9 Code availability
Users with similar data can use the implemented workflow to get canopy cover from their experiments. The code
is available on: https://gitlab.ethz.ch/crop_phenotyping/fip-soybean-canopycover
10 Author contributions
BK: Developed algorithm, analyzed data and drafted manuscript; NK, LR, AH, AM: FIP development, BK, NK,
CO, LK, LR, OZ, SC, FL, HA, NS, FT, HZ, CAB, CB, AH: Collected and prepared data; Experimental design:
BK, LK, LR, AH; all authors improved and approved the manuscript
5Open asset ↗gitlab.ethz.ch · crop_phenotyping/fip-soybean-canopycoverpdf-raw-page:7 lines:1-46Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Aug 2025Plant breeding = Zeitschrift fur Pflanzenzuchtung
In recent years, phenomic prediction has emerged as a new method in plant breeding that has been shown to have great potential. However, there are still many open questions regarding its practical application. For example, in the field of spectroscopy, it is standard practice to optimize the preprocessing of spectra, which so far has only been done to a limited extent for phenomic prediction. In this study, we therefore used three different data sets of soybean, triticale and maize to identify the best combinations of Savitzky–Golay filter parameters for preprocessing near‐infrared spectra for phenomic prediction. We tested 677 combinations of polynomial order, derivative and window size and evaluated them with Monte Carlo cross‐validation. Our results showed that the predictive ability can be improved with the right settings. However, there was no global optimum that gave the best results for all data sets. Even for different traits within the same data set, different combinations of parameters were necessary to achieve the highest predictive ability. Nevertheless, we show that some combinations generally result in a very low predictive ability and should not be used for preprocessing. In addition, we used the normalized discounted cumulative gain to assess whether preprocessing affected the ranking of individuals, which revealed no major changes in the top 1%, 10% or 20% of predicted individuals. Taken together, our results show the potential of preprocessing near‐infrared spectroscopy data to improve the phenomic predictive ability, but there appears to be no global optimum of parameter settings across data sets and traits.
Why it matches plant phenotyping methods植物育種におけるNIRSスペクトル前処理を最適化し、複数作物・形質で交差検証して予測性能と個体順位への影響を評価しており、表現型予測手法が研究の中心である。
abstractwe therefore used three different data sets of soybean, triticale and maize to identify the best combinations of Savitzky–Golay filter parameters for preprocessing near‐infrared spectra for phenomic prediction.
In order to intelligently and non-destructively estimate soybean yield, the fusion of multicolor space and texture feature parameters were considered to construct a yield prediction model. In this study, the yield prediction model was developed through different stands formed by a management experiment of different planting densities and nitrogen application strategies, and validated through a variety test of 28 soybean varieties. Images of the soybean canopy were collected during the key period for yield (the florescence, podding, and grain-filling stages) by unmanned aerial vehicle (UAV). The multicolor space and texture feature parameters of the RGB images of soybean canopy were extracted for these three periods, and soybean yield prediction models were constructed for the florescence, podding, grain-filling, and multiple growth stages based on different parameters combinations, by using methods of multiple linear regression (SMLR), random forest (RF), and back propagation neural networks (BPNN). The results showed that the color space and texture features of the soybean canopy images exhibited significant differences and different trends during the florescence, podding, and grain-filling stages. Models built with the combinations of texture features (TF) and Hue, Saturation and Value (HSV) parameters had little changes in accuracy, while those incorporating skewed parameters (SP) had better model accuracy. Among all the models, the model combining the SP, TF and HSV parameters demonstrated significantly greater accuracy. The accuracy of models based on individual reproductive stage was lower than those based on entire reproductive period. Thus, the best-performing model was a BPNN model using a combination of the SP, HSV and TF parameters of entire reproductive period as input factors, achieving an R² of 0.765, a prediction accuracy (PA) of the validation set of 88.2 %, and a root mean square error (RMSE) of 430.50 kg/ha. The accuracy of this model in predicting soybean yields across different varieties was PA = 80.4 %, and the RMSE = 514.28 kg/ha. This article provides an effective and low-cost method for accurately estimating soybean yield in the field. This method performs robustly in different varieties and agricultural practices, and has practical value.
Why it matches plant phenotyping methodsUAV画像から色・テクスチャ特徴を抽出し、収量という植物形質を予測する手法を開発・検証しており、表現型取得・推定が研究の中心です。
abstractIn order to intelligently and non-destructively estimate soybean yield, the fusion of multicolor space and texture feature parameters were considered to construct a yield prediction model.
Soybean [Glycine max (L.) Merr.] seed morphology markedly influences yield, productivity, and nutritional value. However, assessing quantitative traits remains challenging due to their complexity and strong genotype-by-environment interactions. In this study, a high-throughput phenotyping (HTP) system was used to evaluate 13 image-based traits and a hundred-seed weight in a soybean mutant diversity pool (MDP) comprising 192 genotypes. All traits exhibited significant variations within the mutant diversity pool across multiple environments. Correlation analysis revealed strong positive and negative correlations among the traits regarding seed size, shape, color, and weight. Genome-wide association studies (GWAS) were conducted using 37,249 single nucleotide polymorphisms (SNPs) generated through genotype-by-sequencing (GBS) to uncover the genetic architecture of seed-related traits. The image-based GWAS identified 79 significant quantitative trait nucleotides (QTNs) that were simultaneously detected under all environments. Notably, five novel pleiotropic QTNs were consistently mapped to chromosomes 7, 10, 15, 18, and 20, each associated with a specific candidate gene. These genes exhibited marked expression differences during the seed developmental stages between the wild-type cultivar and its mutant. The HTP-integrated GBS demonstrates a powerful approach for precise trait dissection and genomic selection. These findings provide critical insights into the genetic architecture underlying desirable seed morphology and offer valuable tools for advancing precision soybean breeding.
Why it matches plant phenotyping methods種子形態を13の画像形質として高スループットに取得するフェノタイピングシステムが研究の中心的手段であり、GWASへの実質的な適用も行っているため。
abstracta high-throughput phenotyping (HTP) system was used to evaluate 13 image-based traits
Background: Food security being one of the prominent global issues requires strategies to maximize plant productivity with due consideration to sustainability. In this regard, precision agricultural practices are incorporated and are quite promising. The use of technology i.e. integrating AI is rapidly changing the complete agricultural scenario. A quick identification of wilting ensures farmers take early action and remedies. An early detection of plant wilting in a real-time scenario can avoid huge food crop losses but the whole task is humongous. The images collected from open fields over large areas can be analyzed via various image processing techniques. Using the CNN model for identification of early images for quick response to plant care saves the time and effort of the farmers. CNN utilization allows prompt wilting detection and early corrective action to protect the crop. In this paper, CNN trained on image data allows high predictability of wilting detection at early stages in Soybean plants. Methods: A well-defined dataset of 6704 pictures of soybean plants from agricultural fields is considered and allocated appropriately to train, test and validate the CNN model. Using image preprocessing, resizing and rescaling of images is done to ensure consistency of image dimensions. Noise elimination from the images is done via a low-pass filtering method to preserve low-frequency information and the images are converted to grayscale. Using standard Python libraries, data augmentation is ensured and 9158 images across all classes are arranged for the model. The performance evaluation matrix indicating accuracy percentage, precision percentage and recall percentage is estimated. Result: The proposed CNN algorithm is first calibrated using a set of images from the dataset and then tested for an entirely different set of images not used earlier. The overall accuracy is 91%. The model promises unambiguous identification of wilting in soybean leaves by appropriately classifying images set in 5 orders using a substantially ample dataset. Early identification in real time can prove to be of utmost benefit to the agricultural community in terms of eradication of the causes and yield retention of the crop.
Why it matches plant phenotyping methodsCNNによる画像解析でダイズの萎凋状態を段階分類する手法を開発・検証しており、植物の病害・生理状態の取得が研究の中心である。
abstractUsing the CNN model for identification of early images for quick response to plant care saves the time and effort of the farmers.
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
Soybean is the most important oilseed and forage crop globally. Advancements in high-throughput phenotyping technologies are critical for accelerating genetic improvement in modern breeding research. However, conventional methods for assessing soybean maturity remain labor intensive. This study developed high-throughput phenotyping algorithms based on unmanned aerial vehicle (UAV) multispectral imagery combined with machine learning to monitor the maturity process of 30 soybean cultivars in large-scale breeding trials. UAV images and plant water content (PWC) data were collected to classify soybean maturity into four distinct phases: immaturity (i.e. the period before R5 stage), late pod filling (i.e. R5 to R6), physiological maturity (i.e. R7), and harvesting maturity (i.e. R8). We evaluated the performance of three classification approaches: (1) a computer vision model utilizing UAV-derived color features, (2) a PWC-based model retrieving PWC dynamics using UAV-derived feature, and (3) a multimodal fusion model integrating computer vision and PWC dynamics. Computer vision model effectively distinguished immature and mature plants but showed limitations in resolving specific maturity phases due to genetic variation in canopy color among cultivars (training set accuracy: 0.71; validation set accuracy: 0.69). The sensitive UAV-derived features were applied to establish the prediction model of PWC using convolutional neural network, which achieved the highest R 2 (training set: R 2 = 0.95; validation set: R 2 = 0.86) between the predicted and measured PWC. The PWC-based algorithm outperformed the computer vision approach, achieving higher classification accuracy (training set: 0.78; validation set: 0.79). Strong correlations between PWC and pod water content, stem water content, and leaf water content underscored the physiological relevance of PWC in tracking maturation dynamics. Further improvement in classification accuracy was achieved with the multimodal fusion model (training set: 0.84; validation set: 0.83), which combined the information of computer vision and PWC dynamics. It was also confirmed that the multimodal fusion model achieved the lowest misclassification rate in the validation analysis across diverse soybean cultivars. These findings emphasize the potential of integrating UAV-based computer vision and PWC features to improve the accuracy and efficiency of soybean maturity classification. The proposed multimodal approach offers a robust framework for phenotypic selection and trait evaluation, providing valuable insights for soybean breeding programs.
Why it matches plant phenotyping methodsUAV画像、植物水分含量推定、機械学習を統合し、ダイズ成熟期という植物形質を高スループットに分類する手法を開発・検証しており、フェノタイピング手法が中心である。
abstractThis study developed high-throughput phenotyping algorithms based on unmanned aerial vehicle (UAV) multispectral imagery combined with machine learning to monitor the maturity process of 30 soybean cultivars in large-scale breeding trials.
Plant roots influence various soil physical properties by altering the soil structure and pore configuration; however, a detailed understanding of these effects remains limited. In this study, we applied a relatively simple approach for segmenting plant roots and soil constituents using X-ray computed tomography (CT) images to evaluate root-induced changes in soil structure. The method combines manual initialization with a layer-wise automated region-growing approach, enabling the extraction of the root systems of soybean, Italian ryegrass, and Guinea grass. The method utilizes freely available software with a simple interface and does not require advanced image analysis skills, making it accessible to a wide range of researchers. The soil particles, pore water, and pore air were segmented using a Kriging-based thresholding technique. The segmented four-phase images allowed for the quantification of the volume fractions of soil constituents, pore size distributions, and coordination numbers. Furthermore, by separating the rhizosphere and bulk soil, we found that the root presence significantly reduced solid fractions and increased water content, particularly in the upper soil layers. Macropores and fine pores were observed near the roots, highlighting the complex structural impacts of root growth. While further validation is needed to assess the method’s applicability across different soil types and imaging conditions, it provides a practical basis for visualizing and quantifying root–soil interactions, and could contribute to advancing our understanding of how plant roots influence key soil hydraulic and thermal properties.
Why it matches plant phenotyping methodsX線CT画像から植物根系を抽出・定量するセグメンテーション手法の開発が中心であり、根系という植物形態形質の取得に直接関係するため。
abstractwe applied a relatively simple approach for segmenting plant roots and soil constituents using X-ray computed tomography (CT) images to evaluate root-induced changes in soil structure.
High temperature stress (HT) plays an important role in soybean selection and breeding, it can cause changes in soybean physiological, biochemical and morphological traits, and directly affect the growth and yield of soybean plants. Among these changes, soybean leaves are particularly sensitive to HT during growth and development. It is important to establish a non-destructive method to distinguish the phenotypic differences between soybean plants under HT and control (CK). In this study, data from two years of soybean field trials were used. In the first year, phenotypic information was collected by near-infrared spectroscopy (NIR), microscopic images, and further difference analysis and classification modelling experiments were conducted. In the second year, multispectral image data were collected and analyzed by Soybean high temperature mask autoencoder (SHT_MAE). The SHT_MAE model with a 75% masking ratio achieved an accuracy of 89.16% and an F1-score of 89.18%. Compared with one-dimensional near-infrared and two-dimensional microscopic image fusion models, the classification accuracy of HT and CK is improved by 2.68%. The accuracy of SHT_MAE multispectral model was improved by 16.84% and 6.88%, respectively, compared with models using only NIR or microscopic images. Both spectral and imaging methods effectively distinguish the phenotypic differences between HT and CK soybean leaves, with the multispectral approach based on the SHT_MAE model demonstrating a clear advantage. This study realized the effective distinction of soybean leaves under HT and CK. It provides theoretical support for HT intelligent breeding (using artificial intelligence and data analysis to optimize breeding decisions) and high temperature grade prediction.
Why it matches plant phenotyping methods高温ストレス下のダイズ葉の表現型差を、近赤外分光・顕微鏡画像・マルチスペクトル画像と分類モデルで非破壊的に抽出・比較する手法研究であり、表現型取得と解析が中心である。
abstractIt is important to establish a non-destructive method to distinguish the phenotypic differences between soybean plants under HT and control (CK).
Crop diseases pose a significant threat to agricultural productivity and global food security. Timely and accurate disease identification is crucial for improving crop yield and quality. While most existing deep learning-based methods focus primarily on image datasets for disease recognition, they often overlook the complementary role of textual features in enhancing visual understanding. To address this problem, we proposed a cross-modal data fusion via a vision-language model for crop disease recognition. Our approach leverages the Zhipu.ai multi-model to generate comprehensive textual descriptions of crop leaf diseases, including global description, local lesion description, and color-texture description. These descriptions are encoded into feature vectors, while an image encoder extracts image features. A cross-attention mechanism then iteratively fuses multimodal features across multiple layers, and a classification prediction module generates classification probabilities. Extensive experiments on the Soybean Disease, AI Challenge 2018, and PlantVillage datasets demonstrate that our method outperforms state-of-the-art image-only approaches with higher accuracy and fewer parameters. Specifically, with only 1.14M model parameters, our model achieves a 98.74%, 87.64% and 99.08% recognition accuracy on the three datasets, respectively. The results highlight the effectiveness of cross-modal learning in leveraging both visual and textual cues for precise and efficient disease recognition, offering a scalable solution for crop disease recognition.
Why it matches plant phenotyping methods作物葉の病徴を画像・テキストから認識するマルチモーダル手法の開発と評価が研究の中心であり、植物の病害状態を直接推定している。
abstractwe proposed a cross-modal data fusion via a vision-language model for crop disease recognition.
Reproduction assets foundThe paper's Data Availability Statement explicitly links the public image datasets used for its crop disease recognition experiments: the Soybean Disease dataset (Dryad DOI) and the PlantVillage dataset (Kaggle). These are the phenotyping image inputs directly used in this study. The AI Challenge 2018 dataset is also公开Dataset · publicThe soybean
dataset is available at https://doi.org/10.5061/dryad.41ns1rnj3 (accessed on 1 April 2025).Open asset ↗Dryad · 10.5061/dryad.41ns1rnj3pdf-page:12 lines:1-58Dataset · publicplantvillage dataset is available at https://www.kaggle.com/datasets/abdallahalidev/plantvillage-Open asset ↗Kagglepdf-page:12 lines:1-58Plant phenotyping relevance match · UnverifiedOpenAlex · checked 15 Sept 2026
Published25 Jun 2025ISPRS Journal of Photogrammetry and Remote SensingCited by 6 · OpenAlex ↗
Abstract has not been obtained from indexed metadata or an accessible article page.
Why it matches plant phenotyping methodsUAV時系列画像と深層学習を用いた育種圃場のセグメンテーションおよび高収量品種スクリーニングが題名の中心であり、植物育種における画像ベースの表現型取得・推定手法に該当する。
titleSTANet-TLA: leveraging deep learning and prior knowledge for large-scale soybean breeding plot segmentation and high-yielding variety screening from UAV time-series data
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 6 Sept 2026
Soybeans are important due to their nutritional benefits, economic role, agricultural contributions, and various industrial applications. Effective leaf detection plays a crucial role in analyzing soybean growth within precision agriculture. This study examines the influence of different labeling methods on the efficiency of artificial intelligence (AI) based soybean leaf detection. We compare a traditional general labeling technique against a new context-aware method that utilizes information about leaf length and bottom extremities. Both approaches were employed to train a YOLOv5L deep learning model using high-resolution soybean imagery. Results show that the general labeling method excelled with soybean varieties that have wider internodes and distinctly separated leaves. In contrast, the context-aware labeling method outperformed the general approach for medium soybean varieties characterized by narrower internodes and overlapping leaves. By optimizing labeling strategies, the accuracy and efficiency of AI-based soybean growth analysis can be significantly improved, particularly in high-throughput phenotyping systems. Ultimately, the findings suggest that a thoughtful approach to labeling can enhance agricultural management practices, contributing to better crop monitoring and improved yields.
Why it matches plant phenotyping methods大豆葉の検出精度を高めるラベリング手法を比較・評価し、AI画像解析とハイスループット表現型解析への適用を中心に扱うため、植物フェノタイピング手法研究に該当する。
abstractThis study examines the influence of different labeling methods on the efficiency of artificial intelligence (AI) based soybean leaf detection.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · Crossref · checked 6 Sept 2026
This paper presents a novel method, Histogram of Angles in Linked Features (HALF), designed for the segmentation of 3D point cloud data of plants for robust sensing. The proposed method leverages local angular features extracted from 3D measurements obtained via sensing technologies such as laser scanning, LiDAR, or photogrammetry. HALF enables efficient identification of plant structures-leaves, stems, and knots-without requiring large-scale labeled datasets, making it highly suitable for applications in plant phenotyping and structural analysis. To enhance robustness and interpretability, we extend HALF to a convolution-based mathematical framework and introduce the Sequential Competitive Segmentation Algorithm (SCSA) for phytomer-level classification. Experimental results using 3D point cloud data of soybean plants demonstrate the feasibility of our method in sensor-based plant monitoring systems. By providing a low-cost and efficient approach for plant structure analysis, HALF contributes to the advancement of sensor-driven plant phenotyping and precision agriculture.
Why it matches plant phenotyping methods植物の3D点群から葉・茎・節などの構造を分割・分類する新規センシング手法を開発し、植物フェノタイピングへの適用可能性を実証しているため。
abstractThis paper presents a novel method, Histogram of Angles in Linked Features (HALF), designed for the segmentation of 3D point cloud data of plants for robust sensing.
Genomic selection (GS) and phenotypic selection (PS) are widely used for accelerating plant breeding. However, the accuracy, robustness, and transferability of these two selection methods are underexplored, especially when addressing complex traits. In this study, we introduce a novel data fusion framework, GPS (genomic and phenotypic selection), designed to enhance predictive performance by integrating genomic and phenotypic data through three distinct fusion strategies: data fusion, feature fusion, and result fusion. The GPS framework was rigorously tested using an extensive suite of models, including statistical approaches (GBLUP and BayesB), machine learning models (Lasso, RF, SVM, XGBoost, and LightGBM), a deep learning method (DNNGP), and a recent phenotype-assisted prediction model (MAK). These models were applied to large datasets from four crop species, maize, soybean, rice, and wheat, demonstrating the versatility and robustness of the framework. Our results indicated that: (1) data fusion achieved the highest accuracy compared with the feature fusion and result fusion strategies. The top-performing data fusion model (Lasso_D) improved the selection accuracy by 53.4% compared to the best GS model (LightGBM) and by 18.7% compared to the best PS model (Lasso). (2) Lasso_D exhibited exceptional robustness, achieving high predictive accuracy even with a sample size as small as 200 and demonstrating resilience to single-nucleotide polymorphism (SNP) density variations, underscoring its adaptability to diverse data conditions. Moreover, the model's accuracy improved with the number of auxiliary traits and their correlation strength with target traits, further highlighting its adaptability to complex trait prediction. (3) Lasso_D demonstrated broad transferability, with substantial improvements in predictive accuracy when incorporating multi-environmental data. This enhancement resulted in only a 0.3% reduction in accuracy compared to predictions generated using data from the same environment, affirming the model's reliability in cross-environmental scenarios. This study provides groundbreaking insights, pushing the boundaries of predictive accuracy, robustness, and transferability in trait prediction. These findings represent a significant contribution to plant science, plant breeding, and the broader interdisciplinary fields of statistics and artificial intelligence.
Why it matches plant phenotyping methods植物の形質予測を目的とするGPSデータ融合フレームワークを開発し、複数作物・モデルで精度、頑健性、環境間移 transferability を評価しており、形質推定手法が研究の中心である。
abstractwe introduce a novel data fusion framework, GPS (genomic and phenotypic selection), designed to enhance predictive performance by integrating genomic and phenotypic data through three distinct fusion strategies: data fusion, feature fusion, and result fusion.
Reproduction assets foundThe paper's authors publicly released their GPS analysis scripts on GitHub, and the study used public genomic+phenotypic datasets (rice, maize, wheat, SoyNAM) with explicit URLs. DNNGP and MAK repositories are cited third-party tools, not paper-specific assets.Code · publicThe GPS scripts are available in the release package on GitHub ( https://github.com/Jinlab-AiPhenomics/BioGPS ). All public datasets used in this study are listed in the main text ( Table 1 ).Open asset ↗Jinlab-AiPhenomics/BioGPSlines:244-249Code / dataset availability confirmedEurope PMC · checked 14 Sept 2026
ArabidopsisSoybeanCell / cellular structureRootMorphology / geometry measurementRoot system architecture
Abstract Root system architecture (RSA), the three-dimensional arrangement of roots in soil, is a critical determinant of plant productivity, resource use efficiency, and resilience to environmental stress. Despite its agronomic importance, RSA remains a largely untapped breeding target due to historical technical barriers in root phenotyping. We present RADICYL (Root Architecture 3D Cylinder), a scalable, non-invasive, gel-based platform enabling high-throughput, high-resolution quantification of 15 RSA traits in intact root systems. Applying RADICYL to a genetically diverse panel of 371 soybean accessions, we combined 3D phenotyping with genome-wide association studies (GWAS), single-nucleus RNA sequencing (snRNA-seq), and gene co-expression network (GCN) analysis to identify RCE1 and NPR3 as central regulators of RSA, suggesting auxin and salicylic acid-mediated signaling impacts RSA in specific root tissues. Functional validation in Arabidopsis mutants revealed conserved effects on root width and lateral root development. Our findings position the endodermis and metaphloem as key regulatory cell types and demonstrate how multi-omic frameworks can accelerate the discovery of functional genes underlying complex traits. This study establishes a foundation for cell-type-targeted genome editing and climate-smart crop engineering, offering actionable genetic targets to optimize root systems for improved nutrient acquisition, drought resilience, and deep carbon sequestration. By bridging genotype, cellular context, and phenotype, this work redefines RSA as a tractable and transformative trait for the future of crop improvement.
Why it matches plant phenotyping methodsRADICYLという根系構造を定量化する高スループット3Dフェノタイピング基盤の開発・適用が研究の中心であり、15形質を測定している。
abstractWe present RADICYL (Root Architecture 3D Cylinder), a scalable, non-invasive, gel-based platform enabling high-throughput, high-resolution quantification of 15 RSA traits in intact root systems.
Reproduction assets foundThe paper's Data and code availability section names public repositories containing the authors' analysis code: a GitLab repo for WGCNA/single-cell network analysis, a GitHub repo for the RADICYL root image segmentation/phenotyping pipeline, and PyGNA2 on PyPI/GitLab. These are paper-specific, publicly actionable code/Code · publicn every 5°, resulting in 72 images per plant per timepoint for subsequent 3D root
1103 reconstruction. Phenotypic traits were quantified using the same automated pipeline described
1104 above for soybean.
1105
1106 Data and code availability
1107 The code to analyze the WGCNA network and single-cell data can be found here:
1108 https://gitlab.com/salk-tm/soybean-root-gwas/. RADYCL Segmentation pipeline for image
1109 analysis can be found here: https://github.com/Salk-Harnessing-Plants-Initiative/SSRAPC-Soy-
1110 Segmentation-Root-Architecture-Phenotyping-for-Cylinder.git. PyGNA2 is available on PyPI
1111 (https://pypi.org/project/pygna2/) and GitLab (https://gitlab.com/salk-tm/pygna2).
1112Open asset ↗salk-tm/soybean-root-gwaspdf-layout-page:30 lines:1-54Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 6 Sept 2026
BACKGROUND: Phenotypic characterization of mature soybean pods is a crucial aspect of breeding programs, yet efficiently obtaining accurate pod phenotypic parameters remains a major challenge. Recent advances in deep learning, particularly in keypoint detection models, have introduced innovative methods for pod phenotype extraction. However, precise identification and analysis of fine-scale phenotypic traits in soybean pods remain challenging in current research. RESULTS: We propose Pod-pose, an innovative top-down keypoint detection model for precise soybean pod phenotyping that adapts human pose estimation techniques to plant phenotyping. Specifically, Pod-pose integrates the architectural strengths of various advanced YOLO (You Only Look Once) models through bottleneck structure optimization and positional feature enhancement to achieve superior detection accuracy. Furthermore, we implemented a two-stage detection method augmented with transfer learning, which not only reduces training complexity but also significantly enhances the model's performance. Extensive evaluation of our custom-built dataset demonstrated Pod-Pose's superior performance, with the X variant achieving an Average Precision of 0.912 at an IoU threshold of 0.5 (AP@IoU = 0.5). Notably, four critical pod-related phenotypic traits were successfully quantified: pod length, bending length, curvature, and inflection point width. CONCLUSIONS: This study establishes Pod-Pose as a viable solution for pod phenotyping, with potential applications in soybean breeding optimization.
Why it matches plant phenotyping methods大豆莢の表現型を抽出する深層学習キーポイント検出モデルを開発し、精度評価と複数形質の定量化を行っており、植物フェノタイピング手法が研究の中心である。
abstractWe propose Pod-pose, an innovative top-down keypoint detection model for precise soybean pod phenotyping
Abstract has not been obtained from indexed metadata or an accessible article page.
Why it matches plant phenotyping methodsトウモロコシとダイズの有効葉面積指数（LAI）をUAV画像から推定する新規手法の開発が題名で明示されており、植物形質の取得・推定が中心である。
titleNeRF-LAI: A hybrid method combining neural radiance field and gap-fraction theory for deriving effective leaf area index of corn and soybean using multi-angle UAV images
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 6 Sept 2026
The use of drones has become a commonly used tool by plant scientists to aid in plant phenotyping endeavors. Iron deficiency chlorosis (IDC) is a commonly observed abiotic stress in soybean fields with high soil pH levels. IDC severity is visually classified, and recent work has shown that digital imaging techniques using both ground and UAS-acquired imagery can be utilized for automated severity ratings. In our study, we compared the classification accuracy of two flight altitudes to determine the optimal flight parameters for IDC phenotyping. In addition to this, we investigated the ability to use image-predicted scores for genome wide association study (GWAS), as well as the effect of IDC on traits such as canopy area and canopy growth and development. We also report a tool for semi-automated plot extraction from orthomosaic images that can be easily integrated with UAS. We noted that 43 days after planting was an ideal time for IDC severity ratings as the highest number of significant SNPs were reported at this timepoint.
Why it matches plant phenotyping methodsUAS画像によるIDC重症度フェノタイピングについて、飛行高度の分類精度を比較検証し、半自動プロット抽出ツールも提示しているため、手法が中心的である。
abstractwe compared the classification accuracy of two flight altitudes to determine the optimal flight parameters for IDC phenotyping.
Plant phenotyping relevance match · UnverifiedCrossref · OpenAlex · Europe PMC · checked 6 Sept 2026
High throughput phenotyping for crop monitoring at both leaf and canopy scales is essential for understanding plant responses to various stresses. PhenoGazer, a high-throughput phenotyping system, enhances crop monitoring in controlled environments by integrating a portable hyperspectral spectrometer with eight fiber optics, four Raspberry Pi cameras, and blue LED lights. This system allows for comprehensive assessment of plant health and development. PhenoGazer features automated moveable upper and lower racks for continuous measurements. The lower rack, equipped with four blue LED lights and spectrometer fiber optics, captures blue light-induced chlorophyll fluorescence at night. The upper rack, carrying four spectrometer fiber optics and cameras, captures hyperspectral reflectance and RGB images during the day. This dual capability enables detailed evaluation of plant phenology, stress responses, and growth dynamics throughout the entire crop growth cycle. Fully automated and managed by a Raspberry Pi running Python scripts, PhenoGazer ensures precise control and data acquisition with minimal human intervention. Additionally, it includes continuous measurements through a datalogger to acquire photosynthetically active radiation (PAR), soil moisture and temperature, and features expansion capability for additional analog or digital sensors as desired by end users. To test the system, soybean plants representing three conditions, healthy well watered, healthy droughted, and diseased, were monitored to evaluate growth and stress responses. PhenoGazer successfully phenotyped plants under different conditions in a walk-in growth chamber. By combining nighttime blue light induced chlorophyll fluorescence, hyperspectral reflectance-based vegetation indices, and RGB imagery, PhenoGazer represented a significant advancement in plant phenotyping technology, enhancing our understanding of crop responses to environmental conditions and supporting optimized crop performance in research and agricultural applications.
Why it matches plant phenotyping methods植物ストレス応答を測定する高スループット表現型解析システムの開発・技術的実証が中心であり、複数センサーと自動取得ワークフローを統合している。
abstractPhenoGazer, a high-throughput phenotyping system, enhances crop monitoring in controlled environments by integrating a portable hyperspectral spectrometer with eight fiber optics, four Raspberry Pi cameras, and blue LED lights.
Noninvasive analysis of pod phenotypic traits under field conditions is crucial for soybean breeding research. However, previous pod phenotyping studies focused on postharvest materials or were limited to indoor scenarios, failing to generalize to real-field environments. To address these issues, this paper employs an instance segmentation approach for the precise extraction of the pod area from multiplant RGB images in preharvest soybean fields. We first introduce a cost-effective workflow for constructing datasets of densely planted crop images with a uniform backdrop. Starting with video recording, high-quality static frames are collected by automatic selection. Then, a large vision model is explored to facilitate dense annotation and build a large-scale soybean dataset comprising 20k pod masks. Second, the pod instance segmentation model PodNet is developed based on the YOLOv8 architecture. We propose a novel hierarchical prototype aggregation strategy to fuse multiscale semantic features and a U-EMA prototype generation network to improve the model's perception performance for small objects. Comprehensive experiments suggest that lightweight PodNet achieves a superior mean average accuracy of 0.786 in the custom pod segmentation dataset. PodNet also performs competitively on in-field images without a backdrop and enables real-time inference on the edge computing platform. To the best of our knowledge, PodNet is the first pod instance segmentation model for preharvest fields. The low-cost and high-precision extraction of pods is not only a prerequisite for phenotypic analysis of the pod organs but also constitutes an important foundation in conducting cross-scale phenotyping from whole-plant to seed levels.
Why it matches plant phenotyping methods大豆莢の表現型形質を抽出する画像セグメンテーション手法を開発し、データセット、精度評価、エッジ推論まで扱っており、表現型取得が研究の中心である。
abstractNoninvasive analysis of pod phenotypic traits under field conditions is crucial for soybean breeding research.
The unmanned aerial vehicle (UAV) platform has emerged as a powerful tool in soybean (Glycine max (L.) Merr.) breeding phenotype research due to its high throughput and adaptability. However, previous studies have predominantly relied on statistical features like vegetation indices and textures, overlooking the crucial structural information embedded in the data. Feature fusion has often been confined to a one-dimensional exponential form, which can decouple spatial and spectral information and neglect their interactions at the data level. In this study, we leverage our team's cross-circling oblique (CCO) route photography and Structure-from-Motion with Multi-View Stereo (SfM-MVS) techniques to reconstruct the three-dimensional (3D) structure of soybean canopies. Newly point cloud deep learning models SoyNet and SoyNet-Res were further created with two novel data-level fusion that integrate spatial structure and color information. Our results reveal that incorporating RGB color and vegetation index (VI) spectral information with spatial structure information, leads to a significant reduction in root mean square error (RMSE) for yield estimation (22.55 kg ha⁻¹) and an improvement in F1-score for five-class lodging discrimination (0.06) at S7 growth stage. The SoyNet-Res model employing multi-task learning exhibits better accuracy in both yield estimation (RMSE: 349.45 kg ha⁻¹) when compared to the H2O-AutoML. Furthermore, our findings indicate that multi-task deep learning outperforms single-task learning in lodging discrimination, achieving an accuracy top-2 of 0.87 and accuracy top-3 of 0.97 for five-class. In conclusion, the point cloud deep learning method exhibits tremendous potential in learning multi-phenotype tasks, laying the foundation for optimizing soybean breeding programs.
Why it matches plant phenotyping methodsUAV・SfM-MVSによる植物キャノピー3D再構成と、収量推定・倒伏識別のための新規点群深層学習モデル開発が研究の中心であり、植物表現型取得・推定手法に該当する。
abstractIn this study, we leverage our team's cross-circling oblique (CCO) route photography and Structure-from-Motion with Multi-View Stereo (SfM-MVS) techniques to reconstruct the three-dimensional (3D) structure of soybean canopies.
High throughput phenotyping for crop monitoring at both leaf and canopy scales is essential for understanding plant responses to various stresses. PhenoGazer, a high-throughput phenotyping system, enhances crop monitoring in controlled environments by integrating a portable hyperspectral spectrometer with eight fiber optics, four Raspberry Pi cameras, and blue LED lights. This system allows for comprehensive assessment of plant health and development. PhenoGazer features automated moveable upper and lower racks for continuous measurements. The lower rack, equipped with four blue LED lights and spectrometer fiber optics, captures blue light-induced chlorophyll fluorescence at night. The upper rack, carrying four spectrometer fiber optics and cameras, captures hyperspectral reflectance and RGB images during the day. This dual capability enables detailed evaluation of plant phenology, stress responses, and growth dynamics throughout the entire crop growth cycle. Fully automated and managed by a Raspberry Pi running Python scripts, PhenoGazer ensures precise control and data acquisition with minimal human intervention. Additionally, it includes continuous measurements through a datalogger to acquire photosynthetically active radiation (PAR), soil moisture and temperature, and features expansion capability for additional analog or digital sensors as desired by end users. To test the system, soybean plants representing three conditions, healthy well watered, healthy droughted, and diseased, were monitored to evaluate growth and stress responses. PhenoGazer successfully phenotyped plants under different conditions in a walk-in growth chamber. By combining nighttime blue light induced chlorophyll fluorescence, hyperspectral reflectance-based vegetation indices, and RGB imagery, PhenoGazer represented a significant advancement in plant phenotyping technology, enhancing our understanding of crop responses to environmental conditions and supporting optimized crop performance in research and agricultural applications.
Why it matches plant phenotyping methods植物ストレス応答を測定する高スループット表現型解析システムの開発・実証が中心であり、複数センサーと自動取得ワークフローを統合している。
abstractPhenoGazer, a high-throughput phenotyping system, enhances crop monitoring in controlled environments by integrating a portable hyperspectral spectrometer with eight fiber optics, four Raspberry Pi cameras, and blue LED lights.
Modern plant phenomics leverages advanced digital technologies to derive qualitative and quantitative traits that define plant phenotypes, offering crucial insights for breeders and farmers in precision agriculture.However, real-field conditions, with their complexity and lack of flexibility, pose significant challenges for machine vision algorithm initially developed in controlled laboratory settings.The objective of this study was to explore the potential of human aided 3D point cloud analysis for phenotyping crop under near real-field conditions using a custom-built desktop application, with sorghum and soybean plants as the case study.Light detection and ranging (LiDAR) data acquisition was performed using a Leica BLK360 imaging laser scanner (Leica Geosystems AG, USA).Coordinate difference measurements were employed in extracting various phenotypic traits from plant point clouds.The sphere outlier removal (SOR) was fundamental in macro-noise reduction, while color-based scatter plot matrix were used for micro-noise isolation.The correlation between point cloud-derived traits and manually measured values was strong, with root mean square error (RMSE) of 17.84 mm for sorghum plant height, 16.28 mm for soybean plant height, 11.65 mm for sorghum panicle height, and 0.967 mm for sorghum stem diameter, and corresponding R-squared values between 0.7334 and 0.9492.However, measuring more complex traits like crown diameter, which are influenced by overlap and occlusion, was less accurate, with an RMSE of 102.4 mm and an R-squared value of 0.3702.While 3D phenotyping in near real-field environment reliably captures linear plant structures, complex morphological traits require improved occlusionhandling algorithms.Future work should prioritize high resolution sensors to capture finer details.Likewise, automated workflows are poised to improve not only throughput but the reliability and reproducibility of the 3D phenotyping approach.
Why it matches plant phenotyping methodsLiDAR点群とカスタムアプリケーションを用いた3D植物形態表現型の取得・解析手法を開発し、手動測定との精度検証も行っており、表現型測定法が研究の中心である。
abstractThe objective of this study was to explore the potential of human aided 3D point cloud analysis for phenotyping crop under near real-field conditions using a custom-built desktop application
The contents and ratios of 7S and 11S globulins are crucial for the nutritional value and functional properties of soybean proteins. Typically sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) is used to detect 7S and 11S globulin in soybeans however this method involves slow analysis procedures and high costs. Near-infrared (NIR) spectroscopy technology has emerged for detecting soybean protein content, enabling rapid non-destructive testing with advantages of convenient measurement, minimal sample processing requirements, and simultaneous determination of multiple components. To resolve the issue of shared quantitative prediction models between NIR spectroscopy-based 7S and 11S protein content predictions for various soybean seed and soybean powders, a transfer method of standard-free model based on transfer learning (TL) was proposed. Firstly, the NIR data of different forms of soybean samples were collected, and the near-infrared prediction models of 7S and 11S protein content were established. Secondly, the direct standardization (DS) and piecewise direct standardization (PDS) algorithms were improved to propose a DS-PDS-based model transfer method, with the influence of the sequence of preprocessing and model transfer algorithm on overall model transfer scheme was explored. Then, IRM is used to force the model to learn invariant features with causal relationship with labels by constraining the optimal classifier consistency of the model in different environments. Finally, aiming at standard sample sets corresponding to master-slave spectra required by traditional model transfer methods, the model transfer effect was investigated using a model transfer method based on standard-free migration learning. Results showed that the model transfer method based on without standard transfer learning was more suitable for 7S and 11S globulin content modeling between soybean seeds and soybean powders. It is intended to provide efficient and accurate 7S and 11S protein content detection methods for soybean processing enterprises and support quality control of soybean protein products and production of functional products.
Why it matches plant phenotyping methods大豆種子・粉末の7S/11Sタンパク質含量という種子形質を対象に、NIR分光モデルの転移学習、DS/PDS改良、標準試料不要のモデル転移を開発・検証しており、表現型取得・推定法が研究の中心である。
abstracta transfer method of standard-free model based on transfer learning (TL) was proposed
Soybean is a crucial global oilseed crop and a vital source of plant protein. As one of the world's largest consumers of soybeans, China heavily relies on soybean imports, making increased soybean yields an effective way to address the shortage of soybean resources. As soil salinization becoming increasingly severe, salt stress has become a major factor affecting soybean yield and quality in China. This paper proposes a deep learning framework for identifying salt stress levels in soybean seedlings using RGB images of their leaves. In this framework, a Convolutional Neural Network combined with a Convolutional Block Attention Module is used to extract image features; a dimensionality reduction method is employed to remove redundancy from the extracted features; and a machine learning classifier is used to classify the reduced features. Experimental results demonstrate that this framework can accurately identify salt stress levels from soybean leaf images while overcoming the overfitting problem associated with small datasets. Compared to existing traditional deep learning models, transfer learning models, and other frameworks, the proposed framework offers better classification performance and generalization ability.
Why it matches plant phenotyping methodsRGB葉画像からダイズ幼苗の塩ストレス状態を推定する深層学習フレームワークの開発・比較が中心であり、植物状態の画像ベースフェノタイピング手法に該当する。
abstractThis paper proposes a deep learning framework for identifying salt stress levels in soybean seedlings using RGB images of their leaves.
Soybean is an important oilseed crop, rich in protein and oil, often referred to as a ``cash crop'' or ``gold bean'' by Indian farmers. In Maharashtra, soybean cultivation spans over approximately 3.8 million hectares, producing 3.07 million tons, placing the state second in India for overall soybean production. However, despite of its significance, several issues such as weeds, diseases, and pests hamper the overall productivity of soybean. Addressing these challenges faced by soybean growers it is essential to enhance yield and improve the crop's overall potential Currently, the farming sector is transitioning towards Agriculture 5.0, also known as digital farming. This approach utilizes data-driven technologies, such as artificial intelligence and computer vision, to transform the agriculture sector. These technologies enable the automation of several farming tasks. To develop accurate and robust machine learning/deep learning models high quality datasets are needed. With this aim, we have created a comprehensive dataset of soybean crop images affected by diseases and pest attacks from original fields of Maharashtra region located in India. Data acquisition was conducted across two seasons through aerial as well as ground-based approaches. The dataset is enriched with 4 types of diseases and 1 pest attack. The proposed dataset will serve as a valuable resource for training and testing machine learning and deep learning models ,enabling accurate detection and classification of diseases and pests attack damage.
Why it matches plant phenotyping methods大豆の病害・害虫被害を対象とした航空・地上画像データセットの構築が中心で、植物の病害状態を画像から評価する再利用可能な資源であるため。
abstractwe have created a comprehensive dataset of soybean crop images affected by diseases and pest attacks from original fields of Maharashtra region located in India.
Reproduction assets foundThe paper's own soybean UAV and leaf image dataset is publicly deposited on Mendeley Data with an explicit direct URL and DOI, matching an allowed URL.Dataset · publicarashtra, India
Banawadi: Longitude 74.1943023 Latitude:17.3179823
Goware: Longitude 74.208238 Latitude:17.286995
Data accessibility
Repository name: Mendeley Data
MH-SoyaHealthVision: An Indian UAV and Leaf Image Dataset for Integrated Crop Health Assessment
Data identification number: 10.17632/hkbgh5s3b7.1
Direct URL to data: https://data.mendeley.com/datasets/hkbgh5s3b7/1
1
Value of the Data
•
The dataset uniquely combines UAV-based aerial images, offering high resolution and a broad spectrum, with ground-level close-up images. UAV imaging is effective for macro level field variability while ground-based images provide micro level finer details of symptoms such as leaf spots, lesions, andOpen asset ↗Mendeley Data · 10.17632/hkbgh5s3b7.1lines:1-55Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
SoybeanRootMorphology / geometry measurementRoot system architecture
Soybean drought tolerance relies on root traits. Genomic prediction (GP) offers a non-destructive alternative to laborious phenotyping. This study explores a multi-kernel GP approach for predicting soybean root traits by also incorporating easily measurable non-destructive aerial traits as secondary covariates. The main idea is to leverage the correlation between the aerial (visible) and root traits (not visible). In addition, we contrasted the predictive ability (PA) shown by the multi-kernel approach to those obtained from single-trait and multi-trait genomic prediction models. Data comprising 100 cultivars evaluated in two years and genotyped for 5,403 single-nucleotide polymorphism markers was analyzed. To comprehensively assess model performance, two cross-validation schemes were considered (CV1 and CV0). CV1 used a five-fold approach, and CV0 used a time-lagged cross-validation (i.e., data from years 1 and 2 were used for training and testing, respectively). Aerial traits added as covariates enhanced the GP predictive ability for all the traits and cross-validation (CV) schemes, outperforming single- and multi-trait models without this information. The inclusion of the interaction term between markers and secondary traits did not improved PA compared to the main effects models.
Why it matches plant phenotyping methods根形質を非破壊の地上部形質とゲノム情報から予測する統計的手法を開発・比較し、交差検証で性能を評価しているため、植物表現型推定法が中心である。
abstractThis study explores a multi-kernel GP approach for predicting soybean root traits by also incorporating easily measurable non-destructive aerial traits as secondary covariates.
Abstract Background Seed quality analysis using X-rays is increasingly explored due to its invasive and rapid nature. Yet, the current absence of reliable and standardised imaging protocols has led to contradictory effects of X-ray exposure in previous studies. Our work systematically investigated the effect of soft X-rays on a wide range of plant materials. Results The baseline of three germination categories was established across seven species before the application of soft X-ray exposure under controlled standard germination conditions. The high inter-varietal and inter-lot variabilities, in addition to the strong interaction between X-ray exposure with variety and lot, reinforced the need to consider genetic and seed quality aspects while evaluating the impacts of X-rays. A slight stimulative effect was observed on most of the species (bean, carrot, fennel, maize, radish, and ryegrass), notably, with a repeated reduction in ungerminated seeds. Intrinsic physical quality holds a crucial value where the minor negative impact observed in soybean originated from its degraded physical quality and not from X-ray exposure, hence, no destructive effects were detected. To understand whether seed size plays a significant role in a seed's response to exposure, linear regression models were built to predict 3D seed traits (volume) from 2D X-ray images. Yet, seed size did not explain the variation in responses to soft X-rays. However, the average density of the seven species explained both their natural germination ( p p Conclusion Soft X-ray exposure is non-destructive with a beneficial effect on germination but can be strongly influenced by underlying genetics and the physical quality of the tested seeds. This study adopted internationally-standardised germination procedures and tested the effect of soft X-rays across diverse botanical, genetic and seed quality profiles. This work addressed important gaps in evaluating X-ray impacts and proposed a robust design and well-examined radiography protocol for a proven non-destructive seed quality analysis.
Why it matches plant phenotyping methods軟X線ラジオグラフィーによる非破壊的な種子品質・3D形質推定プロトコルの検討と検証が中心的に含まれており、単なる発芽試験ではない。
abstractSeed quality analysis using X-rays is increasingly explored due to its invasive and rapid nature.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · bioRxiv · checked 6 Sept 2026
Abstract The stink bug complex is a major agricultural pest for soybean crops, significantly reducing productivity. Genetic resistance is the most effective control strategy, but its quantitative nature and labor-intensive phenotyping make its implementation in breeding programs challenging. This study explored high-throughput phenotyping (HTP) using unmanned aerial vehicles (UAVs) equipped with RGB cameras to evaluate a soybean population and identify stink bug resistance by correlating image-derived features and machine learning (ML) models. Using an alpha-lattice design with three replications, we phenotyped 304 soybean lines over two seasons under natural stink bug infestations. We manually evaluated five traits associated with stink bug resistance and correlated them with color, texture, and histogram features from aerial images. Three ML models—AdaBoost, SVM, and MLP— were tested to predict these traits. VIs, especially the Visible Atmospherically Resistant Index at the first percentile (VARI_P25) and texture-based indices at 45° and 135°, effectively predicted traits in stressed environments, particularly during flights near maturation. While ML models showed good predictive ability for yield, healthy seed weight, and maturity, they were less effective for stink bug resistance. Increasing the number of UAV flights modestly improved predictive accuracy, though predicting traits across different seasons remained challenging. Despite this, indices like VARI_25P were valuable for screening and excluding less promising genotypes, optimizing breeding program resources. This pioneering work offers valuable insights and highlights the need for further research to optimize resistance selection, promising significant advances in soybean breeding for stink bug resistance.
Why it matches plant phenotyping methodsUAV RGB画像と機械学習による形質推定を、ソイビーン育種集団の高スループット表現型解析として評価しており、画像特徴量・モデル性能・季節間予測を検討する方法中心の研究である。
abstractThis study explored high-throughput phenotyping (HTP) using unmanned aerial vehicles (UAVs) equipped with RGB cameras to evaluate a soybean population and identify stink bug resistance by correlating image-derived features and machine learning (ML) models.
Noninvasive analysis of pod phenotypic traits under field conditions is crucial for soybean breeding research. However, previous pod phenotyping studies focused on postharvest materials or were limited to indoor scenarios, failing to generalize to real-field environments. To address these issues, this paper employs an instance segmentation approach for the precise extraction of the pod area from multiplant RGB images in preharvest soybean fields. We first introduce a cost-effective workflow for constructing datasets of densely planted crop images with a uniform backdrop. Starting with video recording, high-quality static frames are collected by automatic selection. Then, a large vision model is explored to facilitate dense annotation and build a large-scale soybean dataset comprising 20k pod masks. Second, the pod instance segmentation model PodNet is developed based on the YOLOv8 architecture. We propose a novel hierarchical prototype aggregation strategy to fuse multiscale semantic features and a U-EMA prototype generation network to improve the model's perception performance for small objects. Comprehensive experiments suggest that lightweight PodNet achieves a superior mean average accuracy of 0.786 in the custom pod segmentation dataset. PodNet also performs competitively on in-field images without a backdrop and enables real-time inference on the edge computing platform. To the best of our knowledge, PodNet is the first pod instance segmentation model for preharvest fields. The low-cost and high-precision extraction of pods is not only a prerequisite for phenotypic analysis of the pod organs but also constitutes an important foundation in conducting cross-scale phenotyping from whole-plant to seed levels.
Why it matches plant phenotyping methods大豆莢の表現型抽出を目的に、データセット構築、インスタンスセグメンテーションモデル、実環境での性能評価を中心的に開発しているため。
abstractNoninvasive analysis of pod phenotypic traits under field conditions is crucial for soybean breeding research.
Reproduction assets foundThe authors open-source the field soybean pod instance segmentation dataset (488 images, 20k pod masks) and PodNet-related resources at their public GitHub repository, explicitly stated in the data availability statement.Dataset · publicefficiency of manual annotation. The average pod number per image is more than 56, and the total number of pod objects is greater than 20k. Fig. 7 (c) shows that most of the pods are located in the upper center region of the image. The field soybean pod instance segmentation dataset is open sourced for the research community at https://github.com/Boatsure/PodNet .
3.2.
Implementation and experiments of PodNet
Considering that instance segmentation is a computationally intensive task, this study selected the lightweight architecture YOLOv8-nano (v8n) as the baseline model for the development of PodNet. Model v8n has simplified module connections and competitive perception accuracy whileOpen asset ↗Boatsure/PodNetlines:96-104Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Nonphotochemical quenching (NPQ) is a critical photoprotective mechanism in plants, safeguarding photosystem II (PSII) and PSI from photodamage under abiotic stress. However, it is unclear if different stressors lead to similar NPQ phenotypes, and the magnitude of natural variation (between and within plant species) in NPQ response to abiotic stress is unknown. Testing a semi-high-throughput leaf-disc approach for examining the NPQ kinetics parameters, we investigated NPQ under chilling, drought and low nitrogen stress across multiple species and/or genotypes. Our results show substantial variation in NPQ phenotypes across species, genotypes and treatments. In C3 crops, tobacco and soybean, multiple NPQ parameters generally increased under chilling and drought, while in C4 crops, maize and sorghum, NPQ traits were more variable including a decrease of multiple NPQ parameters. Low-N stress revealed genotype- and developmental stage-specific effects on NPQ, potentially reflecting distinct adaptive strategies and regulatory changes in NPQ stress response. A significant effect of ecotype and stress treatment was detected on most NPQ kinetics traits in Arabidopsis thaliana , however, the interaction between ecotype and treatment was stronger in drought than in chilling. Differential regulation of NPQ could be associated with a combination of changes in proton motive, ATPase synthase activity, and PSI redox state. Our findings highlight that interpreting relative changes in NPQ under abiotic stress is inherently complex and demands a broader integration of physiological data across multiple regulatory layers.
Why it matches plant phenotyping methods半高速スループットの葉ディスク法を用いてNPQ動態形質を測定し、複数種・遺伝子型・ストレス条件で適用しているため、植物生理フェノタイピング手法の実質的応用と判断します。
abstractTesting a semi-high-throughput leaf-disc approach for examining the NPQ kinetics parameters, we investigated NPQ under chilling, drought and low nitrogen stress across multiple species and/or genotypes.
Root system architecture (RSA) underpins plant access to water and nutrients, making its characterization critical for improving crop performance in environments with limited soil fertility. However, current methods for quantifying root features face several challenges. They may rely on 2D images that suffer from occlusion, use expensive sensing technologies like X-ray computed tomography, or depend on 3D modeling approaches with assumptions about branching that make them difficult to generalize. To address these challenges, we introduce an open-source Python framework for quantifying RSA samples from 3D point clouds generated from low-cost photogrammetry. Critically, this method incorporates no assumptions about taxon-specific branching orientation, making it both well-suited for modeling naturally grown annual dicots such as soybean and generalizable across species. Using field-grown soybean as a test case, we demonstrate the utility of this framework to extract biologically meaningful 3D features of divergent root systems sampled across developmental stages and soil environments, and enable new analyses not possible with 2D approaches, such as modeling metabolic scaling relationships. Results indicate that, in our soybean samples, while certain individual features like taproot tortuosity are potentially influenced by the soil environment, and while roots in sandy loam exhibited greater feature plasticity, fundamental scaling properties remain consistent. By combining low-cost photogrammetry with 3D reconstruction of root systems from point clouds, this approach provides the plant science community with new opportunities for more comprehensive root studies.
Why it matches plant phenotyping methods植物根系構造を3D点群から定量化するオープンソース手法の開発が中心であり、低コスト写真測量と3D再構成による形態形質抽出を実証している。
abstractwe introduce an open-source Python framework for quantifying RSA samples from 3D point clouds generated from low-cost photogrammetry.
To address the issues of low soybean self-sufficiency and drought - related production constraints, multispectral imaging can non - invasively detect crop characteristics, while deep convolutional networks identify drought from multispectral images. However, large models face mobility limitations due to high computational requirements.This study presents a mobile detection approach integrating ReliefF feature screening and lightweight convolutional neural networks, and develops the "Early Acknowledgment for Soybean Drought" App. It embeds a lightweight model that optimizes 37 - dimensional soybean canopy multispectral features via ReliefF and uses a three - layer 1D convolutional network for drought identification.The model achieves 96.88% classification accuracy on the self - built dataset, with an inference time of 18 ms, a size under 30 MB, and less than 60 MB memory usage on mobile. The APP integrates the multispectral camera SDK and PyTorch inference engine, enabling real - time spectral analysis. Field tests show its one - button operation, low learning curve for farmers, and significant water - saving and yield - increasing effects, offering a lightweight, high - precision mobile solution for soybean drought management and promoting smart agriculture development.
Why it matches plant phenotyping methodsマルチスペクトル画像からダイズの干ばつ状態を推定する軽量深層学習モデルとモバイルアプリを開発しており、植物状態の取得・推定手法が研究の中心である。
abstractmultispectral imaging can non - invasively detect crop characteristics, while deep convolutional networks identify drought from multispectral images
Salinity is a significant factor limiting the cultivation of soybean, a globally important cash crop. However, efficient assessment and genetic dissection of soybean response to salt stress remain challenging. This study leveraged high-throughput phenotyping (HTP) and traditional physiological methods for comprehensive phenotyping of salt tolerance using 261 diverse soybean germplasms and dissected the genetic basis through GWAS. A highly efficient rail-based HTP system with depth-sensing and RGB cameras was developed to collect horizontal and vertical growth and leaf health information. Machine learning pipeline facilitated canopy detection, segmentation, and phenotype extraction processes. Three HTP traits and five traditional physiological traits related to salt tolerance were collected. Divergence between growth status and chlorophyll content was observed, indicating the importance of HTP and the genetic complexity of salt tolerance in soybean. A stepwise regression analysis indicated that "Vegetation color index" (VEG), "Anthocyanin Reflectance Index" (ARI), and "Cyan, Magenta, Yellow" (CMY_Yellow) are the most informative indices of soybean foliar health under salt tolerance. GWAS identified 46 loci for salt tolerance-related traits. Fifteen potential candidate genes were proposed, including Glyma.18g238700 which is known to be involved in salt tolerance mechanisms. Field test indicated that two of the top five tolerant accessions at seedling stage are salt tolerant at full growth stages with high yield potential. Additionally, best crosses were predicted from random mating of the association panel by using linkage and independent assortment models for salt tolerance improvement breeding. This study provided tolerant genotypes, promising candidates and optimized crosses for further exploration.
Why it matches plant phenotyping methods塩耐性の表現型取得を目的に、深度・RGBカメラ搭載のHTPシステムを開発し、機械学習でキャノピー検出・分割・形質抽出を行っており、フェノタイピング手法が研究の中心である。
abstractA highly efficient rail-based HTP system with depth-sensing and RGB cameras was developed to collect horizontal and vertical growth and leaf health information.
SoybeanTomatoWatermelonField / plotFruitPhysiological trait estimationStress / disease detectionGrowth / development / phenologyStomatal traitsWater status / transpiration
The integration of flexible electronics with plant science has generated various plant-wearable sensors, yet challenges persist in their application to real-world agriculture, particularly in high-throughput settings. Overcoming the trade-off between sensing sensitivity and range, adapting sensors to a wide range of crop types, and bridging the gap between sensor measurements and biological understandings remain primary obstacles. Here, we introduce PlantRing, an innovative, nano-flexible sensing system designed to address these challenges. PlantRing employs bio-sourced carbonized silk georgette as the strain-sensing material, offering an exceptional detection limit (0.03%–0.17% strain, depending on sensor model), high stretchability (tensile strain up to 100%), and remarkable durability (season-long use). PlantRing effectively monitors plant growth and water status by measuring organ circumference dynamics, performing reliably under harsh conditions, and adapting to a wide range of plant species. Applying PlantRing to study fruit cracking in tomato and watermelon has revealed a novel hydraulic mechanism characterized by genotype-specific excess sap flow within the plant to fruiting branches. Its high-throughput application has enabled large-scale quantification of stomatal sensitivity to soil drought—a long-standing aspiration in plant biology—facilitating the selection of drought-tolerant germplasm. Combining PlantRing with a soybean mutant has led to the discovery of a potential novel function of the circadian clock gene GmLNK2 in stomatal regulation. More practically, integrating PlantRing into feedback irrigation achieves simultaneous water conservation and quality improvement, signifying a paradigm shift from reliance on experience or environmental cues to plant-based feedback control. Collectively, PlantRing represents a groundbreaking tool poised to revolutionize botanical studies, agriculture, and forestry.
Why it matches plant phenotyping methods植物器官周径を測定するウェアラブル高スループットセンサーを開発し、成長・水分状態・気孔感度などの表現型を定量化する方法が研究の中心である。
abstractHere, we introduce PlantRing, an innovative, nano-flexible sensing system designed to address these challenges.
In this study, an effective method for detecting purine in soybean seeds and products by ultra-micro spectrophotometry (UMS) was established. Addition of 35 % HClO₄ to 0.1 g of soybean flour and incubation in a water bath at 100 °C for 30 min and adjustment of the pH to 4 was optimal for purine hydrolysis and extraction of the total purine content of soybean seeds. No significant difference (P > 0.05) in total purine detection was observed between UMS and UPLC, and the relative standard deviation was less than 3 %, ensuring optimal precision and accuracy. Seed from 2083 soybean accessions grown in two years were analysed using UMS. Total purine content ranged from 65.49 to 379.84 mg/100 g, with 5.8-fold variation, and 11 elite accessions with under 100 mg/100 g of total purine were identified. The agronomic traits, environmental factors, and geographical significantly affected soybean total purine content. Furthermore, total purine content correlated positively with seed protein (r = 0.342∗∗∗) and negatively with oil (r = −0.330∗∗∗). The geographical distribution revealed that accessions from Southern and Northern China contain lower amounts of total purine content. This study provides a technical and material foundation for breeding low purine soybean.
Why it matches plant phenotyping methodsダイズ種子の総プリン含量という育種対象の種子形質について、超微量分光法を開発し、UPLCとの比較で精度・正確性を検証しているため、測定法が研究の中心である。
abstractan effective method for detecting purine in soybean seeds and products by ultra-micro spectrophotometry (UMS) was established
Early detection of nutrient deficiencies is crucial for optimizing crop yields and ensuring sustainable agricultural practices. This study presents a novel application of the YOLOv8s object detection model for identifying nitrogen, phosphorus, and potassium deficiencies in soybean plants. Employing a unique dataset from a long-term nutrient-deficient field maintained for over 40 years, we trained and evaluated the model on 6,020 red, green, and blue images of soybean leaves exhibiting nutrient stress conditions. The YOLOv8s model achieved exceptional performance, with a mean average precision (mAP@0.5) of 99.18% during training and 98.51% for validation. Precision rates for individual nutrient deficiencies ranged from 90.03 to 96.54%, with highly accurate potassium deficiency detection. The model demonstrated robust generalization across diverse field conditions, processing images in 3.46 ms each, making it suitable for real-time applications. This research significantly advances the field of precision agriculture by providing a fast, accurate, and scalable method for detecting early nutrient deficiency in soybean crops, potentially revolutionizing fertilizer management practices and contributing to more sustainable farming systems.
Why it matches plant phenotyping methods大豆葉の栄養欠乏状態を画像から検出するYOLOv8s手法を開発・評価しており、表現型取得とモデル性能検証が研究の中心である。
abstractThis study presents a novel application of the YOLOv8s object detection model for identifying nitrogen, phosphorus, and potassium deficiencies in soybean plants.
Background: Soybean is one of the important leguminous crops grown mainly in the middle states of India. Soybean plants are prone to leaf diseases like spot and bacterial blight. Early detection of such diseases is an important task for the farmers to avoid loss in production. In this background, deep learning techniques are used for identifying the diseases in leaves of Soybean plants. Methods: This study investigates the utilization of convolutional autoencoder in extracting features from the lesion leaf images. Images are preprocessed and converted to latent space features using convolutional autoencoder. Multinomial logistic regression model is employed over the extracted features to find the type of disease in Soybean plants. Result: The experimental result shows that convolutional autoencoder model along with multinomial logistic regression model achieves an average accuracy of 92% in disease identification. In addition to providing disease identification, the ideas in this paper provide support for food security through the application of deep learning techniques.
Why it matches plant phenotyping methods大豆葉の病斑画像から病害状態を推定する画像・機械学習手法が研究の中心であり、植物病害表現型の取得・分類に該当する。
abstractdeep learning techniques are used for identifying the diseases in leaves of Soybean plants
As one of the important indicators of soybean seed quality identification, the appearance of soybeans has always been of great concern to people, and in traditional detection, it is mainly through the naked eye to check whether there are defects on its surface. The field of machine learning, particularly deep learning technology, has undergone rapid advancements and development, making it possible to detect the defects of soybean seeds using deep learning technology. This method can effectively replace the traditional detection methods in the past and reduce the human resources consumption in this work, leading to decreased expenses associated with agricultural activities. In this paper, we propose a Yolov9-c-ghost-Forward model improved by introducing GhostConv, a lightweight convolutional module in GhostNet, which enhances the recognition of soybean seed images through grayscale conversion, filtering processing, image segmentation, morphological operations, etc. and greatly reduces the noise in them, to separate the soybean seeds from the original images. Based on the Yolov9 network, the soybean seed features are extracted, and the defects of soybean seeds are detected. Based on the experiments' findings, the recall rate can reach 98.6%, and the mAP0.5 can reach 99.2%. This shows that the model can provide a solid theoretical foundation and technical support for agricultural breeding screening and agricultural development.
Why it matches plant phenotyping methods大豆種子表面の欠陥という植物器官の状態を、画像処理と改良YOLOv9で抽出・検出する手法開発が研究の中心であり、性能評価も行っている。
abstractwe propose a Yolov9-c-ghost-Forward model improved by introducing GhostConv, a lightweight convolutional module in GhostNet, which enhances the recognition of soybean seed images through grayscale conversion, filtering processing, image segmentation, morphological operations, etc.
Reproduction assets foundThe paper's soybean seed defect detection study uses a public Kaggle dataset of 4,388 soybean seed images (intact, broken, skin-damaged, spotted), explicitly stated in the Data Availability statement. No author analysis code or trained model checkpoints are deposited; makesense.ai is a generic annotation tool, not a deDataset · publicuthors reviewed and approved the final manuscript.
Funding
This work was supported in part by the Special Support Plan for High level Talents in Zhejiang Province (2021R52019), and the Education Department of Hainan Province (Hnky2024-18).
Data availability
The data used in this article can be downloaded from the following link https://www.kaggle.com/datasets/warcoder/soyabean-seeds .
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Xia Yu, Email: 100170@hainnu.edu.cn.
Qi Dai, Email: daiqi@zstu.edu.cn.
ReferenOpen asset ↗Kaggle · warcoder/soyabean-seedslines:388-411Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 13 Sept 2026
Rapid detection of crop grain components is crucial for effective production and energy conversion. We used the sample set division method to divide multiple sample sets and optimize NIRS models for rapid prediction of protein and fat content. 1243 and 415 crop grain samples were screened and divided into 5 and 4 sets, respectively. The aim was to establish NIRS models for protein and fat content prediction. The best modeling methods for protein were N (Norris Derivative)+D (detrending)-C (CARS)-P (PLS) and N+M (MC-UVE)-C-P, while those for fat were N+M-C-P and N+S (Savitzky-Golay)-C-P. The SS (Soybean Set), KS (Sorghum Set), and FS (Full Samples Set) data sets provided accurate protein content analysis, while the FS and SS data sets were suitable for both protein content prediction and evaluation. For fat, the FS, SS, and CS (Cereal Set) models met content analysis requirements, with the FS model suitable for external validation. It compared and analyzed the fitness, robustness, and accuracy of different NIRS set models, employing various division methods in this study, which provided a new idea of green method theoretical and technical support for major component rapid detection of biomass raw materials.
Why it matches plant phenotyping methods作物穀粒のタンパク質・脂肪という種子形質をNIRSで迅速推定するモデルを開発し、適合性・頑健性・精度を比較評価しており、形質取得法が研究の中心です。
abstractThe aim was to establish NIRS models for protein and fat content prediction.
The rapid and accurate identification of soybean diseases is critical for optimizing both yield and quality. Traditional image recognition techniques face notable limitations in terms of generalization and accuracy, particularly when tasked with identifying small-scale targets or distinguishing diseases with similar characteristics in large, heterogeneous, and complex environments. To address these challenges, this study proposes the YOLOv8-DML model for soybean leaf disease recognition. Building upon YOLOv8n, this model integrates a DWR module that replaces the high-level C2f module with C2f-DWR, enhancing feature extraction across varied receptive fields. Additionally, modifications to the neck structure incorporate a Multi-scale Enhanced Feature Pyramid (MEFP), which improves detection performance across targets of varying sizes by enabling effective multi-scale information fusion. A lightweight detection head (LSCD) is further introduced to facilitate multiscale feature interactions while reducing the overall model parameter count. Lastly, the WIoUv3 loss function is employed to place greater emphasis on small targets and moderate-quality samples, thereby enhancing detection precision. Experimental results demonstrate that YOLOv8-DML achieves a mAP50 of 96.9%, marking a 1.8% improvement over the original YOLOv8 algorithm, while also achieving an 18.6% reduction in parameters. Comparative analysis with other mainstream object detection models indicates that YOLOv8-DML delivers superior overall performance, highlighting its significant potential for effective soybean leaf disease identification.
Why it matches plant phenotyping methods大豆葉の病害状態を画像から認識するYOLOv8改良モデルを開発し、他モデルとの性能比較・検証を行っているため、植物病害フェノタイピング手法が中心である。
abstractthis study proposes the YOLOv8-DML model for soybean leaf disease recognition.
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
Crop models are essential for evaluating the effects of climate change on crop yields, optimizing agronomic practices, and guiding policy decisions to enhance food security. However, using traditional crop models, including both process-based and statistical models, for regional applications presents significant challenges. Process-based crop models often require extensive, locally-sensed inputs to drive the models, which are generally lacking at the regional level. Meanwhile, statistical crop models depend heavily on training data, but it is often difficult, or even impossible, to find high-quality training data on a large scale. Solar-induced chlorophyll fluorescence (SIF), a more physiologically based proxy for gross primary production (GPP), has shown good potential for estimating GPP and crop yield. We developed a practical SIF-based crop model driven by satellite SIF observations and three readily available datasets: air temperature, vapor pressure deficit, and soil moisture content. The key improvement of our research is to parameterize the fraction of open PSII reaction centers (qL) for crops, and incorporate variations in qL into the SIF-based estimation of crop GPP and yield. Using a leaf-level measurement system, we provided parameters for qL in corn and soybean. We showed that the simulated qL closely matches the measured qL, with R² > 0.95 and RMSE <0.05, even under conditions of high light and/or high temperature, whereas the performance of SIF alone significantly decreased under stress. By using SIF and qL within the mechanistic light response model, one can accurately estimate crop GPP without the need to parameterize various plant physiological processes or nutrient dynamics and management practices. This improvement substantially simplifies the model, reduces the need for driving variables and calibration data, and minimizes associated uncertainties. We applied the model to estimate corn and soybean yields in the U.S. Midwest for the period 2018–2023. A comparison with eddy covariance-based GPP measurements reveals that the simulated GPP accounts for 85 % of the variability in daily observed GPP for corn and 81 % for soybean. The model's performance at the regional scale was assessed by comparing it against county-level crop yield statistics. On average, the model captures 78 % of the county-level yield variability across more than 700 counties during the study period, achieving 76 % for corn and 81 % for soybean, with RMSE values of 14.47 Bu/Acre, and 4.09 Bu/Acre, respectively. The practical, yet mechanistic, SIF-based model introduced in this study represents a significant advance in regional and national crop yield estimation.
Why it matches plant phenotyping methodsSIF衛星観測とqL測定を統合した作物生理・収量推定モデルを開発し、葉レベル測定およびGPP・郡別収量で技術検証しており、植物状態の取得・推定手法が中心である。
abstractWe developed a practical SIF-based crop model driven by satellite SIF observations and three readily available datasets: air temperature, vapor pressure deficit, and soil moisture content.
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.
Intercropping is a key cultivation strategy for safeguarding national food and oil security. Accurate early-stage yield prediction of intercropped soybeans is essential for the rapid screening and breeding of high-yield soybean varieties. As a widely used technique for crop yield estimation, the accuracy of 3D reconstruction models directly affects the reliability of yield predictions. This study focuses on optimizing the 3D reconstruction process for intercropped soybeans to efficiently extract canopy structural parameters throughout the entire growth cycle, thereby enhancing the accuracy of early yield prediction. To achieve this, we optimized image acquisition protocols by testing four imaging angles (15°, 30°, 45°, and 60°), four plant rotation speeds (0.8 rpm, 1.0 rpm, 1.2 rpm, and 1.4 rpm), and four image acquisition counts (24, 36, 48, and 72 images). Point cloud preprocessing was refined through the application of secondary transformation matrices, color thresholding, statistical filtering, and scaling. Key algorithms—including the convex hull algorithm, voxel method, and 3D α-shape algorithm—were optimized using MATLAB, enabling the extraction of multi-dimensional canopy parameters. Subsequently, a stepwise regression model was developed to achieve precise early-stage yield prediction for soybeans. The study identified optimal image acquisition settings: a 30° imaging angle, a plant rotation speed of 1.2 rpm, and the collection of 36 images during the vegetative stage and 48 images during the reproductive stage. With these improvements, a high-precision 3D canopy point-cloud model of soybeans covering the entire growth period was successfully constructed. The optimized pipeline enabled batch extraction of 23 canopy structural parameters, achieving high accuracy, with linear fitting R2 values of 0.990 for plant height and 0.950 for plant width. Furthermore, the voxel volume-based prediction approach yielded a maximum yield prediction accuracy of R2 = 0.788. This study presents an integrated 3D reconstruction framework, spanning image acquisition, point cloud generation, and structural parameter extraction, effectively enabling early and precise yield prediction for intercropped soybeans. The proposed method offers an efficient and reliable technical reference for acquiring 3D structural information of soybeans in strip intercropping systems and contributes to the accurate identification of soybean germplasm resources, providing substantial theoretical and practical value.
Why it matches plant phenotyping methodsインタクロップ大豆の3D画像取得・点群処理・形質抽出パイプラインを中心に最適化し、構造形質の精度検証と収量予測まで行っているため、植物フェノタイピング手法研究に該当する。
abstractThis study focuses on optimizing the 3D reconstruction process for intercropped soybeans to efficiently extract canopy structural parameters throughout the entire growth cycle
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · OpenAlex · checked 6 Sept 2026
High-quality 3D reconstruction and accurate 3D organ segmentation of plants are crucial prerequisites for automatically extracting phenotypic traits. In this study, we first extract a dense point cloud from implicit representations, which derives from reconstructing the maize plants in 3D by using the Nerfacto neural radiance field model. Second, we propose a lightweight point cloud segmentation network (PointSegNet) specifically for stem and leaf segmentation. This network includes a Global-Local Set Abstraction (GLSA) module to integrate local and global features and an Edge-Aware Feature Propagation (EAFP) module to enhance edge-awareness. Experimental results show that our PointSegNet achieves impressive performance compared to five other state-of-the-art deep learning networks, reaching 93.73%, 97.25%, 96.21%, and 96.73% in terms of mean Intersection over Union (mIoU), precision, recall, and F1-score, respectively. Even when dealing with tomato and soybean plants, with complex structures, our PointSegNet also achieves the best metrics. Meanwhile, based on the principal component analysis (PCA), we further optimize the method to obtain the parameters such as leaf length and leaf width by using PCA principal vectors. Finally, the maize stem thickness, stem height, leaf length, and leaf width obtained from our measurements are compared with the manual test results, yielding R 2 values of 0.99, 0.84, 0.94, and 0.87, respectively. These results indicate that our method has high accuracy and reliability for phenotypic parameter extraction. This study throughout the entire process from 3D reconstruction of maize plants to point cloud segmentation and phenotypic parameter extraction, provides a reliable and objective method for acquiring plant phenotypic parameters and will boost plant phenotypic development in smart agriculture.
Why it matches plant phenotyping methods3D再構成、茎葉分割、形質抽出を一体化した植物フェノタイピング手法を開発し、他手法および手測定と比較検証している。
abstractwe propose a lightweight point cloud segmentation network (PointSegNet) specifically for stem and leaf segmentation.
SoybeanTomatoWatermelonFruitPhysiological trait estimationGrowth / development / phenologyStomatal traitsWater status / transpiration
The integration of flexible electronics with plant science has generated various plant-wearable sensors, yet challenges persist in their application to real-world agriculture, particularly in high-throughput settings. Overcoming the trade-off between sensing sensitivity and range, adapting sensors to a wide range of crop types, and bridging the gap between sensor measurements and biological understandings remain primary obstacles. Here, we introduce PlantRing, an innovative, nano-flexible sensing system designed to address these challenges. PlantRing employs bio-sourced carbonized silk georgette as the strain-sensing material, offering an exceptional detection limit (0.03%-0.17% strain, depending on sensor model), high stretchability (tensile strain up to 100%), and remarkable durability (season-long use). PlantRing effectively monitors plant growth and water status by measuring organ circumference dynamics, performing reliably under harsh conditions, and adapting to a wide range of plant species. Applying PlantRing to study fruit cracking in tomato and watermelon has revealed a novel hydraulic mechanism characterized by genotype-specific excess sap flow within the plant to fruiting branches. Its high-throughput application has enabled large-scale quantification of stomatal sensitivity to soil drought-a long-standing aspiration in plant biology-facilitating the selection of drought-tolerant germplasm. Combining PlantRing with a soybean mutant has led to the discovery of a potential novel function of the circadian clock gene GmLNK2 in stomatal regulation. More practically, integrating PlantRing into feedback irrigation achieves simultaneous water conservation and quality improvement, signifying a paradigm shift from reliance on experience or environmental cues to plant-based feedback control. Collectively, PlantRing represents a groundbreaking tool poised to revolutionize botanical studies, agriculture, and forestry.
Why it matches plant phenotyping methodsPlantRingは植物器官の周径変化を測定し、成長・水分状態・気孔感度などの表現型を高スループットに取得するウェアラブルセンサーシステムであり、センサー開発と実証が研究の中心です。
abstractHere, we introduce PlantRing, an innovative, nano-flexible sensing system designed to address these challenges.
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.
The unmanned aerial vehicle (UAV) platform has emerged as a powerful tool in soybean (Glycine max (L.) Merr.) breeding phenotype research due to its high throughput and adaptability. However, previous studies have predominantly relied on statistical features like vegetation indices and textures, overlooking the crucial structural information embedded in the data. Feature fusion has often been confined to a one-dimensional exponential form, which can decouple spatial and spectral information and neglect their interactions at the data level. In this study, we leverage our team's cross-circling oblique (CCO) route photography and Structure-from-Motion with Multi-View Stereo (SfM-MVS) techniques to reconstruct the three-dimensional (3D) structure of soybean canopies. Newly point cloud deep learning models SoyNet and SoyNet-Res were further created with two novel data-level fusion that integrate spatial structure and color information. Our results reveal that incorporating RGB color and vegetation index (VI) spectral information with spatial structure information, leads to a significant reduction in root mean square error (RMSE) for yield estimation (22.55 kg ha -1 ) and an improvement in F1-score for five-class lodging discrimination (0.06) at S7 growth stage. The SoyNet-Res model employing multi-task learning exhibits better accuracy in both yield estimation (RMSE: 349.45 kg ha -1 ) when compared to the H2O-AutoML. Furthermore, our findings indicate that multi-task deep learning outperforms single-task learning in lodging discrimination, achieving an accuracy top-2 of 0.87 and accuracy top-3 of 0.97 for five-class. In conclusion, the point cloud deep learning method exhibits tremendous potential in learning multi-phenotype tasks, laying the foundation for optimizing soybean breeding programs.
Why it matches plant phenotyping methodsUAV・SfM-MVSによるダイズ群落の3D構造再構成と、収量推定・倒伏判別のための専用深層学習モデル開発が研究の中心であり、再利用可能な表現型取得・推定手法に該当する。
abstractIn this study, we leverage our team's cross-circling oblique (CCO) route photography and Structure-from-Motion with Multi-View Stereo (SfM-MVS) techniques to reconstruct the three-dimensional (3D) structure of soybean canopies.
Reproduction assets foundThe article's Data availability statement explicitly says the code and data used in the study (soybean UAV point cloud phenotyping, SoyNet/SoyNet-Res models, yield/lodging analysis) are publicly downloadable from the authors' GitLab repository.Code · publicData availability
The code and data mentioned in the article can be downloaded from https://gitlab.com/zlyzly28/plant-phenomics .Open asset ↗gitlab.com/zlyzly28/plant-phenomicslines:588-659Plant phenotyping relevance match · UnverifiedEurope PMC · checked 14 Sept 2026
This study introduces a plant sensor utilizing an array of microneedles to monitor hydrogen peroxide (H 2 O 2 ) in tobacco and soybean plants under biotic stress response. The microneedle array features a biohydrogel layer composed of the natural biopolymer chitosan (Cs) and reduced graphene oxide (rGO), functionalized with horseradish peroxidase (HRP) (HRP/Cs-rGO). This HRP/Cs-rGO biohydrogel combines biocompatibility, hydrophilicity, porosity, and electron transfer ability, making it a suitable bioelectrode material for an electrochemical sensor. The sensor detects H 2 O 2 through the catalytic reaction of the enzyme, either by direct attachment to the plant leaf with the inserted microneedle or by exposure to the solution extracted from plant parts such as leaves. Utilizing chronoamperometry, the sensor demonstrates high sensitivity of 14.7 μA/μM across a concentration range of 0.1-4500 μM with a low detection limit of 0.06 μM. The sensor enables rapid detection of H 2 O 2 levels by exposing the sensor to extracted leaf solutions. For in situ measurements within the leaf, results are obtained in approximately 1 min, eliminating the need for sample preparation. H 2 O 2 levels in leaves following bacterial pathogen inoculation are evaluated alongside results from qualitative histological staining and quantitative fluorescence-based Amplex Red Assay, validating the ability of the sensor to detect changes in H 2 O 2 concentrations during plant defense responses. This sensor technology has the potential to function as a portable device for on-site measurement of reactive oxygen species in plants, providing a rapid and cost-effective solution for H 2 O 2 quantification.
Why it matches plant phenotyping methods植物体内の過酸化水素濃度という生理状態を測定するマイクロニードル電気化学センサーを開発し、既存アッセイ等で妥当性を検証しており、測定手法が研究の中心である。
abstractThis study introduces a plant sensor utilizing an array of microneedles to monitor hydrogen peroxide (H 2 O 2 ) in tobacco and soybean plants under biotic stress response.
The real-time and non-invasive visualization and quantification of symbiotic nitrogen fixation (SNF) in nodulated roots of soybean plants using Positron Emission Tomography (PET) imaging, coupled with the application of [ 13 N]N 2 gas as a PET radiotracer, has been explored in only a few studies. In these studies, [ 13 N]N 2 was delivered to nodulated soybean roots suspended in air within gas-tight acrylic boxes, followed by two-dimensional (2D) PET imaging to visualize the assimilated [ 13 N]N 2 in the air-suspended root nodules. In this paper, we introduce the In-Media Plant PET Root Imaging System (IMP 2 RIS), a novel gas delivery system designed and constructed in-house. Unlike the previous methods, IMP 2 RIS allows for non-intrusive delivery and exposure of [ 13 N]N 2 gas to the nodulated roots of soybean plants grown in a clay-rich, soil-like and visually opaque growth medium. This advancement enabled in-soil, three-dimensional (3D) visualization of SNF in soybean root nodules using Sofie, a preclinical PET scanner. Equipped with automated controls, IMP 2 RIS ensures ease of operation and operator safety during the [ 13 N]N 2 delivery process. We describe the components and functionalities of IMP 2 RIS, supported by experimental results showcasing its successful application in efficient delivery and exposure of [ 13 N]N 2 gas to nodulated roots of three soybean plant cultivars that vary in rates of N 2 fixation. The in-soil quantitative PET imaging of SNF, aided by IMP 2 RIS, holds promise for enhancing the integration of SNF as a functional phenotypic trait into breeding programs, aiming to enhance SNF efficiency by identifying breeding materials with high SNF capacities.
Why it matches plant phenotyping methods根圏内の窒素固定という植物生理形質をPETで可視化・定量するためのガス供給システムを開発し、実験的に適用しており、フェノタイピング手法が研究の中心である。
abstractwe introduce the In-Media Plant PET Root Imaging System (IMP 2 RIS), a novel gas delivery system designed and constructed in-house.
Digital phenotyping is a fast-growing area of hardware and software research and development. Phenotypic studies usually require determining whether there is a difference in some trait between plants with different genotypes or under different conditions. We developed StatFaRmer, a user-friendly tool tailored for analyzing time series of plant phenotypic parameters, ensuring seamless integration with common tasks in phenotypic studies. For maximum versatility across phenotypic methods and platforms, it uses data in the form of a set of spreadsheets (XLSX and CSV files). StatFaRmer is designed to handle measurements that have variation in timestamps between plants and the presence of outliers, which is common in digital phenotyping. Data preparation is automated and well-documented, leading to customizable ANOVA tests that include diagnostics and significance estimation for effects between user-defined groups. Users can download the results from each stage and reproduce their analysis. It was tested and shown to work reliably for large datasets across various experimental designs with a wide range of plants, including bread wheat (Triticum aestivum), durum wheat (Triticum durum), and triticale (× Triticosecale); sugar beet (Beta vulgaris), cocklebur (Xanthium strumarium) and lettuce (Lactuca sativa), corn (Zea mays) and sunflower (Helianthus annuus), and soybean (Glycine max). StatFaRmer is created as an open-source Shiny dashboard, and simple instructions on installation and operation on Windows and Linux are provided.
Why it matches plant phenotyping methods植物フェノタイピングの時系列データ解析を目的とするオープンソースShinyダッシュボードを開発し、データ準備・統計解析・再現可能なワークフローを提供しており、方法・ソフトウェアが中心である。
abstractWe developed StatFaRmer, a user-friendly tool tailored for analyzing time series of plant phenotypic parameters
Reproduction assets foundThe paper's authors publicly release StatFaRmer, an open-source R Shiny dashboard for phenotyping data analysis, via GitHub with installation instructions and a sample phenotypic dataset, and host a live deployment on shinyapps.io.Code · publicThe resulting tool can be accessed at 9 https://github.com/Stathmin/StatFaRmer ), with the instructions on installation and the sample dataset provided.Open asset ↗Stathmin/StatFaRmerlines:521-528Dataset · publicA sample dataset of different plant species (bread wheat ( Triticum aestivum ), durum wheat ( Triticum durum ), and triticale (× Triticosecale )), cultivars (35 variants) and plant genotypes (allelic state of 3 genes), with different treatments (3 variants), and the time series of morphological and spectral parameters of these plants is loaded in this tool as an example and available on GitHub.Open asset ↗lines:340-350Code / dataset availability confirmedOpenAlex · checked 6 Sept 2026
Recent years have seen significant technological advancements in precision farming and plant phenotyping. Remote sensing along with deep learning (DL) techniques can increase phenotyping efficiency and help on-farm decision making with rapid stress detection. In this work, we use these techniques to evaluate drought stress in soybean plants, a crop whose yield is significantly affected by water availability. Images were taken from a high vantage in the field at various times throughout the day. Each image is given a wilting score ranging from 0 to 4 by expert scorers. We implement a DL method called multiple instance learning (MIL) to perform wilt classification as well as generate heat maps that highlight wilt levels in specific regions of the image. Given the significant overlap between adjacent classes in our dataset, we were able to achieve an overall classification accuracy of 64% and a one-off accuracy of 96% on our holdout test set. Our model outperformed DenseNet121 in most metrics, and provided comparable performance to a vision transformer (ViT) while having fewer parameters overall, less complexity (useful for edge implementations), and some interpretability. Furthermore, we were able to show that our model outperformed expert human annotators by predicting more consistent and accurate wilt levels when considering single-image re-annotation. The results show that our proposed methodology can be a useful approach in detecting drought stress in soybean fields to facilitate efficient crop management and aid selection of drought-resilient varieties.
Why it matches plant phenotyping methods画像からダイズ葉の萎凋・干ばつストレスを推定するMIL手法の開発と性能比較が中心であり、植物状態の表現型推定に該当する。
abstractWe implement a DL method called multiple instance learning (MIL) to perform wilt classification as well as generate heat maps that highlight wilt levels in specific regions of the image.
Reproduction assets foundThe paper's soybean leaf-wilt image dataset (1788 field images with expert wilt scores) is openly available on Zenodo, and the authors' MIL classification/analysis code is publicly available on GitHub, both explicitly stated in the Data Availability Statement.Dataset · publicThe original data presented in the study are openly available on the
data sharing platform Zenodo, accessed on 6 September 2023 https://zenodo.org/records/8256382
with DOI 10.5281/zenodo.8256382.Open asset ↗Zenodo · 10.5281/zenodo.8256382pdf-page:16 lines:1-58Code · publicWe have also made our code available on github and can be accessed
at https://github.com/ARoS-NCSU/Soybean-Leaf-Wilt-Classification, accessed on 4 March 2025.Open asset ↗GitHubpdf-page:16 lines:1-58Plant phenotyping relevance match · UnverifiedOpenAlex · checked 6 Sept 2026
Achieving global sustainable agriculture requires farmers worldwide to adopt smart agricultural technologies, such as autonomous ground robots. However, most ground robots are either task- or crop-specific and expensive for small-scale farmers and smallholders. Therefore, there is a need for cost-effective robotic platforms that are modular by design and can be easily adapted to varying tasks and crops. This paper describes the hardware design of a unique, low-cost multiaxial modular agricultural robot (ModagRobot), and its field evaluation for soybean phenotyping. The ModagRobot’s chassis was designed without any welded components, making it easy to adjust trackwidth, height, ground clearance, and length. For this experiment, the ModagRobot was equipped with an RGB-Depth (RGB-D) sensor and adapted to safely navigate over soybean rows to collect RGB-D images for estimating soybean phenotypic traits. RGB images were processed using the Excess Green Index to estimate the percent canopy ground coverage area. 3D point clouds generated from RGB-D images were used to estimate canopy height (CH) and the 3D Profile Index of sample plots using linear regression. Aboveground biomass (AGB) was estimated using extracted phenotypic traits. Results showed an R2, RMSE, and RRMSE of 0.786, 0.0181 m, and 2.47%, respectively, between estimated CH and measured CH. AGB estimated using all extracted traits showed an R2, RMSE, and RRMSE of 0.59, 0.0742 kg/m2, and 8.05%, respectively, compared to the measured AGB. The results demonstrate the effectiveness of the ModagRobot for in-row crop phenotyping.
Why it matches plant phenotyping methods低コスト移動ロボット、RGB-D撮像、画像・点群解析によるダイズ形質推定を開発・評価しており、植物フェノタイピング手法が中心である。
abstractThis paper describes the hardware design of a unique, low-cost multiaxial modular agricultural robot (ModagRobot), and its field evaluation for soybean phenotyping.
The segmentation of plant disease images enables researchers to quantify the proportion of disease spots on leaves, known as disease severity. Current deep learning methods predominantly focus on single diseases, simple lesions, or laboratory-controlled environments. In this study, we established and publicly released image datasets of field scenarios for three diseases: soybean bacterial blight (SBB), wheat stripe rust (WSR), and cedar apple rust (CAR). We developed Plant Disease Segmentation Networks (PDSNets) based on LinkNet with ResNet-18 as the encoder, including three versions: ×1.0, ×0.75, and ×0.5. The ×1.0 version incorporates a 4 × 4 embedding layer to enhance prediction speed, while versions ×0.75 and ×0.5 are lightweight variants with reduced channel numbers within the same architecture. Their parameter counts are 11.53 M, 6.50 M, and 2.90 M, respectively. PDSNetx0.5 achieved an overall F1 score of 91.96%, an Intersection over Union (IoU) of 85.85% for segmentation, and a coefficient of determination (R2) of 0.908 for severity estimation. On a local central processing unit (CPU), PDSNetx0.5 demonstrated a prediction speed of 34.18 images (640 × 640 pixels) per second, which is 2.66 times faster than LinkNet. Our work provides an efficient and automated approach for assessing plant disease severity in field scenarios.
Why it matches plant phenotyping methods植物病害画像から病斑割合と病害重症度を推定する画像セグメンテーション手法を開発し、野外データセット、精度、速度を評価しており、植物表現型取得法が中心である。
abstractThe segmentation of plant disease images enables researchers to quantify the proportion of disease spots on leaves, known as disease severity.
Reproduction assets foundThe paper's field-scenario plant disease image dataset (SBB, WSR, CAR with three-color pixel labels) is publicly released on Kaggle via DOI, as stated in the Data Availability Statement. No author analysis code or trained model checkpoints are explicitly deposited.Dataset · publicData Availability Statement: The original data presented in this study are openly available in Kaggle
at https://doi.org/10.34740/kaggle/ds/6620728, accessed on 9 March 2025.Open asset ↗Kaggle · 10.34740/kaggle/ds/6620728pdf-page:15 lines:1-58Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
The vast volumes of data are needed to train Deep Learning Models from scratch to identify illnesses in soybean leaves. However, there is still a lack of sufficient high-quality samples. To overcome this problem, we have developed the real-life SoyLeaf dataset and used the pre-trained Deep Learning Models to identify leaf diseases. In this paper, we have initially developed the real-life SoyLeaf dataset collected from the ICAR-Indian Institute of Soybean Research (IISR) Center, Indore field. This SoyLeaf dataset contains 9786 high-quality soybean leaf images, including healthy and diseased leaves. Following this, we have adapted data preprocessing techniques to enhance the quality of images. In addition, we have utilized several Deep Learning Models, i.e., fourteen Keras Transfer Learning Models, to determine which model best fits the dataset on SoyLeaf diseases. The accuracies of the proposed fine-tuned models using the Adam optimizer are as follows: ResNet50V2 achieves 99.79%, ResNet101V2 achieves 99.89%, ResNet152V2 achieves 99.59%, InceptionV3 achieves 99.83%, InceptionResNetV2 achieves 99.79%, MobileNet achieves 99.82%, MobileNetV2 achieves 99.89%, DenseNet121 achieves 99.87%, and DenseNet169 achieves 99.87%. Similarly, the accuracies of the proposed fine-tuned models using the RMSprop optimizer are as follows: ResNet50V2 achieves 99.49%, ResNet101V2 achieves 99.45%, ResNet152V2 achieves 99.45%, InceptionV3 achieves 99.58%, InceptionResNetV2 achieves 99.88%, MobileNet achieves 99.73%, MobileNetV2 achieves 99.83%, DenseNet121 achieves 99.89%, and DenseNet169 achieves 99.77%. The experimental results of the proposed fine-tuned models show that only ResNet50V2, ResNet101V2, InceptionV3, InceptionResNetV2, MobileNet, MobileNetV2, DenseNet121, and DenseNet169 have performed better in terms of training, validation, and testing accuracies than other state-of-the-art models.
Why it matches plant phenotyping methods大豆葉画像から病害状態を推定するデータセットと前処理・深層学習ワークフローを開発・評価しており、植物フェノタイピング手法が中心である。
abstractwe have developed the real-life SoyLeaf dataset and used the pre-trained Deep Learning Models to identify leaf diseases.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 6 Sept 2026
Vegetative indices (VIs) are widely used in high-throughput phenotyping (HTP) for the assessment of plant growth conditions; however, a range of VIs among diverse soybeans is still an unexplored research area. For this reason, we investigated a range of four major VIs: normalized difference vegetation index (NDVI), photochemical reflectance index (PRI), anthocyanin reflectance index (ARI), and change to carotenoid reflectance index (CRI) in diverse soybean accessions. Furthermore, we ensured the correct positioning of the region of interest (ROI) on the soybean leaf and clarified the effect of choosing different ROI sizes. We also developed a Python algorithm for ROI selection and automatic VIs calculation. According to our results, each VI showed diverse ranges (NDVI: 0.60-0.84, PRI: -0.03 to 0.05, ARI: -0.84 to 0.85, CRI: 2.78-9.78) in two different growth stages. The size of pixels in ROI selection did not show any significant difference. In contrast, the shaded part and the petiole part had significant differences compared with the non-shaded and tip, side, and center of the leaf, respectively. In the case of the Python algorithm, algorithm-derived VIs showed a high correlation with the ENVI software-derived value: NDVI -0.97, PRI -0.96, ARI -0.98, and CRI -0.99. Moreover, the average error was detected to be less than 2.5% in all these VIs than in ENVI.
Why it matches plant phenotyping methods大豆葉のハイパースペクトル画像から植生指数を抽出するROI自動選択・計算法を開発し、既存ソフトウェアとの相関および誤差で検証しており、植物表現型取得法が中心である。
abstractWe also developed a Python algorithm for ROI selection and automatic VIs calculation.
Unmanned aerial vehicle (UAV) platforms are increasingly used to obtain plant phenotypes in crop breeding for their efficiency and versatility. A lightweight UAV was used to collect high-precision RGB images, multispectral and point cloud data of soybeans (Glycine max (L.) Merr.) across fields at various growth stages, utilizing an innovative cross-circling oblique (CCO) route. A multi-modal data fusion deep learning model was proposed based on the self-supervised contrastive learning strategy with fine-tuning for yield estimation and lodging discrimination in soybean germplasm resources. During the soybean growth stages of flowering (R1) to maturity (R8), the contrastive learning effectively captured the decoupling characteristics of different soybean varieties in the feature space. Higher accuracy in yield estimation was obtained combined contrastive learning with the traditional features. Correlations were significantly reduced between features among varieties (Pearson’s mean 0.27–0.62) and feature separations were achieved after dimension reduction (R8: CH = 12.4, DB = 51.8). RMSE of yield estimation was 591.39 kg ha⁻¹ at high density and 532.75 kg ha⁻¹ at low density at R8 growth stages. Lodging discrimination achieved the highest accuracy with an F1-score of 0.57 at high density and 0.64 at low density. The results demonstrated that utilizing contrastive learning for extraction of deep soybean features holds significant potential in supporting traditional features for yield estimation and lodging discrimination.
Why it matches plant phenotyping methodsUAVによるマルチモーダル植物表現型取得と、対照学習を用いた収量推定・倒伏判別モデルが研究の中心であり、手法性能も定量評価している。
abstractA multi-modal data fusion deep learning model was proposed based on the self-supervised contrastive learning strategy with fine-tuning for yield estimation and lodging discrimination in soybean germplasm resources.
Stomatal conductance (g s ) quantifies the rate of exchange of carbon dioxide for photosynthesis and water vapor for transpiration between plant leaves and the atmosphere. g s is usually measured by handheld devices like porometers , and readings are manually taken in the field, which is time-consuming and labor-intensive. In this study, we investigated the use of high-throughput phenotyping (HTP) data combined with weather data to estimate g s through machine-learning (ML) modeling. The experiment was conducted in a research field equipped with an HTP platform in 2020 and 2021 involving maize, sorghum, soybean, sunflower , and winter wheat . Weather variables including dew point temperature, wind speed , air temperature, solar radiation, and relative humidity were collected by an onsite weather station . Plot-level canopy temperature, soil temperature , and seven vegetation indices were acquired using a thermal infrared camera, a multispectral camera, and a visible near-infrared spectrometer integrated on the HTP platform. Three supervised ML methods (Partial Least Squares Regression (PLSR), Random Forest Regression (RFR), and Support Vector Regression (SVR)) were employed to train the estimation models for g s , and model performance was evaluated by Coefficient of Determination (R 2 ) and Root Mean Squared Error (RMSE). The result showed that RFR and SVR outperformed PLSR in g s modeling. The RFR model achieved R 2 of 0.63 and RMSE of 0.16 mol m −2 ·s −1 with the combination of phenotyping data and weather data. It outperformed the model using only the weather data (R 2 =0.35 and RMSE=0.21 mol m −2 ·s −1 ), or the model using only the phenotyping data (R 2 =0.46 and RMSE=0.19 mol m −2 ·s −1 ). This result suggested that high-throughput plant phenotyping data effectively complement weather data in estimating g s rapidly and non-destructively through ML. With the wide adoption of HTP technologies in aerial and ground-based platforms, this research provides a practical framework to estimate g s at large scale for crop breeding and irrigation management .
Why it matches plant phenotyping methodsHTPセンサーデータと機械学習を用いて、植物の生理形質である気孔コンダクタンスを大規模・非破壊推定する方法が研究の中心であり、モデル性能も評価している。
abstractIn this study, we investigated the use of high-throughput phenotyping (HTP) data combined with weather data to estimate g s through machine-learning (ML) modeling.
CowpeaSoybeanRootMorphology / geometry measurementSegmentationRoot system architecture
A simple Python algorithm was used to estimate the four major root traits: total root length (TRL), surface area (SA), average diameter (AD), and root volume (RV) of legumes (adzuki bean, mung bean, cowpea, and soybean) based on two-dimensional images. Four different thresholding methods; Otsu, Gaussian adaptive, mean adaptive and triangle threshold were used to know the effect of thresholding in root trait estimation and to optimize the accuracy of root trait estimation. The results generated by the algorithm applied to 400 legume root images were compared with those generated by two separate software (WinRHIZO and RhizoVision), and the algorithm was validated using ground truth data. Distance transform method was used for estimating SA, AD, and RV and ConnectedComponentsWithStat function for TRL estimation. Among the thresholding methods, Otsu thresholding worked well for distance transform, while triangle threshold was effective for TRL. All the traits showed a high correlation with an R² ≥0.98 (p < 0.001) with the ground truth data. The root mean square error (RMSE) and mean bias error (MBE) were also minimal when comparing the algorithm-derived values to the ground truth values, with RMSE and MBE both < 10 for TRL, < 6 for SA, and < 0.5 for AD and RV. This lower value of error metrics indicates smaller differences between the algorithm-derived values and software-derived values. Although the observed error metrics were minimal for both software, the algorithm-derived root traits were closely aligned with those derived from WinRHIZO. We provided a simple Python algorithm for easy estimation of legume root traits where the images can be analyzed without any incurring expenses, and being open source; it can be modified by an expert based on their requirements.
Why it matches plant phenotyping methods根の二次元画像から主要形質を抽出するPythonアルゴリズムを開発し、既存ソフトウェアおよびグラウンドトゥルースで検証しており、植物フェノタイピング手法が研究の中心です。
abstractA simple Python algorithm was used to estimate the four major root traits: total root length (TRL), surface area (SA), average diameter (AD), and root volume (RV) of legumes
Reproduction assets foundThe authors publicly release their Python root-trait analysis source code together with the 400 legume root images and validation images on GitHub, as stated in the article text and Data availability statement. The Zenodo DOI cited for ground-truth images is a third-party dataset from Rose and Lobet (2018), i.e., citedCode · publicThe source code along with the root images and the validation images can be downloaded from ( https://github.com/AG9843/Legume-Root-Analysis.git ).Open asset ↗AG9843/Legume-Root-Analysislines:65-75Plant phenotyping relevance match · UnverifiedEurope PMC · checked 15 Sept 2026
Published1 Mar 2025Computers and Electronics in Agriculture.
Yield and lodging are crucial indicators in soybean breeding. The development of unmanned aerial vehicle (UAV) equipped with hyperspectral imaging technologies provides high-throughput data for estimating these factors. Previous studies have primarily focused on using hand-crafted band reflectance information, vegetation indices, and texture features to construct empirical models for yield and lodging estimation. However, few studies have directly employed deep learning techniques to automatically extract features from raw hyperspectral images in this context. The objectives were to investigate the potential of combining hyperspectral images with deep learning for soybean yield and lodging prediction, and to avoid the complex process of traditional feature extraction. A novel Prototype Contrastive Learning (PCL) network was proposed to learn representations from raw images. For comparison, hand-crafted vegetation indices and texture features, selected for their effectiveness in crop growth monitoring, were extracted and input into the same machine learning model. The impact of different growth stages on yield and lodging prediction was then investigated. Results demonstrated that the PCL network can effectively capture the similarities within the same class and the differences between different classes. The PCL representations exhibited more distinct clusters according to class labels compared to hand-crafted features. At 86 days after emergence (DAE), the PCL method achieved optimal yield prediction accuracy (R² = 0.65, RMSE = 507.56 kg/ha) and was significantly higher than the hand-crafted features method (R² = 0.55, RMSE = 581.37 kg/ha). The highest performance of lodging grades classification was achieved at 65 DAE, and the PCL representations (F1-score = 0.80) achieved a 48 % accuracy improvement compared to hand-crafted features (F1-score = 0.54). This study pioneered the use of deep learning for automatic hyperspectral feature extraction in real-world breeding scenarios, providing valuable insights and strategies to improve yield prediction and lodging classification, thereby more effectively supporting soybean breeding and field management.
Why it matches plant phenotyping methodsUAVハイパースペクトル画像からダイズの収量と倒伏を推定する深層学習手法を提案・比較評価しており、植物形質の取得・抽出が研究の中心である。
abstractA novel Prototype Contrastive Learning (PCL) network was proposed to learn representations from raw images.
Developments in genomics and phenomics have provided valuable tools for use in cultivar development. Genomic prediction (GP) has been used in commercial soybean [Glycine max L. (Merr.)] breeding programs to predict grain yield and seed composition traits. Phenomic prediction (PP) is a rapidly developing field that holds the potential to be used for the selection of genotypes early in the growing season. The objectives of this study were to compare the performance of GP and PP for predicting soybean seed yield, protein, and oil. We additionally conducted genome-wide association studies (GWAS) to identify significant single-nucleotide polymorphisms (SNPs) associated with the traits of interest. The GWAS panel of 292 diverse accessions was grown in six environments in replicated trials. Spectral data were collected at two time points during the growing season. A genomic best linear unbiased prediction (GBLUP) model was trained on 269 accessions, while three separate machine learning (ML) models were trained on vegetation indices (VIs) and canopy traits. We observed that PP had a higher correlation coefficient than GP for seed yield, while GP had higher correlation coefficients for seed protein and oil contents. VIs with high feature importance were used as covariates in a new GBLUP model, and a new random forest model was trained with the inclusion of selected SNPs. These models did not outperform the original GP and PP models. These results show the capability of using ML for in-season predictions for specific traits in soybean breeding and provide insights on PP and GP inclusions in breeding programs.
Why it matches plant phenotyping methods大豆の収量・種子成分を対象に、圃場スペクトルデータ、植生指数、樹冠形質、機械学習を用いたフェノミック予測を構築・比較しており、表現型取得・推定手法が研究の中心である。
abstractThe objectives of this study were to compare the performance of GP and PP for predicting soybean seed yield, protein, and oil.
Published1 Mar 2025Journal of food composition and analysis : an official publication of the United Nations University, International Network of Food Data Systems
Soybean is an important oil crop with significant economic value worldwide, the breeding of soybean varieties requires not only high oil content, but also the appropriate ratio of fatty acids. In this study, a rapid and nondestructive detection method for oil content and fatty acids of soybean was developed using hyperspectral imaging (HSI) technology. Five wavelength selection methods, including competitive adaptive re-weighted sampling, random frogs, iteratively retaining informative variables, uninformative variable elimination, and genetic algorithm, were used to select the important variables, then partial least squares was used to build the prediction models. Among five methods, uninformative variable elimination provided with satisfactory results for the prediction of oil content and fatty acid contents of soybean. The validation results showed that oil content and linolenic acid had good performance with correlation coefficient for cross-validation (R²cᵥ) values of 0.90 and 0.92, and correlation coefficient predictive (R²ₚ) values of 0.93 and 0.93, respectively. The relative errors between the predicted and actual values of oil and linolenic acid content ranged from 0.05% to 5.68 % and from 0.11% to 11.87 %, respectively. In addition, oleic acid had better results with R²cᵥ, residual predictive deviation for cross validation (RPDcᵥ), and R²ₚ values of 0.84, 2.45, and 0.85, respectively. Furthermore, compared the models developed using near infrared (NIR), the average relative errors of the established HSI models for oil content, oleic acid, linoleic acid and linolenic acid in soybean decreased by 48.94 %, 21.85 %, 37.98 % and 39.31 %, respectively. Therefore, HSI technology has great potential to detect oil content and major fatty acids in soybeans.
Why it matches plant phenotyping methods大豆種子の油含量・脂肪酸という植物器官形質を、ハイパースペクトル画像から非破壊推定する手法を開発し、波長選択法と予測モデルを比較・検証しているため、測定法が中心である。
abstracta rapid and nondestructive detection method for oil content and fatty acids of soybean was developed using hyperspectral imaging (HSI) technology.
Abstract has not been obtained from indexed metadata or an accessible article page.
Why it matches plant phenotyping methodsハイパースペクトル画像とコントラスト学習を用いてダイズの収量および倒伏を推定・分類する方法が題名の中心であり、植物表現型の取得・抽出に該当する。
titleBridging the gap between hyperspectral imaging and crop breeding: soybean yield prediction and lodging classification with prototype contrastive learning
The rapid advancement of artificial intelligence has paved the way for innovative solutions in agriculture, particularly in crop disease detection. Diagnosing plant diseases often rely on manual inspection and expert knowledge that time-consuming and prone to errors. As agriculture faces increasing challenges from pests, diseases and climate change, there is a pressing require for efficient, automated systems to monitor crop health. In this manuscript, Development of an AI-Based Pyramid Convolutional Neural Network Model with ResNet and Coati Optimization for Multi-Crop, Multi-Disease Identification in Leaf Images is (PCNN-ResNet-COA) proposed. Initially the data is collected from Plant Disease Classification Merged Dataset. This dataset includes both healthy and diseased leaves for various crop types. The collected images are organized into categories based on the crop type such as crop, grape and soybean and disease condition healthy or various types of diseases. Then the categorized images are fed to Pyramid Convolutional Neural Network with Residual Network (PCNN-ResNet), for identifying and classifying the Leaf Images as Corn_healthy, Corn_northern_leaf_blight, Corn_gray_leaf_spot, Grape_healthy, Grape_leaf_blight, Grape_black_rot, Grape_black_measles, corn_common_rust, soybean_bacterial_blight, Soybean_downy_mildew, Soybean_mosaic_virus, Soybean_powdery_mildew, Soybean_healty, Soybean_rust, and Soybean_southern_blight. In general, PCNN-ResNet does not express any adaption of optimization methods for determining optimal parameters to assure precise detection and classification of Leaf Images. Coati Optimization Algorithm (COA) is proposed for improving the weight parameter of PCNN-ResNet classifier that accurately predicts crop yield. The proposed PCNN-ResNet-COA method is implemented and analyzed with help of performance metrics like accuracy, precision, F1-score, computational time is evaluated. The proposed ResNet-COA approach attains 18.97%, 24.57% and 32.68% higher accuracy and 19.84%, 24.93% and 31.62% lower computational time with existing method respectively.
Why it matches plant phenotyping methods葉画像から植物病害状態を自動分類する新規CNN手法を開発・評価しており、植物フェノタイピング手法が中心です。
titleDevelopment of an AI-Based Pyramid Convolutional Neural Network Model with ResNet and Coati Optimization for Multi-Crop, Multi-Disease Identification in Leaf Images
Reproduction assets foundThe paper's leaf-image disease classification experiments are built entirely on the public Kaggle 'Plant Disease Classification Merged Dataset' (88 classes, >76,000 images), which is explicitly cited with its Kaggle URL in the references. No author code, models, or checkpoints are reported as available.Dataset · publicThe input data are obtained from Plant Disease Classification Merged Dataset [17]. A huge number of images, at
least one healthy plant and one disease per plant, the most prevalent diseases, annotated images, laboratory and
field photographs, significant staple foods and the plant species with the highest worldwide production were the
self-imposed conditions for the dataset.Open asset ↗Plant Disease Classification Merged Datasetpdf-layout-page:4 lines:1-64Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Accurate detection of soybean diseases is a critical component in achieving intelligent agricultural management. However, traditional methods often underperform in complex field scenarios. This paper proposes a diffusion-based object detection model that integrates the endogenous diffusion sub-network and the endogenous diffusion loss function to progressively optimize feature distributions, significantly enhancing detection performance for complex backgrounds and diverse disease regions. Experimental results demonstrate that the proposed method outperforms multiple baseline models, achieving a precision of 94%, recall of 90%, accuracy of 92%, and mAP@50 and mAP@75 of 92% and 91%, respectively, surpassing RetinaNet, DETR, YOLOv10, and DETR v2. In fine-grained disease detection, the model performs best on rust detection, with a precision of 96% and a recall of 93%. For more complex diseases such as bacterial blight and Fusarium head blight, precision and mAP exceed 90%. Compared to self-attention and CBAM, the proposed endogenous diffusion attention mechanism further improves feature extraction accuracy and robustness. This method demonstrates significant advantages in both theoretical innovation and practical application, providing critical technological support for intelligent soybean disease detection.
Why it matches plant phenotyping methods大豆病害を対象に、画像ベースの病徴・病害状態推定モデルを開発し、複数モデルとの性能比較で検証しているため、植物フェノタイピング手法が中心である。
abstractThis paper proposes a diffusion-based object detection model that integrates the endogenous diffusion sub-network and the endogenous diffusion loss function to progressively optimize feature distributions, significantly enhancing detection performance for complex backgrounds and diverse disease regions.
Soybean stands out for being the most economically important oilseed in the world. Remote sensing techniques and precision agriculture are being analyzed through research in different agricultural regions as a technological system aiming at productivity and possible low-cost reduction. Machine learning (ML) methods, together with the advent of demand for remotely piloted aircraft available on the market in the recent decade, have been conducive to remote sensing data processes. The objective of this work was to evaluate the best ML and input configurations in the classification of agronomic variables in different phenological stages. The spectral variables were obtained in three phenological stages of soybean genotypes: V8 (at 45 days after emergence—DAE), R1 (60 DAE), and R5 (80 DAE). A Sensefly eBee fixed-wing RPA equipped with the Parrot Sequoia multispectral sensor coupled to the RGB sensor was used. The Sequoia multispectral sensor with an RGB sensor acquired reflectance at wavelengths of blue (450 nm), green (550 nm), red (660 nm), near-infrared (735 nm), and infrared (790 nm). The following were used to evaluate the agronomic traits: days to maturity, number of branches, productivity, plant height, height of the first pod insertion and diameter of the main stem. The random forest (RF) model showed greater accuracy with data collected in the R5 stage, whose accuracies were close to 56 for the percentage of correct classifications (CC), close to 0.2 for Kappa, and above 0.55 for the F-score. Logistic regression (RL) and support vector machine (SVM) models showed better performance in the early reproductive stage R1, with accuracies above 55 for CC, close to 0.1 for Kappa, and close to 0.4 for the F-score. J48 performed better with data from the V8 stage, with accuracies above 50 for CC and close to 0.4 for the F-score. This reinforces that the use of different specific spectra for each model can enhance accuracy, optimizing the choice of model according to the phenological stage of the plants.
Why it matches plant phenotyping methodsマルチスペクトルセンサー搭載RPAによる植物形質取得と、各生育段階での機械学習モデル比較が研究の中心であり、形質推定ワークフローを実質的に評価している。
titleHigh-Precision Phenotyping in Soybeans: Applying Multispectral Variables Acquired at Different Phenological Stages
The breeding of high-yield varieties is a core objective of soybean breeding programs, and phenotypic trait-based selection offers an effective pathway to achieve this goal. The aim of this study was to identify the key phenotypic traits of high-yield soybean varieties and to utilize these traits for screening high-yield soybean varieties. In this study, the UAV (unmanned aerial vehicle)- and field-based phenotypic data were collected from 1923 and 1015 soybean breeding plots at the Xuzhou experimental site in 2022 and 2023, respectively. First, the soybean varieties were grouped by using a self-organizing map and K-means clustering to investigate the relationships between various traits and soybean yield and to identify the key ones for selecting high-yield soybean varieties. It was shown that the duration of canopy coverage remaining above 90% (Tcc90) was a critical phenotypic trait for selecting high-yield varieties. Moreover, high-yield soybean varieties typically exhibited several key phenotypic traits such as rapid development of canopy coverage (Tcc90r, the time when canopy coverage first reached 90%), prolonged duration of high canopy coverage (Tcc90), a delayed decline in canopy coverage (Tcc90d, the time when canopy coverage began to decline below 90%), and moderate-to-high plant height (PH) and hundred-grain weight (HGW). Based on these findings, a method for screening high-yield soybean varieties was proposed, through which 87% and 72% of high-yield varieties (top 5%) in 2022 and 2023, respectively, were successfully selected. Additionally, about 9% (in 2022) and 10% (in 2023) of the low-yielding (bottom 60%) were misclassified as high-yielding. This study demonstrates the benefit of high-throughput phenotyping for soybean yield-related traits and variety screening and provides helpful insights into identifying high-yield soybean varieties in breeding programs.
Why it matches plant phenotyping methodsUAV・圃場フェノタイピングによる形質抽出と、高収量品種を選抜するスクリーニング手法の提案・評価が研究の中心であるため。
abstractThe aim of this study was to identify the key phenotypic traits of high-yield soybean varieties and to utilize these traits for screening high-yield soybean varieties.
Several home pesticides are organophosphorus compounds. These compounds inhibit the enzyme acetylcholinesterase, causing harmful effects on the health of biota. Through this research, the usefulness of Glycine max (soybean) and Cichorium intybus (chicory) plants as sentinels of organophosphorus compounds in the environment was successfully tested. Different concentrations of the insecticide chlorpyrifos were tried out. Damage to plants at the photosynthetic apparatus level was evaluated by measuring the high temporal resolution variable chlorophyll fluorescence (OJIP test). Several parameters derived from this test indicated a high level of damage in both species even at the mean dose recommended for use in the field. However, a few parameters did not consistently reflect damage in leaves. A drop in the values of the maximum fluorescence (F M ), the quantum yield of electron transport flux, transport between quinones A and B (ET 0 /ABS) and the maximal quantum yield of PSII (TR 0 /ABS) could alert us about the presence of organophosphates in the environment. An increase in the dissipated energy flux per reaction center (DI 0 /RC) values was also observed. The species showed different sensitivities, with soybean plants being the most sensitive. The OJIP transient thus becomes a valuable rapid, non-destructive tool for biomonitoring this class of pesticides in the environment.
Why it matches plant phenotyping methods植物の光合成状態を測定する高時間分解クロロフィル蛍光法を、農薬による植物損傷の評価・環境バイオモニタリング手法として検証しており、表現型取得が中心です。
abstractDamage to plants at the photosynthetic apparatus level was evaluated by measuring the high temporal resolution variable chlorophyll fluorescence (OJIP test).
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:139-197Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
The portable X-ray fluorescence (pXRF) spectrometry has been very useful for the characterization of different earth materials, and its application for foliar analysis is really promising. The performance of pXRF for foliar analysis depends on several factors such as concentration of the elements, fluorescence yield which is influenced by atomic number, spectral interference, and water content. Mn is one of the elements that present a prominent fluorescence peak. In this sense, it was hypothesized that pXRF can directly determine the Mn concentration on foliar samples, even when used on intact leaves (fresh or dry) being a useful tool for agronomic and environmental purposes. Thus, the objective was to assess the performance of a pXRF to determine Mn concentration in two different foliar datasets from Brazil/South America and Mali/Africa. In the Brazilian dataset, leaves from eight crops (common bean, castor plant, coffee, eucalyptus, guava tree, maize, mango, and soybean) were scanned via pXRF at the following conditions: intact and fresh leaves, intact and dry leaves, and powdered samples). In the Malian dataset, powdered samples from cotton and maize were analyzed via pXRF. For comparison, Mn concentration was also determined after nitro-perchloric digestion followed by quantification via inductively coupled plasma optical emission spectroscopy (ICP-OES). After descriptive statistics, linear regressions were performed for all sample preparation conditions in both datasets, using Mn concentrations obtained through pXRF and the acid digestion method. The data quality level of all linear regressions was considered quantitative with high R (0.93 to 0.98) and R 2 (0.87 to 0.96) values. The direct analysis of Mn via pXRF on intact and fresh leaves yielded R of 0.93, R 2 of 0.87, and a low relative standard deviation (< 10%). The manufactured pXRF calibration used in this work allowed an accurate direct Mn determination in plant leaves. Considering the importance of Mn as a plant micronutrient and its potential toxicity depending on soil redox conditions, the fast, in situ, non-destructive, and eco-friendly determination via pXRF has a tremendous agronomic and environmental application worldwide.
Why it matches plant phenotyping methods植物葉のMn濃度という生理・元素形質を、携帯型XRFで非破壊測定する方法の性能評価と検証が中心であり、単なる生物学的実験での routine 測定ではない。
abstractThe direct analysis of Mn via pXRF on intact and fresh leaves yielded R of 0.93, R 2 of 0.87, and a low relative standard deviation (< 10%).
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · OpenAlex · checked 15 Sept 2026
• UAV-derived canopy model quantified radiation availability of intercropped soybean. • Shadow fraction method was developed to calculate direct, diffuse radiation and RUE. • RUE of intercropped soybean was higher than that in monoculture. • Fraction of diffuse in intercropping was slightly lower than that in monoculture. • Other factors leading to the higher RUE of soybean in intercropping systems. Shading is an unavoidable phenomenon in strip intercropping systems for lower crops, which affects the amount and component of solar radiation, and thus the radiation use efficiency (RUE). The higher crop is usually treated as a homogeneous block instead of the actual canopy structure to calculate lower crop radiation availability (block-based method, BM), which underestimates the amount of light passing through gaps in the canopy. Here we proposed a new shadow fraction method (SFM) to separately quantify direct and diffuse radiation on lower crops. The SFM considered shadow fraction dynamic and view factor within a day, which was calculated based on UAV-derived canopy structural models. To test this method, UAV images and crop data were collected from a maize-soybean intercropping experiment with six planting configurations. For daily total radiation, as the width of the soybean strip decreased from 3.8 m to 1.6 m, the relative difference between BM and SFM increased from about 11.10% to 20.36%. Accordingly, the RUE of soybean calculated by the SFM was 0.2–0.3 g/MJ lower than the BM. Consistent with previous studies, the RUE of soybean in strip intercropping systems (1.36–1.61 g/MJ) calculated by the SFM was higher than that in monoculture (0.98 g/MJ). The higher RUE was usually attributed to the increasing fraction of diffuse in strip intercropping systems. However, SFM showed that the fraction of diffuse on intercropped soybean (ranged from 37.42% to 38.58%) was slightly lower than that in monoculture (39.48%), implying that other factors, such as light intensity and quality, may have an impact on soybean performance and warrant further investigation. The SFM was theoretically more accurate than BM as it considered the actual 3D canopy structure. This method can enhance the understanding of light distribution and use efficiency in intercropping systems, which can be integrated with crop growth models or functional structural plant models to optimize intercropping configurations for improved resource use efficiency.
Why it matches plant phenotyping methodsUAV由来の3Dキャノピー構造モデルを用いて、下層作物の光環境を推定する新しいshadow fraction法を開発・検証しており、植物キャノピー構造と放射利用効率の定量化が研究の中心である。
abstractHere we proposed a new shadow fraction method (SFM) to separately quantify direct and diffuse radiation on lower crops.
SoybeanLaboratory / benchtopSeed / grainClassificationGrowth / development / phenology
High-vigor soybean seeds are critical for efficient production owing to their favorable growth properties and high yield potential. The evaluation and identification of high-vigor germplasms are essential for increasing soybean production capacity. Currently, there is no universally accepted evaluation system to test for soybean seed vigor. In this study, 11 seed vigor-related traits were measured across 126 soybean landraces via an artificial accelerated aging technique. The ratios of these 11 traits, which were calculated before and after artificial accelerated aging, were used as vigor indicators in principal component analysis (PCA), ultimately yielding two principal component factors. These factors were then combined via membership function standardization to calculate a comprehensive seed vigor evaluation value (V value), thereby establishing an evaluation system. Cluster analysis based on the V value was used to classify seed vigor into five levels and identify seven high-vigor germplasms: ZDD12322, ZDD06438, ZDD11951, ZDD08251, ZDD12436, ZDD02315, and ZDD15624. Through stepwise regression analysis, the optimal seed vigor predictive model was defined as V = −0.026 + 0.625 × RSL + 0.485 × RGI. This model revealed that the relative seedling length (RSL) and relative germination index (RGI) had significant positive effects on seed vigor. This study provides a valuable framework for seed quality control and selection, facilitating presowing vigor assessments to increase soybean planting efficiency and yield.
Why it matches plant phenotyping methods種子活力を評価するための統合評価システムと予測モデルの開発が研究の中心であり、発芽・幼苗形質から植物状態を抽出・推定する方法論的貢献がある。
titleDevelopment of a comprehensive evaluation system and models to determine soybean seed vigor
Agriculture contributes 18% to India's GDP, with soybean production at 14 million metric tons annually, making it a major crop for farmers, though the sector's share is decreasing. Bacterial, fungal, and viral diseases, along with nematode infestations, can affect soybeans throughout the growing season. Accurate disease identification and appropriate treatment improve soybean production by stopping the spread of infections, reducing crop loss, enhancing plant health, boosting yields, and providing better economic benefits for farmers. The fifty different soybean farms in Maharashtra were surveyed to construct a dataset on the major diseases affecting the soybean crop. The collected images are preprocessed through resizing for uniformity, data augmentation for diversity, and normalization to scale pixel values, facilitating efficient training. The collected dataset is used to train several deep learning algorithms, such as AlexNet, VGG-16, Inception-v3, EfficientNetV2B0, and ResNet-50, to predict diseases. To evaluate the model's effectiveness, the study analyzed the training and validation loss and accuracy. The real-time soybean plant images were utilized, and YOLO was employed for leaf extraction, generating test images that were then fed into the trained models. The outcomes show that ResNet-50 predicts soybean conditions from the captured pictures from the soybean farm more effectively than cutting-edge methods. Performance metrics for classification were calculated for each model, with ResNet-50 yielding the most accurate predictions across all metrics. A confusion matrix was also generated to assess the model's classification accuracy, further confirming ResNet-50’s robustness. These results suggest that deep learning models, especially ResNet-50, can serve as effective tools for early and accurate detection of soybean diseases, offering valuable support for precision agriculture.
Why it matches plant phenotyping methods大豆葉画像から病害状態を抽出・分類するYOLOおよび深層学習ワークフローが研究の中心であり、植物病害表現型の画像ベース推定を評価している。
abstractThe collected images are preprocessed through resizing for uniformity, data augmentation for diversity, and normalization to scale pixel values, facilitating efficient training.
Plant phenotyping relevance match · UnverifiedCrossref · Europe PMC · checked 6 Sept 2026
Phenotypic traits like plant height are crucial in assessing plant growth and physiological performance. Manual plant height measurement is labor and time-intensive, low throughput, and error-prone. Hence, aerial phenotyping using aerial imagery-based sensors combined with image processing technique is quickly emerging as a more effective alternative to estimate plant height and other morphophysiological parameters. Studies have demonstrated the effectiveness of both RGB and LiDAR images in estimating plant height in several crops. However, there is limited information on their comparison, especially in soybean ( Glycine max [L.] Merr.). As a result, there is not enough information to decide on the appropriate sensor for plant height estimation in soybean. Hence, the study was conducted to identify the most effective sensor for high throughput aerial phenotyping to estimate plant height in soybean. Aerial images were collected in a field experiment at multiple time points during soybean growing season using an Unmanned Aerial Vehicle (UAV or drone) equipped with RGB and LiDAR sensors. Our method established the relationship between manually measured plant height and the height obtained from aerial platforms. We found that the LiDAR sensor had a better performance (R 2 = 0.83) than the RGB camera (R 2 = 0.53) when compared with ground reference height during pod growth and seed filling stages. However, RGB showed more reliability in estimating plant height at physiological maturity when the LiDAR could not capture an accurate plant height measurement. The results from this study contribute to identifying ideal aerial phenotyping sensors to estimate plant height in soybean during different growth stages.
Why it matches plant phenotyping methodsUAVのRGB・LiDAR画像を用いたダイズ草高推定法を開発・比較検証しており、植物形質の取得方法が研究の中心である。
abstractHence, the study was conducted to identify the most effective sensor for high throughput aerial phenotyping to estimate plant height in soybean.
X-ray fluorescence (XRF) is a well-established technique for elemental determination. This study evaluates the ability of XRF to quantify soybean protein content based on elemental composition, particularly sulfur emission. Univariate linear regression, multiple linear regression, and partial least squares regression (PLS) were compared. Two scenarios were considered: scenario A used 108 soybean samples for calibration and 54 for validation; scenario B expanded the protein content range of scenario A, including 32 new samples of soybean mixed with concentrates. PLS showed the best performance in validation, with R 2 of 0.73 and 0.89 in scenarios A and B, respectively. The results indicate that protein quantification by XRF has relative prediction errors below 3.1 %. The developed methods provide an alternative for monitoring soybean protein content, suitable for screening applications such as integrating XRF sensors on soybean harvesters.
Why it matches plant phenotyping methods大豆子実のタンパク質含量という植物形質をXRFで推定する手法を開発・検証しており、検量・独立検証・予測性能評価が研究の中心である。
abstractThis study evaluates the ability of XRF to quantify soybean protein content based on elemental composition, particularly sulfur emission.
Abstract. Breeding climate-robust crops is one of the needed pathways for adaptation to the changing climate. To speed up the breeding process, it is important to understand how plants react to extreme weather events such as drought or waterlogging in their production environment, i.e. under field conditions in real soils. Whereas a number of techniques exist for aboveground field phenotyping, simultaneous non-invasive belowground phenotyping remains difficult. In this paper, we present the first data set of the new HYDRAS (HYdrology, Drones and RAinout Shelters) open-access field-phenotyping infrastructure, bringing electrical resistivity tomography, alongside drone imagery and environmental monitoring, to a technological readiness level closer to what breeders and researchers need. This paper investigates whether electrical resistivity tomography (ERT) provides sufficient precision and accuracy to distinguish between belowground plant traits of different genotypes of the same crop species. The proof-of-concept experiment was conducted in 2023, with three distinct soybean genotypes known for their contrasting reactions to drought stress. We illustrate how this new infrastructure addresses the issues of depth resolution, automated data processing, and phenotyping indicator extraction. The work shows that electrical resistivity tomography is ready to complement drone-based field-phenotyping techniques to accomplish whole-plant high-throughput field phenotyping.
Why it matches plant phenotyping methodsERTとドローン画像を用いた非侵襲的な地下部形質の取得・抽出精度を検証する新規圃場フェノタイピング基盤の研究であり、方法と技術性能が中心です。
abstractwe present the first data set of the new HYDRAS (HYdrology, Drones and RAinout Shelters) open-access field-phenotyping infrastructure
Introduction Advancements in machine learning (ML) algorithms that make predictions from data without being explicitly programmed and the increased computational speeds of graphics processing units (GPUs) over the last decade have led to remarkable progress in the capabilities of ML. In many fields, including agriculture, this progress has outpaced the availability of sufficiently diverse and high-quality datasets, which now serve as a limiting factor. While many agricultural use cases appear feasible with current compute resources and ML algorithms, the lack of reusable hardware and software components, referred to as cyberinfrastructure (CI), for collecting, transmitting, cleaning, labeling, and training datasets is a major hindrance toward developing solutions to address agricultural use cases. This study focuses on addressing these challenges by exploring the collection, processing, and training of ML models using a multimodal dataset and providing a vision for agriculture-focused CI to accelerate innovation in the field. Methods Data were collected during the 2023 growing season from three agricultural research locations across Ohio. The dataset includes 1 terabyte (TB) of multimodal data, comprising Unmanned Aerial System (UAS) imagery (RGB and multispectral), as well as soil and weather sensor data. The two primary crops studied were corn and soybean, which are the state's most widely cultivated crops. The data collected and processed from this study were used to train ML models to make predictions of crop growth stage, soil moisture, and final yield. Results The exercise of processing this dataset resulted in four CI components that can be used to provide higher accuracy predictions in the agricultural domain. These components included (1) a UAS imagery pipeline that reduced processing time and improved image quality over standard methods, (2) a tabular data pipeline that aggregated data from multiple sources and temporal resolutions and aligned it with a common temporal resolution, (3) an approach to adapting the model architecture for a vision transformer (ViT) that incorporates agricultural domain expertise, and (4) a data visualization prototype that was used to identify outliers and improve trust in the data. Discussion Further work will be aimed at maturing the CI components and implementing them on high performance computing (HPC). There are open questions as to how CI components like these can best be leveraged to serve the needs of the agricultural community to accelerate the development of ML applications in agriculture.
Why it matches plant phenotyping methods農業向けサイバーインフラの開発が中心で、UAS画像処理パイプラインとMLモデルにより作物の生育ステージおよび収量を推定しており、植物形質の取得・抽出方法が実質的に扱われている。
abstractThis study focuses on addressing these challenges by exploring the collection, processing, and training of ML models using a multimodal dataset and providing a vision for agriculture-focused CI to accelerate innovation in the field.
The leaf area index (LAI) is a critical parameter for characterizing plant foliage abundance, canopy structure changes, and vegetation productivity in ecosystems. Traditional phenological measurements are often destructive, time-consuming, and labor-intensive. This paper proposes a high-throughput 3D point cloud data processing pipeline to segment field soybean plants and estimate their LAI. The 3D point cloud data is obtained from a UAV equipped with a LiDAR camera. First, The PointNet++ model was applied to simplify the segmentation process by isolating field soybean plants from their surroundings and eliminating environmental complexities. Subsequently, individual segmentation was achieved using the Watershed approach and k-means clustering algorithms, segmenting the field soybeans into individual plants. Finally, the LAI of soybean plant was estimated using a machine learning method and validated against measured values. The PointNet++ model improved segmentation accuracy by 6.73%, and the watershed algorithm achieved F1 scores of 0.89-0.90, outperforming k-means in complex adhesion cases. For LAI estimation, the SVM model showed the highest accuracy (R² = 0.79, RMSE = 0.47), with RF and XGBoost also performing well (R² > 0.69, RMSE< 0.65). This indicates that the individual segmentation algorithm, Watershed-based approach combined with PointNet++, can serve as a crucial foundation for extracting high-throughput plant phenotypic data. The experimental results confirm that the proposed method can rapidly calculate the morphological parameters of each soybean plant, making it suitable for high-throughput soybean phenotyping.
Why it matches plant phenotyping methodsUAV-LiDAR点群、深層学習・クラスタリングを用いて個体分割とLAI推定を開発・検証する、植物表現型取得手法が中心の研究。
abstractThis paper proposes a high-throughput 3D point cloud data processing pipeline to segment field soybean plants and estimate their LAI.
The recent climate dynamics characterized by unpredictability and a series of extreme events pose challenges to society at various levels, particularly threatening agricultural production. The development of increasingly sophisticated models and computers combined with remote sensing techniques can serve as a means to safeguard the agricultural domain.The aim of this work is to develop a computational tool, named CROPORBIT, designed to operate at a regional scale for estimating crop yield. The capabilities of this tool have a significant positive impact on water management, crop health monitoring, and quantifying damage from extreme meteorological events, such as high temperatures.CROPORBIT combined the radiative model METRIC with a Photosynthetically Active Radiation-based model. Essential inputs for the tool include Landsat 8 and 9 satellite imagery and daily meteorological data retrieved from the regional network stations.The tool performs a multi-temporal analysis of crop growth, involving the interpolation of ET, stress coefficient, and dry biomass accumulation maps, which are then transformed into crop yield maps by applying a harvest index coefficient.CROPORBIT underwent validation in a series of soybean and corn fields situated in the low-lying plain of the Veneto Region, where crop yield maps were recorded by combine harvesters.The preliminary results have shown that CROPORBIT can predict the average crop yield with a good approximation while it was less performing in capturing the field yield variability. The main issues have proven to be the scarcity of clear-sky conditions imagery and the estimation of the harvest index variability.This research establishes the foundation for future investigations, emphasizing the need for improvements in spatial and time resolution. Enhancements in these aspects may lead to improved outcomes in terms of both accuracy and spatial variability.
Why it matches plant phenotyping methodsCROPORBITは衛星画像・放射モデル・気象データから圃場レベルの作物収量を推定する計算ツールであり、収量マップ生成と収穫機データによる検証が中心です。単なる地域資源監視ではなく、植物・圃場の収量形質を明示的に推定しています。
abstractThe aim of this work is to develop a computational tool, named CROPORBIT, designed to operate at a regional scale for estimating crop yield.
Phenotypic analysis of mature soybeans is a critical aspect of soybean breeding. However, manually obtaining phenotypic parameters not only is time-consuming and labor intensive but also lacks objectivity. Therefore, there is an urgent need for a rapid, accurate, and efficient method to collect the phenotypic parameters of soybeans. This study develops a novel pipeline for acquiring the phenotypic traits of mature soybeans based on three-dimensional (3D) point clouds. First, soybean point clouds are obtained using a multi-view stereo 3D reconstruction method, followed by preprocessing to construct a dataset. Second, a deep learning-based network, PVSegNet (Point Voxel Segmentation Network), is proposed specifically for segmenting soybean pods and stems. This network enhances feature extraction capabilities through the integration of point cloud and voxel convolution, as well as an orientation-encoding (OE) module. Finally, phenotypic parameters such as stem diameter, pod length, and pod width are extracted and validated against manual measurements. Experimental results demonstrate that the average Intersection over Union (IoU) for semantic segmentation is 92.10%, with a precision of 96.38%, recall of 95.41%, and F1-score of 95.87%. For instance segmentation, the network achieves an average precision (AP@50) of 83.47% and an average recall (AR@50) of 87.07%. These results indicate the feasibility of the network for the instance segmentation of pods and stems. In the extraction of plant parameters, the predicted values of pod width, pod length, and stem diameter obtained through the phenotypic extraction method exhibit coefficients of determination (R2) of 0.9489, 0.9182, and 0.9209, respectively, with manual measurements. This demonstrates that our method can significantly improve efficiency and accuracy, contributing to the application of automated 3D point cloud analysis technology in soybean breeding.
Why it matches plant phenotyping methods成熟ダイズの3D点群取得・分割・形質抽出パイプラインを開発し、手動測定と検証しており、植物フェノタイピング手法が研究の中心である。
abstractThis study develops a novel pipeline for acquiring the phenotypic traits of mature soybeans based on three-dimensional (3D) point clouds.
With the increasing global demand for food, breeding soybean varieties resistant to dense planting is crucial for achieving high and stable yields. Traditional phenotyping methods are limited by insufficient temporal resolution and challenges in dynamic modeling continuity, making it difficult to elucidate the intrinsic relationship between canopy development rate and yield stability. Moreover, existing machine learning models often neglect temporal dependencies in time series predictions, leading to insufficient biological interpretability. This study proposes an innovative approach integrating spatiotemporal deep learning and dynamic modeling to quantify the dynamic changes in canopy parameters using UAV high-throughput phenotyping technology, revealing the key regulatory mechanisms of traits associated with resistance to dense planting. Based on a two-year field experiment (2022-2023) in northeast China (Qiqihaer, black soil region), this study set high (50w plants/ha) and low density (30w plants/ha) treatments across 208 soybean varieties, combined with multispectral UAV imagery (15-18 times per season) and ground-truth data, to develop a time series prediction model for leaf area index (LAI). Comparing the performance of spatiotemporal residual networks (ST-ResNet), long short-term memory networks (LSTM), and traditional random forests (RF), the ST-ResNet model demonstrated significantly superior prediction accuracy (R² = 0.90, RMSE = 0.23 m²/m²), effectively capturing the continuous dynamics of canopy growth through its spatiotemporal feature fusion ability. By fitting the time series curves of LAI, canopy cover (CC), and plant height (PH) with P-spline, 15 intermediate traits (e.g., ΔMeanLAI₋ₘᵢd) were extracted. Mixed models and SHAP interpretability analysis showed that ΔMeanLAI₋ₘᵢd was most correlated with the dense planting yield index (ΔYield, r = 0.51). Furthermore, the high-frequency data acquisition and automated analysis framework using UAVs enabled high-throughput phenotypic screening for 208 varieties per year, significantly improving efficiency compared to traditional methods that rely on manual sampling. This study pioneers the integration of spatiotemporal deep learning with dynamic trait modeling, markedly improving the temporal continuity and stability of LAI estimation compared to traditional single-time-point prediction methods. This advancement allows for more precise quantification of canopy development rates across various growth stages, enabling a systematic analysis of how these dynamic patterns influence resistance to dense planting. By elucidating the dynamic relationship between intermediate traits and yield, this approach offers a high-precision, interpretable phenotypic analysis framework for effectively screening soybean varieties resilient to dense planting.
Why it matches plant phenotyping methodsUAV画像と時系列深層学習を用いてLAI・被覆率・草丈などの植物形質を推定し、検証・比較した高スループット表現型解析フレームワークが研究の中心であるため。
abstractThis study proposes an innovative approach integrating spatiotemporal deep learning and dynamic modeling to quantify the dynamic changes in canopy parameters using UAV high-throughput phenotyping technology
Highlights The performance of RTK-GNSS-enabled drones without GCPs was assessed in agricultural applications. A centimeter-level positioning accuracy was achieved when drones were functioning with RTK-GNSS-enabled. RTK-GNSS-enabled drones without GCPs performed competitively with GCP-based methods in plant height estimation. Using RTK-enabled drones without GCPs provides a time- and labor-saving solution for agricultural practices. ABSTRACT. Commercial drones equipped with Real-Time Kinematic GNSS (RTK-GNSS) technology have been available for several years and are now a standard feature in modern models. This technology has the potential to eliminate the need for ground control points (GCPs). However, despite its widespread adoption and the positioning accuracy claimed in product specifications, limited literature exists on quantifying and systematically evaluating its performance in agricultural field settings for specific applications, such as measuring canopy height profiles. This study aimed to evaluate the potential of RTK-GNSS-enabled drones to estimate plant height without GCPs by comparing it to a benchmark method that uses GCPs across three crops: triticale, maize, and soybean. Three methods were compared: (A) regular-GNSS with GCPs, (B) RTK-GNSS with GCPs, and (C) RTK-GNSS without GCPs. Prior to evaluating the performance in height estimation, the positioning accuracy of RTK-GNSS-enabled drones was assessed, and results confirmed the centimeter-level positioning accuracy achieved by drones with RTK-GNSS-enabled. For plant height estimation, Method-C (RTK-GNSS without GCPs) performed competitively with GCP-based methods, achieving R2 values from 0.84 to 0.99 for drone-estimated versus manually measured plant heights for the three crops, with a mean RMSE of 12.15 cm (compared to Method-A R2 of 0.86 to 0.99 with a mean RMSE of 11.86 cm, and Method-B R2 of 0.84 to 0.97 with a mean RMSE of 11.49 cm). Results suggest that RTK-GNSS-enabled drones can reliably estimate plant height without GCPs, offering a time- and labor-saving solution for field measurements in breeding research and agricultural production. Keywords: Keywords.,Global positioning system (GPS), Ground control point (GCP), High-throughput plant phenotyping, Plant height, Structure-from-motion, Unmanned aerial vehicle (UAV).
Why it matches plant phenotyping methodsRTK-GNSSドローンによる植物高推定法をGCP方式と比較検証し、位置精度と推定性能を評価しており、植物フェノタイピング手法が研究の中心である。
abstractThis study aimed to evaluate the potential of RTK-GNSS-enabled drones to estimate plant height without GCPs by comparing it to a benchmark method that uses GCPs across three crops: triticale, maize, and soybean.
Spectral imaging has been widely applied for soybean phenotyping to find and maintain favorable traits. Specifically, in soybean phenotyping, hyperspectral imaging through contact-based proximal sensing demonstrates better signal-to-noise ratio and resolution compared to remote sensing. However, it has not been adapted for large-scale field applications due to its low throughput and high labor costs. Additionally, no automation solution has been developed to collect in vivo contact-based hyperspectral images of soybean plants. In this study, a novel drone-based robotic system was developed to automate the collection of in vivo contact-based hyperspectral images in the field. The system consists of a machine vision system to detect and estimate the pose of soybean leaflets, an articulated robotic arm with specialized control and path planning algorithms to operate contact-based sensors to grasp and image the leaf, and a customized high-payload drone to provide mobility for sampling at different locations across a field. The average accuracy of the optimized machine vision algorithm is 95.88% for leaf detection and 97.54% for leaf pose estimation, and the average success rate of leaf grasping is 90.55%. This study presents an innovative method for expanding the applicability in vivo contact-based hyperspectral imaging for extensive agricultural applications.
Why it matches plant phenotyping methods植物葉の接触型ハイパースペクトル画像を自動取得するロボットシステムと、葉検出・姿勢推定・把持の技術を開発しており、植物フェノタイピング手法が研究の中心である。
abstractIn this study, a novel drone-based robotic system was developed to automate the collection of in vivo contact-based hyperspectral images in the field.
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 14 Sept 2026
Abstract Soybean ( Glycine max (L.) Merr.) breeding programs face challenges in evaluating large progeny populations, which is labor‐ and resource‐intensive. This study addresses these challenges using high‐throughput phenotyping and machine learning (ML) models to predict phenotypic traits in soybeans. We developed and validated ML models using vegetation indices and canopy images from aerial imagery. A total of 275 soybean genotypes were characterized across two environments and management practices. A total of 11 classical traits were measured, and five vegetation indices were calculated from aerial images at different growth stages. ML algorithms, including support vector machine for regression, random forest (RF), multilayer perceptron (MLP), and adaptive boosting, were employed. Additionally, convolutional neural networks with transfer learning were used to extract features from the images. Significant correlations were found between agronomic traits, vegetation indices, and canopy characteristics. The high heritability of the red–green–blue vegetation index and green leaf index (mean broad‐sense heritability of 0.56) compared to other RGB‐based indices indicates their potential usefulness in genetic evaluations. Advanced ML techniques, particularly transfer learning with ResNet 50, enhanced the prediction of phenotypic traits such as days to the R7 growth stage (DR7) and plant height at maturation (PHM). The integration of ResNet 50 with RF achieved a prediction accuracy of 0.64 for DR7, while ResNet 50 with MLP reached an accuracy of 0.68 for PHM. These findings highlight the potential of these techniques to improve decision‐making in soybean breeding. Lastly, principal component analysis identified genotypes with desirable trait combinations, advancing soybean development.
Why it matches plant phenotyping methods航空画像・植生指数・機械学習を用いた植物形質予測モデルを開発・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。
abstractWe developed and validated ML models using vegetation indices and canopy images from aerial imagery.
Cotton and soybeans are important crops for the country's economic growth. Due to the rapid spread of disease, plants are susceptible to bacterial and viral diseases. Early identification and classification using machine or deep learning models aid farmers in reducing potential losses. Model‐based detection necessitates a large number of training samples and high‐quality images. Thus, this study generates new datasets to diagnose soybean and cotton plant diseases. The images are collected with the help of the Central Institute for Cotton Research (CICR) in Nagpur, Maharashtra, to create a clean and comprehensive dataset for research purposes. The dataset contains 5200 images, including both diseased and healthy images. The collected images are labelled using the Robo flow tool, masked with the Photoshop tool and stored in the dataset. The generated dataset is examined through pre‐processing and classification using the novel proposed algorithms. Initially, the Gabor filter is used for pre‐processing to eliminate unwanted noise from the collected images. Afterwards, the Position attention‐based capsule network (PA‐CapNet) model is proposed to perform multidisease classification for the soybean and cotton datasets. Finally, the performances are assessed by evaluating varied metrics. The result analysis shows that the proposed method obtains better results than the other existing models. The proposed method obtains an accuracy of 98% for the soybean dataset and 96.89% for the cotton dataset.
Why it matches plant phenotyping methods植物葉の画像から病害状態を分類するデータセットの生成・前処理・評価が中心であり、植物病害フェノタイピング手法および再利用可能なデータセットに該当する。
abstractThus, this study generates new datasets to diagnose soybean and cotton plant diseases.
Artificial intelligence (AI) in soybean research has revolutionized various crop improvement and production aspects. This review provides predominant areas that have seen the use of AI. AI applications in phenomics have enabled collecting and analyzing high-dimensional data in soybean plants, from below- to above-ground traits, predicting phenotypes, and identifying complex patterns. In genomics, AI has improved genomic selection accuracy and identified genomic regions associated with traits of interest, such as resistance to biotic and abiotic stresses. AI has also been extensively used in detecting and managing biotic and abiotic plant stresses using RGB, multispectral, and thermal imagery from ground-based and aerial platforms. Additionally, AI has shown significant potential in yield prediction, incorporating factors such as vegetation indices, weather data, and soil properties. This review explains the concept of cyber-agricultural systems (CAS) that integrates AI, advanced sensing, computational modeling, and scalable cyberinfrastructure to optimize soybean production, enhance resource management, reduce environmental impact, and improve farm efficiency. We explain the use of CAS in crop improvement as well. We provide an exhaustive listing of challenges and future direction in the integration of AI in soybean production and crop improvement, including multi-modal and layered sensing, data availability and quality, computational modeling, AI models and tools, Cyberinfrastructure, Explainability and interpretability of AI models, AI-related impacts on privacy, ethics, and policy, Impact on Smallholder Farmers, Digital Twin, Large Soybean Datasets for community usage, and Immersive environments.
Why it matches plant phenotyping methods大豆育種・生産におけるAIの総説であり、植物フェノミクス、画像センシング、表現型予測を主要な対象として扱っているため、フェノタイピング方法レビューに該当する。
abstractAI applications in phenomics have enabled collecting and analyzing high-dimensional data in soybean plants, from below- to above-ground traits, predicting phenotypes, and identifying complex patterns.
High‐throughput phenotyping is an emerging tool that allows access to identify simple and complex traits, accelerating genetic discoveries and selection. Vegetation indices strongly correlate with several economic crop traits, allowing plant breeders to detect variation in breeding populations. Thus, this study used red–green–blue (RGB) vegetation indices to evaluate the influence of the stink bug complex (Euschistus heros, Piezodorus guildinii, Nezara viridula, Dichelops melacanthus, and Edessa meditabunda) on the agronomical traits of soybean (Glycine max) lineages. For instance, two experiments were conducted to assess soybean resistance to the stink bug complex, (1) with and (2) without pesticide control. An unmanned aerial vehicle coupled with an RGB camera acquired aerial photography over the field during the R5 stage. Four vegetation indices and canopy were estimated from the orthomosaic, and the genotypes were evaluated based on agronomical traits. Linear mixed models were used to estimate the variance and significance test of each trait using the likelihood ratio test, and the principal component analysis was performed to verify the multivariate pattern among genotypes. The results showed significant genotypic effects for most traits with high broad‐sense heritability for agronomical traits and moderate for vegetation indices. Significant correlations using best linear unbiased predictions were observed among the agronomical traits with the vegetation indices and canopy coverage, which can be used as a tool for the indirect selection of soybean lineages in the breeding pipeline.
Why it matches plant phenotyping methodsUAV搭載RGBカメラとオルソモザイクから植生指数・キャノピーを抽出し、ダイズの農業形質との相関および育種選抜への利用を評価しており、表現型取得法の実質的応用が中心である。
abstractAn unmanned aerial vehicle coupled with an RGB camera acquired aerial photography over the field during the R5 stage.
Genomic selection (GS) and phenotypic selection (PS) are widely used for accelerating plant breeding. However, the accuracy, robustness, and transferability of these two selection methods are underexplored, especially when addressing complex traits. In this study, we introduce a novel data fusion framework, GPS (Genomic and Phenotypic Selection), designed to enhance predictive performance by integrating genomic and phenotypic data through three distinct fusion strategies: data fusion, feature fusion, and result fusion. The effectiveness and generalizability of GPS framework were rigorously tested using an extensive suite of models, including statistical approaches (GBLUP and BayesB), machine learning models (Lasso, RF, SVM, XGBoost, and LightGBM), a deep learning method (DNNGP) and a latest phenotype-assisted prediction model (MAK). These models were applied to large-scale datasets from four crop species: maize, soybean, rice, and wheat, demonstrating the versatility and robustness of the framework. The results indicated that: (1) Data fusion achieved the best accuracy than the feature fusion and result fusion strategies. The top-performing data fusion model (Lasso_D) improved the selection accuracy by 53.4% compared to the best GS models (LightGBM) and by 18.7% compared to the best PS models (Lasso). (2) Lasso_D exhibited exceptional robustness, achieving high predictive accuracy even in sample size as small as 200. Additionally, the model demonstrated resilience to variations in SNP (single nucleotide polymorphism) density, underscoring its adaptability to diverse data conditions. Moreover, the model’s accuracy improved with the number of auxiliary traits and their correlation strength with target traits, further highlighting its adaptability to complex trait prediction. (3) The best-performing data fusion model demonstrated broad transferability, with substantial improvements in predictive accuracy when incorporating multi-environmental data. Notably, this enhancement resulted in only a 0.3% reduction in accuracy compared to predictions generated using data from the same environment, affirming the model’s reliability in cross-environmental scenarios. This study provides groundbreaking insights, pushing the boundaries of predictive accuracy, robustness, and transferability in trait prediction. These findings represent a significant contribution to plant science, breeding, and the broader interdisciplinary fields of statistics and artificial intelligence.
Why it matches plant phenotyping methods植物の形質予測を目的とするデータ融合・機械学習フレームワークを開発し、複数作物・モデルで精度、頑健性、移植性を検証しており、形質推定手法が研究の中心である。
abstractwe introduce a novel data fusion framework, GPS (Genomic and Phenotypic Selection), designed to enhance predictive performance by integrating genomic and phenotypic data through three distinct fusion strategies: data fusion, feature fusion, and result fusion.
Plant phenotyping relevance match · UnverifiedOpenAlex · Europe PMC · checked 6 Sept 2026
This study investigates the effectiveness of high-throughput phenotyping (HTP) using RGB images from unmanned aerial vehicles (UAVs) to assess vegetation indices (VIs) in different soybean pure lines. The VIs were accessed at various stages of crop development and correlated with agronomic performance traits. The field research was conducted in the experimental area of the Mato Grosso do Sul Foundation, Brazil, with 60 soybean pure lines. RGB images were captured at multiple stages of development (28, 37, 49, 70, 86, 105, 115, and 120 days after sowing). We used a linear mixed effects model, with restricted maximum likelihood (REML)/best linear unbiased prediction (BLUP) methods, to estimate variance components and genetic correlations, and to predict genotypic values. Significant genetic differences were identified among genotypes for all agronomic traits evaluated (p< 0.001), with high accuracy and heritability for plant height, maturity at R8, and 100-seed weight. There was a significant genotype × flight data interaction impact on VI expression, emphasizing the importance of timing data collection to enhance HTP with VIs in agronomic performance evaluation. In the early stages, the indices varied depending on the environment. On the other hand, the indices showed higher correlations with the traits of plant height and maturity at the R8 stage, at 105, 115, and 120 days after sowing. HTP with VIs based on RGB images from UAVs has proven to be more effective in the early and final stages of soybean development, providing essential information for the selection of superior genotypes. This study highlights the importance of the temporal approach in HTP, optimizing the selection of soybean genotypes and refining agricultural management strategies.
Why it matches plant phenotyping methodsUAVのRGB画像から植生指数を抽出するハイスループット植物表現型解析を、複数時期・遺伝子型で評価し、農業形質との関連および遺伝的評価に用いており、表現型取得法が研究の中心である。
titleHigh throughput phenotyping in soybean breeding using RGB image vegetation indices based on drone.
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.
LiDAR sensors have great potential for enabling crop recognition (e.g., plant height, canopy area, plant spacing, and intra-row spacing measurements) and the recognition of agricultural working environments (e.g., field boundaries, ridges, and obstacles) using agricultural field machinery. The objective of this study was to review the use of LiDAR sensors in the agricultural field for the recognition of crops and agricultural working environments. This study also highlights LiDAR sensor testing procedures, focusing on critical parameters, industry standards, and accuracy benchmarks; it evaluates the specifications of various commercially available LiDAR sensors with applications for plant feature characterization and highlights the importance of mounting LiDAR technology on agricultural machinery for effective recognition of crops and working environments. Different studies have shown promising results of crop feature characterization using an airborne LiDAR, such as coefficient of determination (R2) and root-mean-square error (RMSE) values of 0.97 and 0.05 m for wheat, 0.88 and 5.2 cm for sugar beet, and 0.50 and 12 cm for potato plant height estimation, respectively. A relative error of 11.83% was observed between sensor and manual measurements, with the highest distribution correlation at 0.675 and an average relative error of 5.14% during soybean canopy estimation using LiDAR. An object detection accuracy of 100% was found for plant identification using three LiDAR scanning methods: center of the cluster, lowest point, and stem–ground intersection. LiDAR was also shown to effectively detect ridges, field boundaries, and obstacles, which is necessary for precision agriculture and autonomous agricultural machinery navigation. Future directions for LiDAR applications in agriculture emphasize the need for continuous advancements in sensor technology, along with the integration of complementary systems and algorithms, such as machine learning, to improve performance and accuracy in agricultural field applications. A strategic framework for implementing LiDAR technology in agriculture includes recommendations for precise testing, solutions for current limitations, and guidance on integrating LiDAR with other technologies to enhance digital agriculture.
Why it matches plant phenotyping methodsLiDARによる植物形質（草丈、樹冠面積、株間など）の取得・評価方法、試験手順、精度ベンチマークを中心にレビューしており、植物フェノタイピング手法が主要テーマである。
abstractThe objective of this study was to review the use of LiDAR sensors in the agricultural field for the recognition of crops and agricultural working environments.
In order to achieve precise discrimination of leaf diseases in the Maize/Soybean intercropping system, i.e. leaf spot disease, rust disease, mixed leaf diseases, this study utilized hyperspectral imaging and deep learning algorithms for the classification of diseased leaves of maize and soybean. In the experiments, hyperspectral imaging equipment was used to collect hyperspectral images of leaves, and the regions of interest were extracted within the spectral range of 400 to 1000 nm. These regions included one or more infected areas on the leaves to obtain hyperspectral data. This approach aimed to enhance the accurate discrimination of different types of diseases, providing more effective technical support for the detection and control of crop diseases. The preprocessing of hyperspectral data involved four methods: Savitzky-Golay (SG), Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC) and 1st Derivative (1st Der). The 1st Der was found to be the optimal preprocessing method for hyperspectral data of maize and soybean diseases. Competitive Adaptive Reweighted Sampling (CARS), Successive Projections Algorithm (SPA) and Principal Component Analysis (PCA) were employed for feature extraction on the optimal preprocessed data. The Support Vector Machines (SVM), Bidirectional Long Short-Term Memory Network (BiLSTM) and Dung Beetle Optimization-Bidirectional Long Short-Term Memory Network (DBO-BiLSTM) were established for the discrimination of maize and soybean diseases. Comparative analysis indicated that, in the classification of maize and soybean diseases, the DBO-BiLSTM model based on the CARS extraction method (1st Der-CARS-DBO-BiLSTM) demonstrated the highest classification rate, reaching 98.7% on the test set. The research findings suggest that integrating hyperspectral imaging with both traditional and deep learning methods is a viable and effective approach for classifying diseases in the intercropping model of maize and soybean. These results offer a novel method and a theoretical foundation for the non-invasive, precise, and efficient identification of diseases in the intercropping model of maize and soybean, carrying positive implications for agricultural production.
Why it matches plant phenotyping methodsトウモロコシ・ダイズ葉の病害状態をハイパースペクトル画像から分類する取得・解析手法が研究の中心であり、前処理、特徴抽出、モデル比較と性能評価を含むため。
abstractthis study utilized hyperspectral imaging and deep learning algorithms for the classification of diseased leaves of maize and soybean.
SoybeanTomatoWatermelonField / plotFruitStomata / guard-cell complexPhysiological trait estimationGrowth / development / phenologyStomatal traitsWater status / transpiration
The combination of flexible electronics and plant science has generated various plant-wearable sensors, yet challenges persist in their applications in real-world agriculture, particularly in high-throughput settings. Overcoming the trade-off between sensing sensitivity and range, adapting them to a wide range of crop types, and bridging the gap between sensor measurements and biological understandings remain the primary obstacles. Here we introduce PlantRing, an innovative, nano-flexible sensing system designed to address the aforementioned challenges. PlantRing employs bio-sourced carbonized silk georgette as the strain sensing material, offering exceptional resolution (tensile deformation: < 100 μm), stretchability (tensile strain up to 100 %), and remarkable durability (season long), exceeding existing plant strain sensors. PlantRing effectively monitors plant growth and water status, by measuring organ circumference dynamics, performing reliably under harsh conditions and being adaptable to a wide range of plants. Applying PlantRing to study fruit cracking in tomato and watermelon reveals novel hydraulic mechanism, characterized by genotype-specific excess sap flow within the plant to fruiting branches. Its high-throughput application enabled large-scale quantification of stomatal sensitivity to soil drought, a traditionally difficult-to-phenotype trait, facilitating drought tolerant germplasm selection. Combing PlantRing with soybean mutant led to the discovery of a potential novel function of the GmLNK2 circadian clock gene in stomatal regulation. More practically, integrating PlantRing into feedback irrigation achieves simultaneous water conservation and quality improvement, signifying a paradigm shift from experience- or environment-based to plant-based feedback control. Collectively, PlantRing represents a groundbreaking tool ready to revolutionize botanical studies, agriculture, and forestry.
Why it matches plant phenotyping methodsPlantRingという高スループットの植物装着型センサーを開発し、器官周径、水状態、気孔感度などの植物形質を直接測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。
abstractHere we introduce PlantRing, an innovative, nano-flexible sensing system designed to address the aforementioned challenges.
Hyperspectral imaging (HSI) is a prevalent method in crop phenotyping. Nevertheless, current HSI remote sensing techniques are compromised by changing ambient lighting conditions, long imaging distances, and comparatively low resolutions. Proximal HSI sensors such as LeafSpec were developed to improve the imaging quality. However, the application of proximal sensors remains contrained by their low throughput and intensive labor costs. Moreover, few automation solutions were available to use LeafSpec in phenotyping dicot plants. In this paper, a novel robotic system is presented as a sensor platform to operate LeafSpec to collect leaf-level hyperspectral images for in vivo phenotyping of soybean. A machine vision algorithm was developed to detect the top mature trifoliate and estimate the poses of the leaflets. A control and motion planning algorithm was developed for an articulated robotic manipulator to grasp the target leaflets. An experiment was conducted in March 2021 in a greenhouse with 64 soybean plants of 2 genotypes and 2 nitrogen treatments. The machine vision detected the target leaflets with a first trial success rate of 84.13% and an overall success rate of 90.66%. The robotic manipulator operated LeafSpec to image the target leaflets with a first trial success rate of 87.30% and an overall success rate of 93.65%. The average cycle time for one soybean plant was 63.20 s. The PLS predictions from the robot-collected data had an R² of 0.84 with the measured nitrogen content and an R² of 0.82 with the predictions from human-collected data. The results demonstrated the potential of applying the system for automated in vivo leaf-level HSI for soybean phenotyping in the field.
Why it matches plant phenotyping methodsロボットによる近接ハイパースペクトル画像取得システムと葉検出・動作計画アルゴリズムを開発し、豆類の表現型計測性能を検証しており、方法が中心的である。
abstractIn this paper, a novel robotic system is presented as a sensor platform to operate LeafSpec to collect leaf-level hyperspectral images for in vivo phenotyping of soybean.
Sun-induced chlorophyll fluorescence (SIF) has recently emerged as a proxy for canopy photosynthesis of vegetation and offers a promising approach for scalable remote crop monitoring. Effective application of SIF for crop monitoring requires better understanding of the processes that cause SIF-photosynthesis decoupling at leaf and canopy scales. To answer this challenge, we developed a novel automated multi-targeting hyperspectral spectrometer (OctoFlox). First, we evaluated the performance of OctoFlox and found high stability and cross-channel comparability. Second, we performed an evaluation of different SIF retrieval methods to identify the best suited retrieval method for our system configuration for both red (SIFʀᴇᴅ) and far-red SIF (SIFꜰʀ). We then deployed OctoFlox within Soil-Plant Atmosphere Research (SPAR) controlled-environment chambers that enable measurement of canopy-scale SIF and photosynthesis with matching footprints. We analyzed the effect of the SPAR chamber tops on the light environment and found minimal impact on the spectral response. Lastly, we examined the response of SIF and canopy photosynthesis using the SPAR chambers. Soybean plants were evaluated at pre-drought, drought (irrigated at 100 % field capacity vs. 33 % field capacity for 2 weeks) and after 1 week recovery from drought. During early growing season, SIFꜰʀ and SIFʀᴇᴅ exhibited similar responses. At peak growing season (R2 growth stage), SIFꜰʀ increased during afternoon depression of photosynthesis, but SIFʀᴇᴅ decreased. We demonstrate that pairing SIF instrumentation with SPAR chambers can accelerate understanding SIF-photosynthesis relationships from diurnal to seasonal scales in relation to crop physiological responses to abiotic stress. We provide user recommendations for future applications using OctoFlox and SPAR chambers for co-measuring SIF and GPP.
Why it matches plant phenotyping methodsOctoFlox分光計の開発・性能評価、SIF検索法の比較、SPARチャンバーとの統合および作物ストレス時のSIF・光合成計測を中心とする植物フェノタイピング手法研究である。
abstractwe developed a novel automated multi-targeting hyperspectral spectrometer (OctoFlox).
Meeting the growing demand for soybeans will require increased production. One approach would be to reduce yield loss from plant diseases. In the U.S., soybean diseases account for approximately 8–25% of average annual yield loss. Early and accurate detection of pathogens is key for effective disease management strategies and can help to minimize pesticide usage and thus boost overall productivity. Recent advancements in computer vision could move us towards that goal by making disease diagnostics expertise more readily accessible to every-one. To that end, we developed an automated classifier of digital images of soybean diseases, based on convolutional neural networks (CNN). For model training and validation, we acquired more than 9,500 original soybean images, representing eight distinct disease and deficiency classes: (1) healthy/asymptomatic, (2) bacterial blight, (3) Cercospora leaf blight, (4) downy mildew, (5) frogeye leaf spot, (6) soybean rust, (7) target spot, and (8) potassium deficiency. To make training more efficient we experimented with a variety of approaches to transfer learning, data engineering, and data augmentation. Our best performing model was based on the DenseNet201 architecture. After training from scratch, it achieved an overall testing accuracy of 96.8%. Experimenting with full or partial freezing of core DenseNet201 model weights did not improve performance. Neither did a deliberate effort to increase the diversity of subject backgrounds in the digital images. Models performed best when trained on datasets composed exclusively of images of soybean leaves still attached to the plant in the field; conversely, mixing in images of detached leaves on simple backgrounds reduced performance. On the other hand, data augmentation to increase representational parity across disease classes provided a substantial performance boost. Our development experience may provide useful insights for researchers considering how to best build and analyze datasets for similar applications.
Why it matches plant phenotyping methods植物葉の病害・欠乏状態を画像から分類するCNN手法の開発と検証が研究の中心であり、植物の病徴状態を直接推定しているため。
abstractwe developed an automated classifier of digital images of soybean diseases, based on convolutional neural networks (CNN).
The growth of precision agriculture has allowed farmers access to more data and greater efficiency for their farms. With consistently tight profit margins, farmers need ways to take advantage of the advancement of technology to lower their costs or increase their revenue. One area where these advancements can prove beneficial are in the measurement of vegetation indices such as the Normalized Difference Vegetation Index (NDVI) and Normalized Difference Red Edge Index (NDRE). Color maps representing these vegetation indices can be used to identify problem areas, plant health, or even places where spot applications are needed. These color maps help farmers to visualize these areas. Currently, a multi-thousand dollar multispectral camera, typically attached to an Unmanned Aerial Vehicle (UAV) during flight, is required for measuring these indices. This makes obtaining NDVI and NDRE somewhat cost prohibitive for most farmers. This work demonstrates a solution to this cost issue. The solution involves the use of a conditional Generative Adversarial Network known as Pix2Pix. By using Pix2Pix along with training data from UAV flights of corn, soybeans, and cotton, this paper highlights the potential for predicting comparable NDVI and NDRE with a low-cost Red-Green-Blue (RGB) camera. This paper proposes and assesses a cost-efficient method that can comparably predict these vegetation indices, resulting in cost-savings in the range of $5000 per UAV system.
Why it matches plant phenotyping methodsRGB航空画像とPix2Pixを用いてNDVI・NDREという植物状態指標を推定する手法を提案・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。
abstractThis work demonstrates a solution to this cost issue.
Rapid and effective identification and diagnosis of soybean drought conditions is crucial for soybean yield and quality. Due to the complexity and diversity of agricultural environments, deep learning models based on three-dimensional data suffer from low accuracy and slow efficiency in practical applications, this paper proposes a three-dimensional image recognition method for soybean canopy based on an improved multi-view network. A lightweight network Res2net was used to reconstruct the feature extraction skeleton network in the MVCNN model, and the group convolution module of the network was optimized by embedding the ECA attention mechanism to propose a new three-dimensional image recognition model based on multi-view network (ECA-MVRes2net). In the study, drought soybeans were used as an example to obtain projected images of soybean canopy in six viewpoints using three-dimensional rotation and image feature theory, and the proposed ECA-MVRes2net was applied to carry out three-dimensional image recognition experiments of drought soybeans, and its recognition accuracy, F1 value and Kappa coefficient reached 96.665 %, 96.7 % and 0.924, respectively, compared with MVCNN, MVResnet, Pointnet++ and PointConv models with 3 evaluation metrics average improved by 17.289 %, 17.43 % and 0.356, respectively. The result realized a lightweight fast and accurate network model suitable for three-dimensional image recognition, which provides a theoretical foundation and technical support for the rapid recognition and accurate management of crops based on three-dimensional image processing.
Why it matches plant phenotyping methods乾燥状態という植物状態を対象に、3次元画像と改良マルチビュー深層学習モデルによる認識手法を開発・評価しており、表現型取得・推定が研究の中心である。
abstractthis paper proposes a three-dimensional image recognition method for soybean canopy based on an improved multi-view network.
SoybeanAerial / UAVField / plotMultispectral / hyperspectralLeafSeed / grainClassificationGrowth / development / phenologyPigment / colour / senescenceWater status / transpiration
Timely and accurate determination of the maturity status during the soybean harvest period is crucial for devising strategic harvesting plans, significantly enhancing the intelligent management of soybean production and minimizing losses. To explore the maturation dynamics of soybeans and pinpoint the precise maturity stage for optimal harvesting, we collected Unmanned Aerial Vehicle (UAV) remote sensing data, along with physical and chemical parameters, during the Beginning maturity(R7)-Full maturity(R8) stages from 2021 to 2022. We analyzed the variations in these parameters throughout the maturation period. Employing UAV multispectral technology and correlation analysis algorithms, we established a relationship between the spectral and physicochemical parameters during the maturation period, which led to the formulation of a soybean maturity evaluation index. Furthermore, we incorporated the concept of precision management zoning. In line with actual field production requirements, this index, combined with real-time moisture content, established a standard for grading soybean maturity. Additionally, our study examined the impact of spatial scale on maturity assessments by conducting a theoretical analysis of regionalized variables and analyzing the spatial correlations and variability of soybean maturity across different field plots. We determined the optimal theoretical model for various sampling distances, thereby establishing feasible zoning ranges. Comparative analysis of zoning algorithms, including Hierarchical Clustering (HC), K-Means, and its optimized versions K-Means++ and Mini Batch K-Means, was conducted using the partition performance index as the criterion. This analysis identified the most suitable algorithm for assessing soybean field maturity, which was subsequently verified through field tests. Results indicated that soybean leaf chlorophyll content and photosynthesis decreased sharply initially, then stabilized at a low rate. Water content in both soybean plants and seeds decreased rapidly before slowing, with plant dehydration rates exceeding those of the seeds. Green Normalized Difference Vegetation Index(GNDVI) showed a positive correlation with various physiological indicators, and soybean maturity was classified into three levels using the unit gridding management method, considering actual moisture content. Spatial correlation analysis of maturity variations in small and large test fields revealed Moran indices ranging from 0.78 to 0.95 and 0.70–0.90, respectively. Optimal detection ranges were determined to be 2.00–30.00 m for small fields and 2.00–67.00 m for large fields. The zoning categories for small and large fields were three and four, respectively. Comparative studies showed that the Mini Batch K-Means algorithm, an optimization of K-Means, achieved Calinski-Harabasz Index(CH) scores and silhouette coefficients of 10521.97, 109508.27, and 0.57, 0.54, respectively, indicating comparable zoning effectiveness to K-Means and K-Means++ and superiority over the HC algorithm. The zoning speeds were 12.50 s and 191.68 s, respectively, underscoring its efficiency in maturity zoning. Furthermore, zoning results corresponded well with actual soybean maturity. Terrain had little impact on soybean maturity, which was more susceptible to slope aspect. This study provides a theoretical foundation and practical guidelines for applying UAV remote sensing technology in determining soybean maturity and optimal harvesting periods, thereby reducing potential harvest losses.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像からダイズ成熟度を推定・分類し、空間スケールやゾーニングアルゴリズムを比較検証しており、植物状態の取得・抽出手法が中心である。
abstractEmploying UAV multispectral technology and correlation analysis algorithms, we established a relationship between the spectral and physicochemical parameters during the maturation period, which led to the formulation of a soybean maturity evaluation index.
Agricultural production models predict crop yield by accounting for a variety of species, cultivar, farming management, and environmental impacts on crop photosynthesis. Without suitable constraints, however, large uncertainties may exist in simulations of crop photosynthesis. Recent advances in retrieving solar-induced chlorophyll fluorescence (SIF) at the top-of-canopy (TOC) have provided a promising measurement for crop photosynthesis. Within the framework of the APSIM (Agricultural Production Systems sIMulator) model, a SIF module was developed to connect crop photosynthesis to TOC SIF emission (SIFₜₒc) which can be measured by remote sensing platforms. The new model (APSIM-SIF) first estimates the leaf-level chlorophyll fluorescence emitted over the full SIF spectrum (SIFₜₒₜ_fᵤₗₗ) according to CO₂ assimilation in crops. The model then mechanistically decomposes the conversion from SIFₜₒₜ_fᵤₗₗ to SIFₜₒc into two factors: the SIF band conversion factor (ɛ) and the fluorescence escape ratio (fₑₛc) that represent the impact of leaf physiological status and plant structure properties, respectively. ɛ can be estimated using leaf structural and biochemical parameters as inputs; fₑₛc for near-infrared SIF can be expressed as a function of directional reflectance in the near-infrared region (RNIR), Normalized Difference Vegetation Index (NDVI), and the fraction of PAR absorbed by crops (fAPAR). The APSIM-SIF model determined more than 90% of the variation in gross primary productivity (GPP), aboveground biomass and leaf area index (LAI) measurements for maize (Zea mays L.) at two AmeriFlux sites in the U.S. Midwest and it also captured the seasonality of SIF (R² = 0.84) and GPP (R² = 0.81) well at an irrigated maize site in China. The APSIM-SIF model was also applied to the simulation of TOC SIF emission of maize and soybean (Glycine max L.) in the U.S. Midwest during the 2018 growing season. The simulated SIFₜₒc accounted for more than 75% of the variability of daily satellite SIF observations for grid squares with more than 70% crop area. The main contribution of this study lies in two aspects: (1) a physically-based framework is proposed to incorporate the SIF module to the APSIM-DCaPST model, and (2) the two important factors used in this framework (ɛ and fₑₛc) remains largely constant during the peak growing season. These findings provide a theoretically robust and operational basis for linking SIF observations with crop growth.
Why it matches plant phenotyping methods作物キャノピーのSIFを推定するAPSIM-SIFモジュールを開発し、光合成・GPP・バイオマス・LAI・季節性との技術的検証も行っているため、表現型取得・推定法が中心である。
abstractWithin the framework of the APSIM (Agricultural Production Systems sIMulator) model, a SIF module was developed to connect crop photosynthesis to TOC SIF emission (SIFₜₒc) which can be measured by remote sensing platforms.
The soybean aphid (SBA), Aphis glycines Matsumura (Hemiptera: Aphididae), is a significant insect pest of soybean, Glycine max (L.) Merrill (Fabales: Fabaceae), and field treatment decisions for this pest are based on average field populations. Previous studies indicated that ground- and drone-based red-edge and near-infrared remote sensing can be used to detect plant stress caused by SBA infestations in soybean. However, it remains to be determined if remote sensing for SBA can be expanded to field or landscape scale using satellite-based platforms. Thus, this research was conducted in three steps to determine the potential of using Sentinel-2 satellite data for the classification of SBA infestations in soybean fields using simulated and actual Sentinel-2 satellite spectral reflectance. In the first step, as a proof of concept, hyperspectral data from cage studies were used to simulate Sentinel-2 bands and vegetation indices (VIs), conducted in nine trials at multiple locations between 2013 and 2021. The effects of SBA from caged plants on simulated data were evaluated with random intercept linear mixed models. The satellite simulation indicated a significant effect of SBA on the spectral reflectance of caged soybean plants (p < 0.05) for four satellite bands (5, 6, 7, and 8A) and five VIs (NDVI, GNDVI, SAVI, OSAVI, and NDRE). In the second step, actual Sentinel-2 spectral reflectance and corresponding aphid counts of commercial soybean fields, collected from 2017 to 2019, were obtained. The relationship between SBA counts and Sentinel-2 spectral reflectance from commercial soybean fields were evaluated with general linear models. A significant effect of SBA was observed for three satellite bands (6, 7, and 8A) and three VIs (NDVI, SAVI, and OSAVI). In the third step, linear support vector machine (LSVM) models for the classification of SBA infestations as above or below a previously determined economic threshold of 250 aphids per plant were developed using simulated Sentinel-2 bands and VIs from the caged plots, and were tested on actual Sentinel-2 data from commercial soybean fields. The best LSVM model for the classification of aphids in soybean reached 91% accuracy, 85.7% sensitivity, and 93.3% specificity. Thus, simulations with caged plots can be used as an indication of the potential of using satellite data for the detection of plant stresses on a larger scale. Furthermore, this study advances decision-making for SBA, and the developed LSVM model can be used to update regional and local monitoring for the management of SBA.
Why it matches plant phenotyping methodsSentinel-2の分光データからアブラムシによるダイズの植物ストレス・被害状態を推定し、LSVM分類器を開発・実データで検証しており、表現型取得手法が中心である。
abstractthis research was conducted in three steps to determine the potential of using Sentinel-2 satellite data for the classification of SBA infestations in soybean fields using simulated and actual Sentinel-2 satellite spectral reflectance.
The soybean grain yield is affected by several factors, among them, the nutritional deficiency caused by low levels of potassium (K⁺) is one of the main responsible for the reduction in grain yield both in Brazil and worldwide. Traditional methods of nutrient determination involve leaf collection and laboratory procedures with toxic reagents, which is a destructive, time-consuming, expensive, and environmentally unfriendly method. In this context, the use of hyperspectral data and machine learning regression models can be a powerful tool in the nutritional diagnosis of plants. However, the comparison among different machine learning algorithms for K⁺ estimation in soybean leaves from hyperspectral reflectance data is yet to be reported. From this, the goal of this research was to obtain K⁺ prediction models in soybean leaves at different stages of development using hyperspectral data and machine learning regression models with wavelength selection algorithms. The experiment was carried out at the National Soybean Research Centre (Embrapa Soja) in the 2017/2018, 2018/2019 and 2019/2020 soybean crop season, at the stages of development V4–V5, R1–R2, R3–R4 and R5.1–R5.3. The experimental plots were managed to obtain different conditions of K⁺ availability for the plants, from severe deficiency level to the appropriate level of nutrient, under the following experimental treatments: severe potassium deficiency, moderate potassium deficiency and adequate supply of potassium. Spectral data were obtained by the ASD Fieldspec 3 Jr. hyperspectral sensor in the visible/near-infrared spectral range (400–1000 nm) and correlated to leaf K⁺ through ten machine learning methods: Partial Least Square Regression (PLSR), interval Partial Least Squares (iPLS), Genetics Algorithm (GA), Competitive Adaptive Reweighted Sampling (CARS), Random Frog (RF, Frog), Variable combination population analysis (VCPA), Principal Component Regression (PCR), Support Vector Machine (SVM), Successive projections algorithm (SPA), and Stepwise. The results showed that K⁺ deficiency significantly reduce grain yield and nutrient content in the leaf, making enabling the clustering separation of all treatments by Tukey’s test. Among the 601 wavelengths obtained by the sensor, the algorithms selected from 1 to 33.28%, largely distributed in the regions of red, green, blue, red-edge and NIR. In all stages of development, it was possible to quantify the nutrient with high accuracy (R² ≅ 0.88). The multivariate regression models from the selection of variables contributed to increase the accuracy (R²) in about 7.65% for the calibration step and 6.45% for the cross-validation step, when compared to the model using the full spectra. The results obtained demonstrate that the monitoring of K⁺ in soybean leaves is possible and has the potential to determine the nutritional content in the early stages of plant development.
Why it matches plant phenotyping methodsハイパースペクトル計測と機械学習により、ダイズ葉のカリウム濃度という植物生理形質を非破壊推定する手法を比較・評価しており、表現型取得・抽出が中心です。
abstractthe use of hyperspectral data and machine learning regression models can be a powerful tool in the nutritional diagnosis of plants
Crop yield forecasting is an essential component of crop production assessment, impacting people at the global scale down to the level of individual farms. Until now, yield forecasting has predominantly relied on optical data, particularly the maximum value of vegetation indexes. However, this approach only presents a short forecasting window, and it is essential to obtain yield estimates as early as possible in the growing season and then further improve forecasting even after the vegetation index has reached its peak. So far, optical satellite data at high-temporal resolution (1–3 days) has been actively used for real time crop yield monitoring, whereas fewer operational models make a use of synthetic aperture radar (SAR). In this study, we explore whether SAR data can capture distinct aspects of crop dynamics, providing new insights for yield estimation depending on the crop's phenological stage. We assess the efficiency of dual- (Sentinel-1) and quad-polarimetric (UAVSAR, RADARSAT-2) data to explain inter-field crop yield variability for corn, soybean, and rice over a test area in Arkansas, US (258 fields, 2019). We used optical imagery acquired by Planet/Dove-Classic, Sentinel-2, and Landsat 8, to establish a baseline performance of satellite-based indicators to explain yield variability and assess dual- and quad-polarimetric SAR data for crop yield assessment. In terms of polarimetric indexes, the results showed that in general the results for rice were mostly stable and better than the other crops (R2adj ∼ 0.4 on average). The best results were obtained for the Sentinel-1 VHasc with R2adj = 0.47 and RADARSAT-2 phase difference with R2adj = 0.45. The results for corn performed the least with an R2adj 0.4. We also investigated the day of year (DOY) with the maximum correlation between optical and SAR-derived features and the final yields for corn, soybean, and rice. The maximum correlation for optical features occurs over a short time between DOY 155 (June 4) and 185 (July 5) for corn and rice, and DOY 190 (July 9) and DOY 211 (July 30) for soybean, with these results being consistent across various optical-based sensors. On the contrary, the maximum correlation for SAR-derived features varied significantly and was between DOY 120 (April 30) to DOY 225 (August 13). A study of the time series parameters cross-correlation showed that the optical parameters were highly correlated, but the SAR parameters showed strong temporal decorrelation. We conducted a comparison between C-band and L-band to assess their sensitivity at each stage of growth. In this experiment, we determined that for low vegetation, the C band will be more useful at the beginning of the growth cycle, while the L band provides more information in later stages of growth. Using a random forest regression model combining SAR parameters with common difference vegetation index (DVI), we improved the error by 50% in comparison to the error using the (DVI) for corn, soybean, and rice.
Why it matches plant phenotyping methodsSAR・光学リモートセンシングを用いた圃場収量推定手法を比較・評価し、センサー特徴量と回帰モデルの性能を検証しており、植物形質（収量）の取得・推定が中心です。
abstractWe assess the efficiency of dual- (Sentinel-1) and quad-polarimetric (UAVSAR, RADARSAT-2) data to explain inter-field crop yield variability for corn, soybean, and rice over a test area in Arkansas, US (258 fields, 2019).
Yield forecasting and within-field yield variation is essential information that helps farmers develop sustainable agriculture. However, such information still needs to be included for most of them, and remote sensing is an alternative to provide it. Our objective was to assess Random Forest regression models composed of unique GLCM texture measures as an alternative to usual empirical models that use spectral response and auxiliary data, which is complex and reaches varied results. Eleven GLCM texture models based on eight texture measures of a single spectral layer were assessed to represent soybean field yield variation in two sites and seasons. Several models achieved satisfactory results, reaching R² from 0.90 to 0.95 and RMSE from 0.06 to 0.26 t/ha. Models above 15-window size are recommended for the soybean yield prediction as window size is an essential attribute to GLCM performance. Models derived from the bands individually (red, red-edge, near-infrared, and short wavelength infrared) were more sensitive to the window size than those derived from vegetation indices (EVI, GNDVI, GRNDVI, NDMI, NDRE, NDVI, SFDVI). The data aggregated by texture measures improve the individual spectral responses, providing alternatives to predict soybean within-field yield variation using random forest models.
Why it matches plant phenotyping methods衛星画像のGLCMテクスチャ特徴量とRandom Forestを用いて、圃場内のダイズ収量変動を推定する手法を評価・比較しており、植物形質（収量）の取得・推定が研究の中心である。
abstractOur objective was to assess Random Forest regression models composed of unique GLCM texture measures as an alternative to usual empirical models
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 6 Sept 2026
[L.] Merr.) production is susceptible to biotic and abiotic stresses, exacerbated by extreme weather events. Water limiting stress, that is, drought, emerges as a significant risk for soybean production, underscoring the need for advancements in stress monitoring for crop breeding and production. This project combined multi-modal information to identify the most effective and efficient automated methods to study drought response. We investigated a set of diverse soybean accessions using multiple sensors in a time series high-throughput phenotyping manner to: (1) develop a pipeline for rapid classification of soybean drought stress symptoms, and (2) investigate methods for early detection of drought stress. We utilized high-throughput time-series phenotyping using unmanned aerial vehicles and sensors in conjunction with machine learning analytics, which offered a swift and efficient means of phenotyping. The visible bands were most effective in classifying the severity of canopy wilting stress after symptom emergence. Non-visual bands in the near-infrared region and short-wave infrared region contribute to the differentiation of susceptible and tolerant soybean accessions prior to visual symptom development. We report pre-visual detection of soybean wilting using a combination of different vegetation indices and spectral bands, especially in the red-edge. These results can contribute to early stress detection methodologies and rapid classification of drought responses for breeding and production applications.
Why it matches plant phenotyping methodsUAVと複数センサー、時系列測定、機械学習を統合した乾燥ストレス表現型の分類・早期検出パイプラインが研究の中心であり、技術的な表現型取得・抽出方法を扱っている。
abstract(1) develop a pipeline for rapid classification of soybean drought stress symptoms, and (2) investigate methods for early detection of drought stress.
SoybeanX-ray / CTRootMorphology / geometry measurementRoot system architecture
Typically, root system architecture (RSA) is not visible, and realistically, high-throughput methods for RSA trait phenotyping should capture key features of developing root systems in solid substrates in 3D. In a published 2-D study using thin rhizoboxes, vermiculite as a growing medium, and photography for imaging, triplicates of 137 soybean cultivars were phenotyped for their RSA. In the transition to 3-D work using X-ray computed tomography (CT) scanning and mineral soil, two research questions are addressed: (1) how different is the soybean RSA characterization between the two phenotyping systems; and (2) is a direct comparison of the results reliable? Prior to a full-scale study in 3D, we grew, in pots filled with sand, triplicates of the Casino and OAC Woodstock cultivars that had shown the most contrasting RSAs in the 2-D study, and CT scanned them at the V1 vegetative stage of development of the shoots. Differences between soybean cultivars in RSA traits, such as total root length and fractal dimension (FD), observed in 2D, can change in 3D. In particular, in 2D, the mean FD values are 1.48 ± 0.16 (OAC Woodstock) vs. 1.31 ± 0.16 (Casino), whereas in 3D, they are 1.52 ± 0.14 (OAC Woodstock) vs. 1.24 ± 0.13 (Casino), indicating variations in RSA complexity.
Why it matches plant phenotyping methods2D写真法と3D X線CT法という根系表現型計測システムを比較・評価し、RSA形質の測定結果の信頼性を検討しているため、フェノタイピング手法が中心である。
abstracthigh-throughput methods for RSA trait phenotyping should capture key features of developing root systems in solid substrates in 3D
Reproduction assets foundThe paper's own supplement (MDPI S1) contains two videos produced in MATLAB from skeletal 3-D images of the root systems reconstructed from this study's CT scanning data — paper-specific phenotyping outputs made publicly available. The figshare links are explicitly described as 'soybean genomic data' from the prior GWASupplement · publicTwo videos (.AVI files), one per soybean cultivar, were produced in MATLAB (MathWorks, Natick, MA, USA) from skeletal 3-D images of the root systems, and are made available as a supplement to the graphical results presented for the 3-D phenotyping system in Figure 2 in the manuscript.Open asset ↗lines:100-116Plant phenotyping relevance match · UnverifiedEurope PMC · checked 6 Sept 2026
Accurate identification of soybean leaf diseases is essential to improving quality and yield. Aiming at the problem of insufficient data volume that may lead to model overfitting and low recognition ability, this paper proposes a hypergraph cell membrane computing network model for soybean disease identification (HcmcNet). The main components of HcmcNet are the pyramid convolutional feature extraction membrane, the ordinary feature extraction membrane, the U-type feature extraction membrane, and the dynamic attention membrane. The three parallel feature extraction membranes are designed to improve the model's ability to capture disease features. The dynamic attention membrane aims to enhance the model's expressiveness and performance by dynamically adjusting the attentional weights of the three feature extraction membranes to fuse the disease features effectively. Soybean leaf disease images were used to create the dataset and conduct experiments. The experimental results show that HcmcNet achieves 98% accuracy on the test set. Compared with classical models, HcmcNet shows obvious advantages in several evaluation metrics. We also conducted experiments on public datasets. The results show that it is feasible to use HcmcNet for soybean leaf disease recognition, and HcmcNet has higher classification accuracy and stronger generalization ability on small sample datasets. HcmcNet has great application prospects in soybean leaf disease recognition.
Why it matches plant phenotyping methods大豆葉の病害状態を画像から認識する深層学習モデルを開発・評価しており、植物病害表現型の取得・抽出手法が中心である。
abstractthis paper proposes a hypergraph cell membrane computing network model for soybean disease identification (HcmcNet).
Background Climate change and the growing demand for agricultural water threaten global food security. Understanding water use characteristics of major crops from leaf to field scale is critical, particularly for identifying crop varieties with enhanced water-use efficiency (WUE) and stress tolerance. Traditional methods to assess WUE are either by gas exchange measurements at the leaf level or labor-intensive manual pot weighing at the whole-plant level, both of which have limited throughput. Results Here, we developed a microcontroller-based low-cost system that integrates pot weighing, automated water supply, and real-time monitoring of plant water consumption via Wi-Fi. We validated the system using major crops (rice soybean, maize) under diverse stress conditions (salt, waterlogging, drought). Salt-tolerant rice maintained higher water consumption and growth under salinity than salt-sensitive rice. Waterlogged soybean exhibited reduced water use and growth. Long-term experiments revealed significant WUE differences between rice varieties and morphological adaptations represented by altered shoot-to-root ratios under constant drought conditions in maize. Conclusions We demonstrate that the system can be used for varietal differences between major crops in their response to drought, waterlogging, and salinity stress. This system enables high-throughput, long-term evaluation of water use characteristics, facilitating the selection and development of water-saving and stress-tolerant crop varieties.
Why it matches plant phenotyping methods作物の水消費量・WUE・成長を高スループットかつ長期的に測定するマイクロコントローラ基盤システムを開発・検証しており、植物表現型取得法が研究の中心である。
abstractwe developed a microcontroller-based low-cost system that integrates pot weighing, automated water supply, and real-time monitoring of plant water consumption via Wi-Fi.
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 15 Sept 2026
In soybean (Glycine max ), limiting whole-plant transpiration rate (TR) response to increasing vapor pressure deficit (VPD) has been associated with the 'slow-wilting' phenotype and with water-conservation enabling higher yields under terminal drought. Despite the promise of this trait, it is still unknown whether it has a genetic basis in soybean, a challenge limiting the prospects of breeding climate-resilient varieties. Here, we present the results of a first attempt at a high-throughput phenotyping of TR and stomatal conductance response curves to increasing VPD conducted on a soybean mapping population consisting of 140 recombinant inbred lines (RIL). This effort was conducted over two consecutive years, using a controlled-environment, gravimetric phenotyping platform that enabled characterizing 900 plants for these responses, yielding regression parameters (R 2 from 0.92 to 0.99) that were used for genetic mapping. Several quantitative trait loci (QTL) were identified for these parameters on chromosomes (Ch) 4, 6, and 10, including a VPD-conditional QTL on Ch 4 and a 'constitutive' QTL controlling all parameters on Ch 6. This study demonstrated for the first time that canopy water use in response to rising VPD has a genetic basis in soybean, opening novel avenues for identifying alleles enabling water conservation under current and future climate scenarios.
Why it matches plant phenotyping methods高スループットの重力測定フェノタイピングプラットフォームを用いて、蒸散・気孔コンダクタンス応答曲線を技術的に取得し、遺伝解析に利用しており、表現型取得法が研究の中心である。
abstractHere, we present the results of a first attempt at a high-throughput phenotyping of TR and stomatal conductance response curves to increasing VPD
The use of high-altitude remote sensing (RS) data from aerial and satellite platforms presents considerable challenges for agricultural monitoring and crop yield estimation due to the presence of noise caused by atmospheric interference, sensor anomalies, and outlier pixel values. This paper introduces a "Quartile Clean Image" pre-processing technique to address these data issues by analyzing quartile pixel values in local neighborhoods to identify and adjust outliers. Applying this technique to 20,946 Moderate Resolution Imaging Spectroradiometer (MODIS) images from 2002 to 2015, improved the mean peak signal-to-noise ratio (PSNR) to 40.91 dB. Integrating Quartile Clean data with Convolutional Neural Networks (CNN) models with exponential decay learning rate scheduling achieved RMSE improvements up to 5.88% for soybeans and 21.85% for corn, while Long Short-Term Memory (LSTM) models demonstrated RMSE reductions up to 11.52% for soybeans and 29.92% for corn using exponential decay learning rates. To compare the proposed method with state-of-the-art technique, we introduce the Vision Transformer (ViT) model for crop yield estimation. The ViT model, applied to the same dataset, achieves remarkable performance without explicit pre-processing, with R2 scores ranging from 0.9752 to 0.9875 for soybean and 0.9540 to 0.9888 for corn yield estimation. The RMSE values range from 7.75086 to 9.76838 for soybean and 26.25265 to 34.20382 for corn, demonstrating the ViT model's robustness. This research contributes by (1) introducing the Quartile Clean Image method for enhancing RS data quality and improving crop yield estimation accuracy, and (2) comparing it with the state-of-the-art ViT model. The results demonstrate the effectiveness of the proposed approach and highlight the potential of the ViT model for crop yield estimation, representing a valuable advancement in processing high-altitude imagery for precision agriculture applications.
Why it matches plant phenotyping methods作物収量という植物形質の推定を対象に、画像前処理法とVision Transformer等の推定手法を中心的に開発・比較しているため、植物フェノタイピング手法研究に該当する。
abstractThis paper introduces a "Quartile Clean Image" pre-processing technique to address these data issues by analyzing quartile pixel values in local neighborhoods to identify and adjust outliers.
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:1-25Plant phenotyping relevance match · UnverifiedEurope PMC · checked 7 Sept 2026
A transcription-aided selection (TAS) strategy is proposed in this paper, which utilizes the positive regulatory roles of genes involved in the plant immunity pathways to screen crops with high disease resistance. Increased evidence has demonstrated that upon pathogen attack, the expression of diverse genes involved in salicylic acid (SA)-mediated SAR are differentially expressed and transcriptionally regulated. The paper discusses the molecular mechanisms of the SA signaling pathway, which plays a central role in plant immunity, and identifies differentially expressed genes (DEGs) that could be targeted for transcriptional detection. We have conducted a series of experiments to test the TAS strategy and found that the level of GmSAGT1 expression is highly correlated with soybean downy mildew (SDM) resistance with a correlation coefficient R 2 = 0.7981. Using RT-PCR, we screened 2501 soybean germplasms and selected 26 collections with higher levels of both GmSAGT1 and GmPR1 (Pathogenesis-related proteins1) gene expression. Twenty-three out of the twenty-six lines were inoculated with Peronospora manshurica (Pm) in a greenhouse. Eight showed HR (highly resistant), four were R (resistant), five were MR (moderately resistant), three were S (susceptible), and three were HS (highly susceptible). The correlation coefficient R 2 between the TAS result and Pm inoculation results was 0.7035, indicating a satisfactory consistency. The authors anticipate that TAS provides an effective strategy for screening crops with broad-spectrum and long-lasting resistance.
Why it matches plant phenotyping methods遺伝子発現を用いて作物の病害抵抗性を推定・選抜するTAS戦略を提案し、RT-PCRスクリーニングと接種試験で相関・妥当性を検証している。分子測定だが、抵抗性という植物状態の推定法が研究の中心である。
abstractA transcription-aided selection (TAS) strategy is proposed in this paper, which utilizes the positive regulatory roles of genes involved in the plant immunity pathways to screen crops with high disease resistance.
Developments in genomics and phenomics have provided valuable tools for use in cultivar development. Genomic prediction (GP) has been used in commercial soybean [Glycine max L. (Merr.)] breeding programs to predict grain yield and seed composition traits. Phenomic prediction (PP) is a rapidly developing field that holds the potential to be used for the selection of genotypes early in the growing season. The objectives of this study were to compare the use and performance of GP and PP for predicting soybean seed yield, protein content, and oil content. We additionally conducted Genome Wide Association Studies (GWAS) to identify significant SNPs associated with the traits of interest. These SNPs were also used to train the GP models. The GWAS panel of 292 diverse accessions was grown in six environments in replicated trials. Spectral data were collected at three timepoints during the growing season. A GBLUP model was trained on 268 accessions, while three separate machine learning (ML) models were trained on vegetation indices (VIs) and canopy traits. We observed that for PP, Random Forest (RF) algorithm had the highest rank correlation between the predicted and the actual phenotype rank. PP had a higher correlation coefficient than GP for seed yield, while GP had higher correlation coefficients for seed protein and oil contents. VIs with high feature importance were used as covariates in a new GBLUP model, and a new RF model was trained with the inclusion of selected SNPs from the GWAS results. These models did not outperform the original GP and PP models. These results show the capability of using ML for in-season predictions for specific traits in soybean breeding and provide insights on PP and GP inclusions in breeding programs.
Why it matches plant phenotyping methodsスペクトルデータ、植生指数、キャノピー形質、機械学習を用いたフェノミック予測を中心に、収量・種子成分の予測性能をゲノム予測と比較しているため、植物表現型取得・推定手法の実質的な評価に該当する。
abstractThe objectives of this study were to compare the use and performance of GP and PP for predicting soybean seed yield, protein content, and oil content.
Abstract Early detection of nutrient deficiencies is crucial for optimizing crop yields and ensuring sustainable agricultural practices. This study presents a novel application of the YOLOv8s object detection model for identifying nitrogen, phosphorus, and potassium deficiencies in soybean plants. Employing a unique dataset from a long-term nutrient-deficient field maintained for over 40 years, we trained and evaluated the model on 6,020 red, green, and blue images of soybean leaves exhibiting nutrient stress conditions. The YOLOv8s model achieved exceptional performance, with a mean average precision (mAP@0.5) of 99.18% during training and 98.51% for validation. Precision rates for individual nutrient deficiencies ranged from 90.03–96.54%, with highly accurate potassium deficiency detection. The model demonstrated robust generalization across diverse field conditions, processing images in 3.46 ms each, making it suitable for real-time applications. This research significantly advances the field of precision agriculture by providing a fast, accurate, and scalable method for detecting early nutrient deficiency in soybean crops, potentially revolutionizing fertilizer management practices and contributing to more sustainable farming systems.
Why it matches plant phenotyping methods大豆葉の栄養欠乏状態を画像とYOLOv8sで直接推定する手法を開発・評価しており、植物状態の取得が研究の中心である。
abstractThis study presents a novel application of the YOLOv8s object detection model for identifying nitrogen, phosphorus, and potassium deficiencies in soybean plants.
X-ray fluorescence (XRF) analyses are fast, clean, non-destructive, and compatible with on-field operations, which are some advantages over traditional determinations using coupled plasma optical emission spectroscopy (ICP-OES). The aim of this study was to advance in situ XRF approaches for assessing the nutritional status of soybean leaves (i.e., P, S, K, Ca, Mn, Fe, Cu and Zn). More specifically, we propose a protocol to ensure accuracy of in-field analysis and then evaluate the predictive performance of XRF via different data modelling strategies for macro- and micronutrient determination. Therefore, the XRF sensor dwell time of 60 s and the maximum time of 5 min were determined for the analysis of the leaves after leaf abscission, taking into account the influence of moisture loss on the signal intensity of the lighter elements. Regarding the predictive performance of XRF data for nutrients determination, multiple linear regression (MLR) models resulted in lower root mean square errors (RMSE) for P (433 mg kg⁻¹), S (204 mg kg⁻¹) and K (1957 mg kg⁻¹); Partial least squares regression (PLS) for Ca (519 mg kg⁻¹); and simple linear regression (SLR) for Mn (9 mg kg⁻¹), Fe (18 mg kg⁻¹), Zn (5 mg kg⁻¹). The different modelling strategies exhibited equivalent RMSE for Cu (2 mg kg⁻¹). These prediction errors are within a ±20% range, demonstrating that the in situ protocols developed in this research are useful for predicting the nutrients concentration in soybean leaves. Our study shows the possibility of using the in situ XRF sensor for the rapid and practical nutrients determination in soybean leaves, presenting good potential as a crop diagnosis tool.
Why it matches plant phenotyping methods大豆葉の栄養状態をXRFセンサーで非破壊測定・予測するプロトコルとモデル性能を開発・評価しており、植物形質取得法が研究の中心である。
abstractThe aim of this study was to advance in situ XRF approaches for assessing the nutritional status of soybean leaves
Plant phenotyping relevance match · UnverifiedOpenAlex · Crossref · Europe PMC · checked 7 Sept 2026
Building models that allow phenotypic evaluation of complex agronomic traits in crops of global economic interest, such as grain yield (GY) in soybean and maize, is essential for improving the efficiency of breeding programs. In this sense, understanding the relationships between agronomic variables and those obtained by high-throughput phenotyping (HTP) is crucial to this goal. Our hypothesis is that vegetation indices (VIs) obtained from HTP can be used to indirectly measure agronomic variables in annual crops. The objectives were to study the association between agronomic variables in maize and soybean genotypes with VIs obtained from remote sensing and to identify computational intelligence for predicting GY of these crops from VIs as input in the models. Comparative trials were carried out with 30 maize genotypes in the 2020/2021, 2021/2022 and 2022/2023 crop seasons, and with 32 soybean genotypes in the 2021/2022 and 2022/2023 seasons. In all trials, an overflight was performed at R1 stage using the UAV Sensefly eBee equipped with a multispectral sensor for acquiring canopy reflectance in the green (550 nm), red (660 nm), near-infrared (735 nm) and infrared (790 nm) wavelengths, which were used to calculate the VIs assessed. Agronomic traits evaluated in maize crop were: leaf nitrogen content, plant height, first ear insertion height, and GY, while agronomic traits evaluated in soybean were: days to maturity, plant height, first pod insertion height, and GY. The association between the variables were expressed by a correlation network, and to identify which indices are best associated with each of the traits evaluated, a path analysis was performed. Lastly, VIs with a cause-and-effect association on each variable in maize and soybean trials were adopted as independent explanatory variables in multiple regression model (MLR) and artificial neural network (ANN), in which the 10 best topologies able to simultaneously predict all the agronomic variables evaluated in each crop were selected. Our findings reveal that VIs can be used to predict agronomic variables in maize and soybean. Soil-adjusted Vegetation Index (SAVI) and Green Normalized Dif-ference Vegetation Index (GNDVI) have a positive and high direct effect on all agronomic variables evaluated in maize, while Normalized Difference Vegetation Index (NDVI) and Normalized Difference Red Edge Index (NDRE) have a positive cause-and-effect association with all soybean variables. ANN outperformed MLR, providing higher accuracy when predicting agronomic variables using the VIs select by path analysis as input. Future studies should evaluate other plant traits, such as physiological or nutritional ones, as well as different spectral variables from those evaluated here, with a view to contributing to an in-depth understanding about cause-and-effect relationships between plant traits and spectral variables. Such studies could contribute to more specific HTP at the level of traits of interest in each crop, helping to develop genetic materials that meet the future demands of population growth and climate change.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像から植生指数を抽出し、作物の収量・形態などの表現型を予測するHTP解析手法が研究の中心である。
titleHigh-throughput phenotyping in maize and soybean genotypes using vegetation indices and computational intelligence.
• Breeding for drought tolerance is becoming a necessity for most of the main crops, including soybean. • Phenotyping for physiological traits is considered unfeasible in breeding strategy focuses on abiotic stress tolerance. • Spectroscopy data can be used for high-throughput phenotyping methodology for hard-to-measure traits. • PLSR models successfully predicted physiological parameters at leaf-level. • Hyperspectral data can be used as a selection methodology for physiological traits in soybean cultivars under drought conditions. Understanding cultivars' physiological traits variations under abiotic stresses is critical to improve phenotyping and selections of resistant crop varieties. Traditional methods of accessing physiological traits in plants are costly and time consuming, which prevents their use in breeding programs. Spectroscopy data and statistical approaches such as partial least square regression could be applied to rapidly collect and predict several physiological parameters at leaf-level, allowing phenotyping several genotypes in a high-throughput manner. We collected spectroscopy data of twenty soybean cultivars planted under well-watered and drought conditions during the reproductive phase. At 20 days after drought was imposed, we measured leaf pigments content (chlorophyll a and b, and carotenoids), specific leaf area, electrons transfer rate, and photosynthetic active radiation. At 28 days after drought imposition, we measured leaf pigments content, specific leaf area, relative water content, and leaf temperature. Partial least square regression models accurately predicted leaf pigments content, specific leaf area, and leaf temperature (cross-validation R 2 ranging from 0.56 to 0.84). Discriminant analysis using 54 wavelengths was able to select the best-performance cultivars regarding all evaluated physiological traits. We showed the great potential of using spectroscopy as a feasible, non-destructive, and accurate method to estimate physiological traits and screening of superior genotypes.
Why it matches plant phenotyping methods葉スペクトロスコピーとPLSRを用いて生理形質を非破壊・高スループットに推定する手法を開発・適用し、予測性能も評価しているため、フェノタイピング手法が中心である。
abstractSpectroscopy data can be used for high-throughput phenotyping methodology for hard-to-measure traits.