Accurate estimation of wheat spike volume is important for yield component analysis and stress resilience assessment, yet field-based measurement remains challenging. Active 3D sensing methods such as Light Detection and Ranging (LiDAR) or time-of-flight (ToF) are sensitive to plant motion or poorly suited to outdoor conditions, while 3D reconstructions are computationally expensive. Direct 2D image processing would offer computational advantages, but image-based models lack explicit geometric information. We therefore propose a hybrid 2D-3D approach with knowledge distillation during training while enabling efficient image-only inference. First, we train a rigid-invariant point cloud network using distance-based histogram features to obtain pose-robust geometric representations. We then combine the 3D model with a proposed multi-view image-based regulated Transformer (RT) in an ensemble architecture. Finally, we distill the ensemble knowledge into a purely image-based student model using either feature-based or label-based distillation. The two distilled RTs reduce the mean absolute error (MAE) from 654.31 mm$^3$ of the non-distilled RT to 639.93 mm$^3$ and 644.62 mm$^3$, and increase correlation from 0.76 to 0.77 and 0.82, respectively. At the same time, inference time is reduced from 160 ms to 1.4 ms per spike. Distillation further mitigates volume-dependent bias and reshapes the latent representation of the image model toward a geometry-aware shape. Our results demonstrate that 3D-informed training of a 2D Transformer allows for scalable and efficient spike volume estimation for high-throughput field phenotyping.
Why it matches plant phenotyping methods小麦穂の体積を画像・3D再構成・知識蒸留で推定する手法の開発と性能評価が中心であり、高スループット植物フェノタイピングへの応用も明示されている。
abstractWe therefore propose a hybrid 2D-3D approach with knowledge distillation during training while enabling efficient image-only inference.
Reproduction assets foundThe paper explicitly states that links to its wheat spike dataset (multi-view images and 3D scans) and its analysis code are available via the authors' project webpage, which is an allowed URL. Other URLs (pyrender, CORDIS projects) are generic libraries or unrelated funding projects, not paper-specific assets.Dataset · publictance of around 2.5 m with a ground sampling distance of 0.3 mm (Fig. S1 a). The tagged and imaged spikes (Fig. S1 b) were sampled and ground truth volumes were acquired with a 3D light scanner (Shining 3D Einscan-SE V2, SHINING3D, Hangzhou, China) following the protocol of [ 76 ] . Links to the dataset and code can be found at https://oliviazum.github.io/3DKD-wheat/ . Detailed information about the dataset can be found in Sec. A .
3.3 Data Pre-Processing
Field images contained approximately 300-500 spikes per genotype within a plot of about 1.5 m 2 m^{2} . To reduce background inference, spike detection was first performed, and all subsequent processing was restricted to the detected regioOpen asset ↗lines:91-104Code / dataset availability confirmedEurope PMC · bioRxiv · checked 15 Sept 2026
1 Summary Zinc (Zn) deficiency limits rice productivity and poses a risk to human health, particularly in populations reliant on rice-based diets. Although rice germplasm exhibits wide variation in Zn-deficiency tolerance, the underlying physiological mechanisms remain poorly resolved. Evidence across the literature for Zn-deficiency–induced secretion of 2′-deoxymugineic acid (DMA) is inconsistent. This study clarifies the role of DMA secretion as a Zn-deficiency stress response. We developed and validated a sensitive LC–ESI–Q–TOF–MS method for selective detection of DMA in rice root exudates. Five rice genotypes with contrasting Zn-deficiency tolerance were grown hydroponically and DMA secretion measured. Zn-deficiency increased DMA exudation across all genotypes, with sensitive genotypes also showing higher secretion compared with control, supporting DMA’s role as a general response to Zn stress rather than being restricted to efficient genotypes. Fold-change responses exceeded previous studies, likely due to more severe stress exposure. Our results confirm that DMA secretion is induced under Zn-deficiency in rice as part of the micronutrient stress response. However, the lack of increased Zn uptake indicates that additional tolerance mechanisms are involved. These findings reconcile inconsistencies in the literature and position DMA secretion as an important, but not exclusive, component of Zn-deficiency adaptation in rice.
Why it matches plant phenotyping methodsイネ根滲出液中のDMAを選択的に検出するLC–MS法を開発・検証し、亜鉛欠乏応答という植物生理状態を測定しているため、化学分析が単なる付随測定ではなく中心的な方法貢献である。
abstractWe developed and validated a sensitive LC–ESI–Q–TOF–MS method for selective detection of DMA in rice root exudates.
Reproduction assets foundThe paper's Data availability statement points to a public Zenodo deposit containing the datasets generated and analysed in this study (DMA exudation and Zn uptake measurements in rice).Dataset · publicthe experiments, developed the
525
methods and analysed the results. The experimental data were collected by C.R. assisted by
526
G.L.M., C.T. and D.J.W. Data analysis and writing of paper by all authors.
527
528
Data availability
529
The data sets generated and/or analysed during the current study are available on Zenodo,
530
https://zenodo.org/uploads/18184803
531
532
533
.
CC-BY 4.0 International license
perpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for this
this version posted March 18, 2026.
;
https://doi.org/10.64898/2026.03.16.71158Open asset ↗Zenodo · 18184803pdf-raw-page:21 lines:1-47Code / dataset availability confirmedOpenAlex · Europe PMC · checked 5 Sept 2026
Efficient large-scale 3D reconstruction of orchard environments is essential for robotic inspection and precision agriculture, yet existing methods struggle with unstructured scenes, variable illumination, and computational bottlenecks. We propose InspectGaussian, a coarse-to-fine Gaussian reconstruction framework tailored for orchard inspection robots. The pipeline integrates an RGB-D-based data acquisition strategy using ORB-SLAM3, which is enhanced by a dense mapping module for robust large-scale pose estimation and point cloud generation. A divide-and-conquer strategy is then employed: individual plant views are extracted via a YOLO-World-based detection and 3D matching algorithm, followed by plant-specific reconstruction using an improved 3D Gaussian Splatting (3DGS) method incorporating depth regularization and region-aware refinement. Experimental results in citrus orchards demonstrate that InspectGaussian achieves 96% average precision and 93% recall in plant view extraction, while surpassing state-of-the-art methods in reconstruction fidelity (31.226 PSNR, 0.915 SSIM, 0.067 LPIPS) and point cloud accuracy (7 mm error). These results confirm its effectiveness in capturing fine structural and textural details while maintaining scalability and efficiency. This framework provides a practical solution for high-throughput, in-field plant phenotyping and lays the foundation for intelligent orchard monitoring and management.
Why it matches plant phenotyping methods植物個体の3D再構成とRGB-D・検出・Gaussian Splattingを統合した手法開発であり、植物の構造的形質取得を目的とするため、フェノタイピング手法が中心である。
abstractWe propose InspectGaussian, a coarse-to-fine Gaussian reconstruction framework tailored for orchard inspection robots.
Reproduction assets foundThe paper's authors explicitly state their analysis code is publicly available on GitHub. Phenotype datasets (RGB-D orchard image sequences, LiDAR point clouds, manual trait measurements) are only available upon request, so they do not qualify as public assets.Code · publicOur code are available at https://github.com/zlhzau/InspectGaussian.git .Open asset ↗zlhzau/InspectGaussianlines:489-515Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
This data article describes a curated RGB-Depth image dataset captured using an Intel RealSense D435 stereo depth camera mounted on an autonomous mobile platform during field deployments at commercial baby broccoli farms in Victoria, Australia. The dataset comprises 1759 paired RGB images (640 × 480 pixels) and corresponding 16-bit depth frames acquired under both daytime (natural sunlight) and night-time (LED illumination) conditions, designed to support research in agricultural computer vision and robotic harvesting. Images were selected from 39,765 raw acquisitions through a reproducible Python curation pipeline applying quality filtering (blur detection, brightness thresholds, corruption detection), perceptual hash-based duplicate removal, and manual review. The final dataset includes 924 daytime and 835 night-time image pairs containing baby broccoli plants at various growth stages. The dataset provides RGB camera intrinsic parameters and pixel-aligned depth maps to enable 3D point cloud reconstruction. Potential applications include developing deep learning models for crop detection and segmentation, validating depth-based size estimation methods, and benchmarking illumination-robust vision systems. All data and curation code are publicly available under a CC BY 4.0 license.
Why it matches plant phenotyping methodsRGB-Depth画像データセットの構築と再現可能なキュレーションを中心とし、作物検出に加えてサイズ推定という植物形質の評価・ベンチマークに利用できるため。
titleA field-acquired RGB-Depth image dataset for computer vision-based baby broccoli detection and size estimation under varying illumination conditions.
Reproduction assets foundThe paper is a data article describing a public Mendeley Data repository containing the authors' field-acquired RGB-D baby broccoli image dataset (1759 image pairs, ground truth diameter annotations, camera intrinsics, and curation/annotation code), directly reproducing the paper's phenotyping measurements and analysisDataset · publicRepository name: Mendeley Data
Data identification number: 10.17632/px5p6zdk6k.3
Direct URL to data: https://data.mendeley.com/datasets/px5p6zdk6k/3Open asset ↗Mendeley Data · 10.17632/px5p6zdk6k.3html-lines:95-155Code / dataset availability confirmedarXiv · OpenAlex · checked 13 Sept 2026
Potato yield is a key indicator for optimizing cultivation practices in agriculture. Potato yield can be estimated on harvesters using RGB-D cameras, which capture three-dimensional (3D) information of individual tubers moving along the conveyor belt. However, point clouds reconstructed from RGB-D images are incomplete due to self-occlusion, leading to systematic underestimation of tuber weight. To address this, we introduce PointRAFT, a high-throughput point cloud regression network that directly predicts continuous 3D shape properties, such as tuber weight, from partial point clouds. Rather than reconstructing full 3D geometry, PointRAFT infers target values directly from raw 3D data. Its key architectural novelty is an object height embedding that incorporates tuber height as an additional geometric cue, improving weight prediction under practical harvesting conditions. PointRAFT was trained and evaluated on 26,688 partial point clouds collected from 859 potato tubers across four cultivars and three growing seasons on an operational harvester in Japan. On a test set of 5,254 point clouds from 172 tubers, PointRAFT achieved a mean absolute error of 12.0 g and a root mean squared error of 17.2 g, substantially outperforming a linear regression baseline and a standard PointNet++ regression network. With an average inference time of 6.3 ms per point cloud, PointRAFT supports processing rates of up to 150 tubers per second, meeting the high-throughput requirements of commercial potato harvesters. Beyond potato weight estimation, PointRAFT provides a versatile regression network applicable to a wide range of 3D phenotyping and robotic perception tasks. The code, network weights, and a subset of the dataset are publicly available at https://github.com/pieterblok/pointraft.git.
Why it matches plant phenotyping methods部分点群からジャガイモ塊茎重量を推定する3D深層学習手法を開発・評価しており、植物形質取得が研究の中心である。
abstractwe introduce PointRAFT, a high-throughput point cloud regression network that directly predicts continuous 3D shape properties, such as tuber weight, from partial point clouds.
Reproduction assets foundThe paper publicly releases its authors' analysis code and trained network weights on GitHub, and a subset of its potato tuber partial point cloud dataset (with ground truth weights) on Hugging Face. Both are paper-specific, public, and actionable.Code · publicThe code, network weights, and a subset of the dataset are publicly available at https://github.com/pieterblok/pointraft.git .Open asset ↗pieterblok/pointraftlines:1-93Dataset · publicA subset of the datasets generated and/or analyzed during this study is publicly available at: https://huggingface.co/datasets/UTokyo-FieldPhenomics-Lab/3DPotatoTwinOpen asset ↗UTokyo-FieldPhenomics-Lab/3DPotatoTwinlines:447-463Code / dataset availability confirmedOpenAlex · Europe PMC · checked 6 Sept 2026
Tomato is a globally significant horticultural crop with substantial economic and nutritional value. High-precision phenotypic analysis of tomato fruit characteristics, enabled by computer vision and image-based phenotyping technologies, is essential for varietal selection and automated quality evaluation. An intelligent detection framework for phenomics analysis of tomato fruits was developed in this study, which combines image processing techniques with deep learning algorithms to automate the extraction and quantitative analysis of 12 phenotypic traits, including fruit morphology, structure, color and so on. First, a dataset of tomato fruit section images was developed using a depth camera. Second, the SegFormer model was improved by incorporating the MLLA linear attention mechanism, and a lightweight SegFormer-MLLA model for tomato fruit phenotype segmentation was proposed. Accurate segmentation of tomato fruit stem scars and locular structures was achieved, with significantly reduced computational cost by the proposed model. Finally, a Hybrid Depth Regression Model was designed to optimize the estimation of optimal depth. By fusing RGB and depth information, the framework enabled efficient detection of key phenotypic traits, including fruit longitudinal diameter, transverse diameter, mesocarp thickness, and depth and width of stem scar. Experimental results demonstrated a high correlation between the phenotypic parameters detected by the proposed model and the manually measured values, effectively validating the accuracy and feasibility of the model. Hence, we developed an equipment automatically phenotyping tomato fruits and the corresponding software system, providing reliable data support for precision tomato breeding and intelligent cultivation, as well as a reference methodology for phenotyping other fruit crops.
Why it matches plant phenotyping methods深度カメラ、画像処理、深層学習を統合し、トマト果実の12形質を自動抽出・定量する装置とソフトウェアを開発しており、表現型取得法が研究の中心である。
abstractAn intelligent detection framework for phenomics analysis of tomato fruits was developed in this study, which combines image processing techniques with deep learning algorithms to automate the extraction and quantitative analysis of 12 phenotypic traits
Reproduction assets foundThe paper's Data Availability Statement explicitly deposits some datasets, model weights, and code used in the study at a public GitHub repository, which is paper-specific and actionable. Full self-developed datasets require contacting the corresponding author.Code · publicSome datasets, model weights, and code used in the present study are available at https://github.com/Snail-code-wq/Plants_Tomato_2025 (accessed on 5 November 2025). All self-developed datasets can be obtained by contacting the corresponding author.Open asset ↗Snail-code-wq/Plants_Tomato_2025lines:466-479Code / dataset availability confirmedarXiv · OpenAlex · checked 15 Sept 2026
Lychee is a high-value subtropical fruit. The adoption of vision-based harvesting robots can significantly improve productivity while reduce reliance on labor. High-quality data are essential for developing such harvesting robots. However, there are currently no consistently and comprehensively annotated open-source lychee datasets featuring fruits in natural growing environments. To address this, we constructed a dataset to facilitate lychee detection and maturity classification. Color (RGB) images were acquired under diverse weather conditions, and at different times of the day, across multiple lychee varieties, such as Nuomici, Feizixiao, Heiye, and Huaizhi. The dataset encompasses three different ripeness stages and contains 11,414 images, consisting of 878 raw RGB images, 8,780 augmented RGB images, and 1,756 depth images. The images are annotated with 9,658 pairs of lables for lychee detection and maturity classification. To improve annotation consistency, three individuals independently labeled the data, and their results were then aggregated and verified by a fourth reviewer. Detailed statistical analyses were done to examine the dataset. Finally, we performed experiments using three representative deep learning models to evaluate the dataset. It is publicly available for academic
Why it matches plant phenotyping methodsライチ果実の成熟段階という植物器官の状態をRGB-D画像から分類するデータセットを構築し、アノテーション検証と深層学習モデル評価を行っており、表現型取得・評価手法が中心である。
abstractwe constructed a dataset to facilitate lychee detection and maturity classification.
Reproduction assets foundThe authors publicly release the paper's lychee RGB-D image dataset (raw/augmented RGB images, depth maps, detection and maturity annotations) and the Python scripts for data augmentation, image similarity comparison, and annotation in the same GitHub repository.Dataset · publicchees, the non-augmented models
produced misclassifications with lower recognition and accuracy, whereas the augmented models
avoided these issues. Overall, the results demonstrate that the data augmentation method effectively
improves the comprehensive performance of the models.
5. Data Availability
The dataset is available at：https://github.com/SeiriosLab/Lychee. The Python scripts for data
augmentation, image similarity comparison, and annotation are available within the same
repository under the tree/main/script directory.Open asset ↗SeiriosLab/Lycheepdf-raw-page:13 lines:1-55Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
Accurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies. While RGB-D cameras enable high-throughput scanning in operational settings like harvesting conveyors, they produce incomplete, low-quality 3D models. Conversely, close-range Structure-from-Motion (SfM) produces high-quality reconstructions but is not suitable for high-throughput field application. This study bridges this gap through 3DPotatoTwin , a paired dataset containing 339 tuber samples across three cultivars collected in Hokkaido, Japan. Our dataset uniquely combines: (1) conveyor-acquired RGB-D point clouds, (2) ground measurement, (3) SfM reconstructions under indoor controlled environment, and (4) aligned model pairs with transformation matrices. The multi-sensory alignment employs an semi-supervised pin-guided pipeline incorporating single-pin extraction and referencing, cross-strip matching, and binary-color-enhanced ICP, achieving 0.59 ± 0.11 mm registration accuracy. Beyond serving as a benchmark for 3D phenotyping algorithms, the dataset enables training of 3D completion networks to reconstruct high-quality 3D models from partial RGB-D point clouds. Meanwhile, the proposed semi-automated annotation pipeline has the potential to accelerate 3D dataset generation for similar studies. The presented methodology demonstrates broader applicability for multi-sensor data fusion across crop phenotyping applications. The dataset and pipeline source code are publicly available at HuggingFace and GitHub, respectively.
Why it matches plant phenotyping methodsジャガイモ塊茎の3D表現型計測を対象に、RGB-D・SfM・地上計測を統合したデータセット、位置合わせパイプライン、ベンチマークを開発しており、表現型取得手法が中心である。
abstractAccurate 3D phenotyping of agricultural produce remains challenging due to the trade-off between reconstruction quality and acquisition throughput in existing sensing technologies.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicAll the batch processing scripts mentioned in this section were provided in the 3dscan folder at Github (https://github.com/UTokyo-FieldPhenomics-Lab/PotatoScan/).Open asset ↗UTokyo-FieldPhenomics-Lab/PotatoScanhtml-lines:119-131Code / dataset availability confirmedCrossref · OpenAlex · Europe PMC · checked 6 Sept 2026
Understanding the genetic basis of quantitative traits related to crop growth, yield, and stress response requires the acquisition of large-scale, high-quality phenotypic datasets. High-throughput phenotyping platforms have become effective tools for meeting this requirement. Autonomous mobile robots have gained prominence owing to their ability to carry heavy payloads, their operational flexibility, and their proximity to crops, which allows for higher imaging resolution. In this study, we introduce PhenoRob-F (a phenotyping robot for the field), a cross-row, wheeled robot designed for efficient and automated phenotyping under field conditions. The mobile platform and phenotyping module of the robot were engineered to meet the specific demands of field phenotyping, with integrated visual and satellite navigation systems enabling autonomous operation. We validated the performance of the robot through a series of experiments involving various crop canopies. By capturing RGB images of rice and wheat, we independently performed wheat ear detection and rice panicle segmentation. For wheat ear detection, we achieve a precision of 0.783, a recall of 0.822, and a mean average precision (mAP) of 0.853 when the YOLOv8m model is used. For rice panicle segmentation, the SegFormer_B0 model yielded a mean intersection over union (mIoU) of 0.949 and an accuracy of 0.987. Additionally, by capturing RGB-D data of maize canopies, we performed 3D reconstructions to calculate plant height, achieving an R 2 of 0.99 compared with manual measurements. Similar experiments with rapeseed yielded an R 2 of 0.97. Near-infrared spectral data collected from drought-stressed rice plants enabled the classification of drought severity into five categories, with classification accuracies ranging from 0.977 to 0.996. Our results reveal that PhenoRob-F is an effective tool for high-throughput phenotyping and is capable of providing precise data to support phenotypic trait analysis and the selection of superior crop genotypes.
Why it matches plant phenotyping methods圃場用自律ロボットと複数の画像・分光センシング、形質抽出手法を開発し、作物キャノピーで性能検証しているため、植物フェノタイピング手法が研究の中心である。
abstractwe introduce PhenoRob-F (a phenotyping robot for the field), a cross-row, wheeled robot designed for efficient and automated phenotyping under field conditions.
Reproduction assets foundThe paper's data availability statement explicitly links a public GitHub repository containing part of the data and code supporting this PhenoRob-F phenotyping study; remaining data are available on request.Code · publicPart of the data and code supporting this study are openly available with the following link: https://github.com/balloonhaha/PhenoRob-F. All other reasonable requests for data and research materials will be fulfilled upon contacting the corresponding authors.Open asset ↗balloonhaha/PhenoRob-Fhtml-lines:193-220Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
The architecture of rice tillers plays a pivotal role in yield potential, yet conventional phenotyping methods have struggled to capture these intricate three-dimensional (3D) structures with high fidelity. In this study, a 3D model reconstruction method was developed specifically for rice tillers to overcome the challenges posed by their slender, feature-poor morphology in multi-view stereo-based 3D reconstruction. By applying strategically designed colorful reference markers, high-resolution 3D tiller models of 231 rice landraces were reconstructed. Accurate phenotyping was achieved by introducing ScaleCalculator, a software tool that integrated depth images from a depth camera to calibrate the physical sizes of the 3D models. The high efficiency of the 3D model-based phenotyping pipeline was demonstrated by extracting the following seven key agronomic traits: flag leaf length, panicle length, first internode length below the panicle, stem length, flag leaf angle, second leaf angle from the panicle, and third leaf angle. Genome-wide association studies (GWAS) performed with these 3D traits identified numerous candidate genes, nine of which had been previously confirmed in the literature. This work provides a 3D phenomics solution tailored for slender organs and offers novel insights into the genetic regulation of complex morphological traits in rice.
Why it matches plant phenotyping methodsイネ分げつの3D再構成とScaleCalculatorによるスケール校正を開発し、7つの形態形質を抽出するフェノタイピング手法が研究の中心であるため。
abstracta 3D model reconstruction method was developed specifically for rice tillers
Reproduction assets foundThe paper's 3D tiller models for 231 rice landraces are publicly deposited on Zenodo, and the authors' ScaleCalculator phenotyping source code is publicly available on GitHub, both explicitly stated in the Data Availability Statement. SNP genotype data are unpublished and excluded.Code · publicvelopment Co. LTD, and
Jiangsu Collaborative Innovation Center for Modern Crop Production.
Data Availability Statement: The 3D tiller models created in this study are available for research pur-
poses at https://zenodo.org/records/16080993 (accessed on 18 July 2025).The source code of ScaleCal-
culator is available on GitHub at https://github.com/ganlab/OSTRA/tree/master/ScaleCalculator
(accessed on 18 July 2025).
Acknowledgments: We thank Jianmin Wan for their valuable suggestions and Jiaqi Deng for their
technical help.
Conflicts of Interest: The authors declare that there are no conflicts of interest regarding the publica-
tion of this article.
References
1. Food and Agriculture OrganizatOpen asset ↗github · ganlab/OSTRApdf-raw-page:16 lines:1-50Code / dataset availability confirmedOpenAlex · Crossref · Europe PMC · checked 13 Sept 2026
Large-scale manual measurements of plant architectural traits in tomato growth are laborious and subjective, hindering deeper understanding of temporal variations in gene expression heterogeneity. This study develops a high-throughput approach for characterizing tomato architectural traits at different growth stages and mapping temporal broad-sense heritability using an unmanned ground vehicle-based plant phenotyping system. The SegFormer with fusion of multispectral and depth imaging modalities was employed to semantically segment plant organs from the registered RGB-D and multispectral images. Organ point clouds were then generated and clustered into instances. Finally, six key architectural traits, including fruit spacing (FS), inflorescence height (IH), stem thickness (ST), leaf spacing (LS), total leaf area (TLA), and leaf inclination angle (LIA) were extracted and the temporal broad-sense heritability folds were plotted. The root mean square errors (RMSEs) of the estimated FS, IH, ST, and LS were 0.014, 0.043, 0.003, and 0.015 m, respectively. The visualizations of the estimated TLA and LIA matched the actual growth trends. The broad-sense heritability of the extracted traits exhibited different trends across the growth stages: (i) ST, IH, and FS had a gradually increased broad-sense heritability over time, (ii) LS and LIA had a decreasing trend, and (iii) TLA showed fluctuations (i.e. an M-shaped pattern) of the broad-sense heritability throughout the growth period. The developed system and analytical approach are promising tools for accurate and rapid characterization of spatiotemporal changes of tomato plant architecture in controlled environments, laying the foundation for efficient crop breeding and precision production management in the future.
Why it matches plant phenotyping methods植物形態形質を取得するUGV型マルチモーダル画像フェノタイピングシステムと解析手法の開発・定量評価が研究の中心であり、誤差検証も行っているため。
abstractThis study develops a high-throughput approach for characterizing tomato architectural traits at different growth stages and mapping temporal broad-sense heritability using an unmanned ground vehicle-based plant phenotyping system.
Reproduction assets foundThe paper's Data availability statement explicitly deposits the authors' trait-extraction pipeline code and example data on a public GitHub repository, matching the allowed URL.Code · publicThe pipeline code and example data related to this project are available as open source on GitHub ( https://github.com/DigBigPigForU/Tomato-architectural-trait-extraction ).Open asset ↗Tomato-architectural-trait-extractionlines:822-958Code / dataset availability confirmedarXiv · checked 13 Sept 2026
Crop yield estimation is a relevant problem in agriculture, because an accurate yield estimate can support farmers' decisions on harvesting or precision intervention. Robots can help to automate this process. To do so, they need to be able to perceive the surrounding environment to identify target objects such as trees and plants. In this paper, we introduce a novel approach to address the problem of hierarchical panoptic segmentation of apple orchards on 3D data from different sensors. Our approach is able to simultaneously provide semantic segmentation, instance segmentation of trunks and fruits, and instance segmentation of trees (a trunk with its fruits). This allows us to identify relevant information such as individual plants, fruits, and trunks, and capture the relationship among them, such as precisely estimate the number of fruits associated to each tree in an orchard. To efficiently evaluate our approach for hierarchical panoptic segmentation, we provide a dataset designed specifically for this task. Our dataset is recorded in Bonn, Germany, in a real apple orchard with a variety of sensors, spanning from a terrestrial laser scanner to a RGB-D camera mounted on different robots platforms. The experiments show that our approach surpasses state-of-the-art approaches in 3D panoptic segmentation in the agricultural domain, while also providing full hierarchical panoptic segmentation. Our dataset is publicly available at https://www.ipb.uni-bonn.de/data/hops/. The open-source implementation of our approach is available at https://github.com/PRBonn/hapt3D.
Why it matches plant phenotyping methodsリンゴ樹・果実・幹を3Dセグメンテーションし、樹ごとの果実数を推定する手法と専用データセットを中心に開発・評価しており、植物の器官形態・収量関連形質の取得に該当する。
abstractwe introduce a novel approach to address the problem of hierarchical panoptic segmentation of apple orchards on 3D data from different sensors.
Reproduction assets foundThe paper introduces the HOPS dataset of annotated 3D apple orchard point clouds (TLS, UAV, UGV, SfM) for hierarchical panoptic segmentation, publicly available at the authors' IPB Bonn page, and releases the open-source implementation (hapt3D) on GitHub. Both are paper-specific, public, and actionable.Code · publicThe open-source implementation of our approach is available at https://github.com/PRBonn/hapt3D .Open asset ↗PRBonn/hapt3Dlines:1-59Code / dataset availability confirmedCrossref · checked 6 Sept 2026
Abstract Monitoring plant growth is crucial for effective crop management, and using color and depth (RGBD) cameras to model lettuce has emerged as one of the most convenient and non-invasive methods. In recent years, deep learning techniques, particularly neural networks, have become popular for estimating lettuce fresh weight. However, these models are typically specific to particular datasets, lack domain adaptation, and are often limited by the availability of open-access datasets. In this study, we propose a method based on plant geometric features for estimating the rosette structure and volume of lettuce. This new approach was compared to existing methods that reconstruct surfaces from point clouds, such as Ball Pivoting and Alpha Shapes. The proposed method creates a tight hull around the plant's point cloud, preserving high detail of the rosette structure while filling in surface holes in areas not visible to 3D cameras. Using a linear regression model, we estimated fresh weight for this dataset, achieving a root mean square error (RMSE) of 18.2 g when using only the estimated plant volume, and 17.3 g when both volume and geometric features were included. Additionally, we introduced new geometric features that characterize leaf density, which could be useful for breeding applications. A dataset of 402 point clouds of lettuce plants, captured before harvest, was compiled using one top-down and three side-view 3D cameras.
Why it matches plant phenotyping methodsRGB-D画像からレタスの構造・体積・葉密度を抽出し、生体重推定を検証する手法開発が研究の中心であり、データセットも構築している。
abstractIn this study, we propose a method based on plant geometric features for estimating the rosette structure and volume of lettuce.
Reproduction assets foundThe paper's own lettuce 3D point cloud dataset (Pii, 402 point clouds with fresh weight references) is deposited on Zenodo, and the vacuum-package surface reconstruction code plus data processing scripts are publicly available on the authors' GitHub repository. Both are paper-specific, public, and actionable.Dataset · publicData used in this study and developed models are available on Zenodo storage service https://zenodo.org/records/8410252 .Open asset ↗Zenodo · 8410252lines:158-220Code · publicThe code used at this study is available at https://github.com/VicB18/LettuceFW (accessed on 1 November 2024).Open asset ↗GitHub · VicB18/LettuceFWlines:158-220Code · publicThe code for the vacuum package method, along with the data processing scripts used in this study, are available in the Supplementary Information.Open asset ↗lines:98-114Code / dataset availability confirmedOpenAlex · Europe PMC · checked 14 Sept 2026
In the era of machine learning-driven plant imaging, the production of annotated datasets is a very important contribution. In this data paper, a unique annotated dataset of seedling emergence kinetics is proposed. It is composed of almost 70,000 RGB-depth frames and more than 700,000 plant annotations. The dataset is shown valuable for training deep learning models and performing high-throughput phenotyping by imaging. The ability of such models to generalize to several species and outperform the state-of-the-art owing to the delivered dataset is demonstrated. We also discuss how this dataset raises new questions in plant phenotyping.
Why it matches plant phenotyping methods植物の出芽速度を対象とする大規模RGB深度画像・アノテーションデータセットを提供し、深層学習および高スループット表現型解析への利用性を実証しており、表現型取得基盤が中心である。
abstracta unique annotated dataset of seedling emergence kinetics is proposed
Reproduction assets foundThis is a data paper whose core contribution is a public annotated RGB-depth seedling dataset (~70,000 frames, >700,000 annotations) deposited in DATA INRAE with DOI 10.57745/AMFJTK, explicitly stated as publicly accessible. Other allowed URLs (license, Intel datasheet, Jülich record) are not paper-specific assets.Dataset · publicSynthesis of the full time-lapse and RGB-Depth full frame quantity per species
Species
Pots time-lapse
Labelled pots time-lapse
RGB-depth full frame
Rapeseed
1 760
336
15 218
Tomatoes
1 960
480
33 283
Beans
2 320
400
21 445
Total
6 040
1 216
69 946
The dataset is publicly accessible in the DATA INRAE repository, DOI: https://doi.org/10.57745/AMFJTK . The file tree structure is illustrated in Fig. 4 . The dataset is organized into 11 compressed .zip files, each corresponding to a distinct trial. Within these files, images are sorted chronologically by acquisition start date, then by camera, and stored in .png format within dedicated color and depth folders. Labels are alsoOpen asset ↗DATA INRAE · 10.57745/AMFJTKlines:105-195Code / dataset availability confirmedEurope PMC · Crossref · checked 6 Sept 2026
Fruit size is crucial for growers as it influences consumer willingness to buy and the price of the fruit. Fruit size and growth along the seasons are two parameters that can lead to more precise orchard management favoring production sustainability. In this study, a Python-based computer vision system (CVS) for sizing apples directly on the tree was developed to ease fruit sizing tasks. The system is made of a consumer-grade depth camera and was tested at two distances among 17 timings throughout the season, in a Fuji apple orchard. The CVS exploited a specifically trained YOLOv5 detection algorithm, a circle detection algorithm, and a trigonometric approach based on depth information to size the fruits. Comparisons with standard-trained YOLOv5 models and with spherical objects were carried out. The algorithm showed good fruit detection and circle detection performance, with a sizing rate of 92%. Good correlations (r > 0.8) between estimated and actual fruit size were found. The sizing performance showed an overall mean error (mE) and RMSE of + 5.7 mm (9%) and 10 mm (15%). The best results of mE were always found at 1.0 m, compared to 1.5 m. Key factors for the presented methodology were: the fruit detectors customization; the HoughCircle parameters adaptability to object size, camera distance, and color; and the issue of field natural illumination. The study also highlighted the uncertainty of human operators in the reference data collection (5–6%) and the effect of random subsampling on the statistical analysis of fruit size estimation. Despite the high error values, the CVS shows potential for fruit sizing at the orchard scale. Future research will focus on improving and testing the CVS on a large scale, as well as investigating other image analysis methods and the ability to estimate fruit growth.
Why it matches plant phenotyping methods果実サイズという植物器官形質を、深度カメラ・物体検出・円検出・三角測量で推定するコンピュータビジョン手法を開発・検証しており、フェノタイピング手法が中心である。
abstracta Python-based computer vision system (CVS) for sizing apples directly on the tree was developed
Reproduction assets foundThe paper's RGB-D apple dataset (RGB/depth frames, annotations, and reference caliper measurements) is explicitly released as open source on GitHub with a Zenodo DOI. The YOLOv5 base model is a generic third-party library, not a paper-specific asset; no author analysis code repository is stated.Dataset · publicThe obtained dataset is open source and available at
https://github.com/ECOPOM/OpenAcces_RGBD_apple_dataset (Bortolotti et al., 2024).Open asset ↗ECOPOM/OpenAcces_RGBD_apple_datasetpdf-page:3 lines:1-52Code / dataset availability confirmedOpenAlex · arXiv · checked 7 Sept 2026
In modern agriculture, precise monitoring of plants and fruits is crucial for tasks such as high-throughput phenotyping and automated harvesting. This paper addresses the challenge of reconstructing accurate 3D shapes of fruits from partial views, which is common in agricultural settings. We introduce CF-PRNet, a coarse-to-fine prototype refining network, leverages high-resolution 3D data during the training phase but requires only a single RGB-D image for real-time inference. Our approach begins by extracting the incomplete point cloud data that constructed from a partial view of a fruit with a series of convolutional blocks. The extracted features inform the generation of scaling vectors that refine two sequentially constructed 3D mesh prototypes - one coarse and one fine-grained. This progressive refinement facilitates the detailed completion of the final point clouds, achieving detailed and accurate reconstructions. CF-PRNet demonstrates excellent performance metrics with a Chamfer Distance of 3.78, an F1 Score of 66.76%, a Precision of 56.56%, and a Recall of 85.31%, and win the first place in the Shape Completion and Reconstruction of Sweet Peppers Challenge.
Why it matches plant phenotyping methods果実の部分RGB-D画像から3D形状を再構成する手法を開発・評価しており、植物器官の形態形質取得が研究の中心である。
abstractThis paper addresses the challenge of reconstructing accurate 3D shapes of fruits from partial views
Reproduction assets foundThe paper's authors publicly release their CF-PRNet source code for sweet pepper point cloud completion. The sweet pepper benchmark dataset is cited prior work (ref [2]), not a paper-specific asset, and the challenge website is a generic event page.Code · publicOur source code is
available at https://github.com/uqzhichen/CF-PRNet/.Open asset ↗uqzhichen/CF-PRNetpdf-page:1 lines:1-50Code / dataset availability confirmedarXiv · OpenAlex · checked 13 Sept 2026
Potato yield is an important metric for farmers to further optimize their cultivation practices. Potato yield can be estimated on a harvester using an RGB-D camera that can estimate the three-dimensional (3D) volume of individual potato tubers. A challenge, however, is that the 3D shape derived from RGB-D images is only partially completed, underestimating the actual volume. To address this issue, we developed a 3D shape completion network, called CoRe++, which can complete the 3D shape from RGB-D images. CoRe++ is a deep learning network that consists of a convolutional encoder and a decoder. The encoder compresses RGB-D images into latent vectors that are used by the decoder to complete the 3D shape using the deep signed distance field network (DeepSDF). To evaluate our CoRe++ network, we collected partial and complete 3D point clouds of 339 potato tubers on an operational harvester in Japan. On the 1425 RGB-D images in the test set (representing 51 unique potato tubers), our network achieved a completion accuracy of 2.8 mm on average. For volumetric estimation, the root mean squared error (RMSE) was 22.6 ml, and this was better than the RMSE of the linear regression (31.1 ml) and the base model (36.9 ml). We found that the RMSE can be further reduced to 18.2 ml when performing the 3D shape completion in the center of the RGB-D image. With an average 3D shape completion time of 10 milliseconds per tuber, we can conclude that CoRe++ is both fast and accurate enough to be implemented on an operational harvester for high-throughput potato yield estimation. CoRe++'s high-throughput and accurate processing allows it to be applied to other tuber, fruit and vegetable crops, thereby enabling versatile, accurate and real-time yield monitoring in precision agriculture. Our code, network weights and dataset are publicly available at https://github.com/UTokyo-FieldPhenomics-Lab/corepp.git.
Why it matches plant phenotyping methodsRGB-D画像からジャガイモ塊茎の3D形状を補完し、体積・収量を推定する手法を開発・検証しており、植物フェノタイピング手法が研究の中心である。
abstractwe developed a 3D shape completion network, called CoRe++, which can complete the 3D shape from RGB-D images.
Reproduction assets foundThe paper's abstract explicitly states that the authors' code, network weights, and the potato tuber RGB-D/3D point cloud dataset are publicly available at the authors' GitHub repository (UTokyo-FieldPhenomics-Lab/corepp), which is a paper-specific, public, actionable asset for the CoRe++ phenotyping analysis.Code · publicOur code, network weights and dataset are publicly available at https://github.com/UTokyo-FieldPhenomics-Lab/corepp.git .Open asset ↗UTokyo-FieldPhenomics-Lab/corepplines:1-93Code / dataset availability confirmedarXiv · checked 14 Sept 2026
As the world population is expected to reach 10 billion by 2050, our agricultural production system needs to double its productivity despite a decline of human workforce in the agricultural sector. Autonomous robotic systems are one promising pathway to increase productivity by taking over labor-intensive manual tasks like fruit picking. To be effective, such systems need to monitor and interact with plants and fruits precisely, which is challenging due to the cluttered nature of agricultural environments causing, for example, strong occlusions. Thus, being able to estimate the complete 3D shapes of objects in presence of occlusions is crucial for automating operations such as fruit harvesting. In this paper, we propose the first publicly available 3D shape completion dataset for agricultural vision systems. We provide an RGB-D dataset for estimating the 3D shape of fruits. Specifically, our dataset contains RGB-D frames of single sweet peppers in lab conditions but also in a commercial greenhouse. For each fruit, we additionally collected high-precision point clouds that we use as ground truth. For acquiring the ground truth shape, we developed a measuring process that allows us to record data of real sweet pepper plants, both in the lab and in the greenhouse with high precision, and determine the shape of the sensed fruits. We release our dataset, consisting of almost 7,000 RGB-D frames belonging to more than 100 different fruits. We provide segmented RGB-D frames, with camera intrinsics to easily obtain colored point clouds, together with the corresponding high-precision, occlusion-free point clouds obtained with a high-precision laser scanner. We additionally enable evaluation of shape completion approaches on a hidden test set through a public challenge on a benchmark server.
Why it matches plant phenotyping methods果実の3D形状という植物器官の形態形質を対象に、RGB-D画像・高精度点群・評価用ベンチマークを構築しており、形状取得と推定の方法論が中心である。
abstractWe provide an RGB-D dataset for estimating the 3D shape of fruits.
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 · publicOur development toolkit including a data loader is available at:
https://github.com/PRBonn/shape_completion_toolkit for handling the dataset and computing metrics.Open asset ↗PRBonn/shape_completion_toolkitlines:55-81Code / dataset availability confirmedEurope PMC · checked 13 Sept 2026
Feeding the increasing global population and reducing the carbon footprint of agricultural activities are two critical challenges of our century. Growing crops under protected horticulture and precise crop monitoring have emerged to address these challenges. Crop monitoring in commercial protected facilities remains mostly manual and labour intensive. Using computer vision to solve specific problems in image-based crop monitoring in these compact and complex growth environments is currently hindered by the scarcity of available data. We collected an RGBD dataset for vertically supported, hydroponically-grown capsicum plants in a commercial-scale glasshouse facility to fill this gap. Data were collected weekly using a single top-angled stereo camera mounted on a mobile platform running between the hydroponic gutters. The RGBD streams covered 80 % of the crop growing season in three different light conditions. The metadata include camera configurations and light condition information. Manually measured plant heights of ten selected plants per gutter are provided as ground truth. The images covered the whole plants and focused on the top third. This dataset will support research on plant height estimation, plant organ identification, object segmentation, organ measurements, 3D reconstruction, 3D data processing, and depth noise reduction. The usability of the dataset has been successfully demonstrated in a previously published study on plant height estimation using machine learning and 3D point cloud.
Why it matches plant phenotyping methods植物の草丈推定や器官計測を目的としたRGBD画像データセットを構築し、地上真値も提供しているため、植物フェノタイピング手法・データセットが中心です。
abstractWe collected an RGBD dataset for vertically supported, hydroponically-grown capsicum plants in a commercial-scale glasshouse facility to fill this gap.
Reproduction assets foundThis data article directly deposits its paper-specific phenotyping assets: the PC4C_CAPSI RGBD image dataset (Rosbag streams, JSON metadata, manual plant-height ground truth) on the Western Sydney University ResearchDirect repository, and the authors' RGBD processing code (image extraction, depth correction, 3D reconssDataset · publicData accessibility
Repository name: Image Data of Capsicum Plant Growth in Protected Horticulture: PC4C_CAPSI. [ 1 ]
Data identification number: 10.26183/1A0R-E318
Direct URL to data: https://rds.westernsydney.edu.au/Institutes/HIE/2024/Jayasuriya_N/Open asset ↗rds.westernsydney.edu.aulines:1-40Code / dataset availability confirmedOpenAlex · arXiv · checked 15 Sept 2026
The use of multiple camera technologies in a combined multimodal monitoring system for plant phenotyping offers promising benefits. Compared to configurations that only utilize a single camera technology, cross-modal patterns can be recorded that allow a more comprehensive assessment of plant phenotypes. However, the effective utilization of cross-modal patterns is dependent on precise image registration to achieve pixel-accurate alignment, a challenge often complicated by parallax and occlusion effects inherent in plant canopy imaging. In this study, we propose a novel multimodal 3D image registration method that addresses these challenges by integrating depth information from a time-of-flight camera into the registration process. By leveraging depth data, our method mitigates parallax effects and thus facilitates more accurate pixel alignment across camera modalities. Additionally, we introduce an automated mechanism to identify and differentiate different types of occlusions, thereby minimizing the introduction of registration errors. To evaluate the efficacy of our approach, we conduct experiments on a diverse image dataset comprising six distinct plant species with varying leaf geometries. Our results demonstrate the robustness of the proposed registration algorithm, showcasing its ability to achieve accurate alignment across different plant types and camera compositions. Compared to previous methods it is not reliant on detecting plant specific image features and can thereby be utilized for a wide variety of applications in plant sciences. The registration approach principally scales to arbitrary numbers of cameras with different resolutions and wavelengths. Overall, our study contributes to advancing the field of plant phenotyping by offering a robust and reliable solution for multimodal image registration.
Why it matches plant phenotyping methods植物フェノタイピングのためのマルチモーダル3D画像登録手法を開発し、複数植物種の画像データセットで性能評価しているため、方法が研究の中心である。
abstractIn this study, we propose a novel multimodal 3D image registration method that addresses these challenges by integrating depth information from a time-of-flight camera into the registration process.
Reproduction assets foundThe paper's multimodal plant image dataset (six plant species recorded with the RGBD/thermal/hyperspectral setup) is publicly available on the authors' GitHub repository, which is an allowed URL.Dataset · publiclity of our registration algorithm across diverse scenarios, we recorded a dataset comprising images of six distinct plant species. This was done to encompass a wide variety of leaf and canopy structures, thus offering a representative sample for evaluation purposes. The recorded dataset can be found on the project github page: https://github.com/eric-stumpe/Plant3DImageReg . The chosen plant species are as follows:
1.
Grapevine ( Vitis vinifera )
2.
Leopard lily ( Dieffenbachia )
3.Open asset ↗eric-stumpe/Plant3DImageReglines:287-310Code / dataset availability confirmedEurope PMC · OpenAlex · checked 7 Sept 2026
Stem diameter is a critical phenotypic parameter for maize, integral to yield prediction and lodging resistance assessment. Traditionally, the quantification of this parameter through manual measurement has been the norm, notwithstanding its tedious and laborious nature. To address these challenges, this study introduces a non-invasive field-based system utilizing depth information from RGB-D cameras to measure maize stem diameter. This technology offers a practical solution for conducting rapid and non-destructive phenotyping. Firstly, RGB images, depth images, and 3D point clouds of maize stems were captured using an RGB-D camera, and precise alignment between the RGB and depth images was achieved. Subsequently, the contours of maize stems were delineated using 2D image processing techniques, followed by the extraction of the stem's skeletal structure employing a thinning-based skeletonization algorithm. Furthermore, within the areas of interest on the maize stems, horizontal lines were constructed using points on the skeletal structure, resulting in 2D pixel coordinates at the intersections of these horizontal lines with the maize stem contours. Subsequently, a back-projection transformation from 2D pixel coordinates to 3D world coordinates was achieved by combining the depth data with the camera's intrinsic parameters. The 3D world coordinates were then precisely mapped onto the 3D point cloud using rigid transformation techniques. Finally, the maize stem diameter was sensed and determined by calculating the Euclidean distance between pairs of 3D world coordinate points. The method demonstrated a Mean Absolute Percentage Error ( MAPE ) of 3.01%, a Mean Absolute Error ( MAE ) of 0.75 mm, a Root Mean Square Error ( RMSE ) of 1.07 mm, and a coefficient of determination ( R ²) of 0.96, ensuring accurate measurement of maize stem diameter. This research not only provides a new method of precise and efficient crop phenotypic analysis but also offers theoretical knowledge for the advancement of precision agriculture.
Why it matches plant phenotyping methodsRGB-Dカメラと画像・3D処理によりトウモロコシ茎径を非破壊測定する手法を開発し、誤差指標で精度検証しており、フェノタイピング手法が中心である。
abstractthis study introduces a non-invasive field-based system utilizing depth information from RGB-D cameras to measure maize stem diameter
Reproduction assets foundThe paper's data availability statement points to a public Figshare deposit (DOI 10.6084/m9.figshare.25450039) containing the study's datasets (RGB/depth imagery and stem diameter measurements used for the maize stem diameter phenotyping analysis). No author analysis code or trained models are explicitly deposited.Dataset · publicThe datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: http://dx.doi.org/10.6084/m9.figshare.25450039 .Open asset ↗figshare · 10.6084/m9.figshare.25450039lines:909-917Code / dataset availability confirmedOpenAlex · checked 15 Sept 2026
Trunk diameter is related to the overall health and level of carbon sequestration in a tree. Trunk diameter measurement, therefore, is a key task in both forest plot and urban settings. Unlike the traditional approach of manual measurement with a measuring tape or calipers, several recent approaches rely on sophisticated technologies such as Terrestrial Laser Scanning (TLS), LiDAR, and time-of-flight sensors that provide fine-grain depth maps, which are used for depth-assisted image segmentation in downstream processing. These technologies are supported only on specialized devices or high-end smartphones. We present a mobile application that uses coarse-grain depth maps derived from an optical sensor, and so can be run on most common Android devices. Moreover, we use a state-of-the-art deep neural network to estimate trunk diameter from an image and its corresponding coarse depth map (RGB-D). We tested our app using a dataset collected from four countries and under challenging conditions including occlusion, leaning trees, and irregular shapes and found that our algorithm has a MAE of 2.58 cm and an RMSE of 3.57 cm, which is comparable to accuracy from fine-grain depth maps. Moreover, diameter measurement using our app is more than 5 times faster than traditional manual surveying.
Why it matches plant phenotyping methodsRGB-D画像と粗い深度マップから樹幹直径を推定するモバイル手法を開発し、複数国のデータセットと困難条件で精度検証しており、植物形質取得が中心である。
abstractWe present a mobile application that uses coarse-grain depth maps derived from an optical sensor
Reproduction assets foundThe paper's DBH estimation evaluation dataset (154 RGB + depth tree images with metadata and ground-truth DBH) is publicly deposited on Zenodo, and the app/algorithm source code is publicly available on GitHub. Both are paper-specific, public, and actionable.Dataset · publicing the quality of the DBH estimate, includ-
197
ing non-cylindrical trunks, burl presence, degrees of leaning and occlusion, and poor lighting, as illustrated in Fig. 4. Sample images
198
from the dataset are available in Supplement 10.9, and the complete set of RGB and depth images, along with metadata, is acces-
199
sible at https://zenodo.org/records/10199711.200
In Thailand, we collected data in Bangkok’s Lumphini Park and Chulalongkorn Centenary Park. As a tropical location, Bangkok is
201
home to many tropical trees, such as rain trees (Samanea saman), banyan trees, palm trees, and coconut trees52. At Lumphini Park,
202
where most of our data came from, small forests grow next to watOpen asset ↗zenodo.org · 10199711pdf-raw-page:8 lines:1-31Code · public; Z.F. analyzed the data and led the writing of the manuscript.
352
A.H. and S.K. reviewed the manuscript and provided constructive suggestions. All authors contributed critically to the drafts and
353
gave final approval for publication.
354
8 DATA AVAILABILITY
355
The algorithm and app code are publicly available on GitHub at https://github.com/MingyueX/GreenLens, with APK available from
356
APKPure at https://apkpure.com/p/com.cleeg.greenlens. All the data for the app evaluation can be accessed at https://zenodo.org/357
records/10199711.
358
9 AUTHOR COMPETING INTERESTS
359
The authors declare no conflict of interest.
360
361
REFERENCES
362
[1] Kenneth G MacDicken. Global forest resourOpen asset ↗github.com/MingyueX/GreenLenspdf-raw-page:15 lines:1-92Code / dataset availability confirmedEurope PMC · checked 13 Sept 2026
The present dataset comprises a collection of RGB-D apple tree images that can be used to train and test computer vision-based fruit detection and sizing methods. This dataset encompasses two distinct sets of data obtained from a Fuji and an Elstar apple orchards. The Fuji apple orchard sub-set consists of 3925 RGB-D images containing a total of 15,335 apples annotated with both modal and amodal apple segmentation masks. Modal masks denote the visible portions of the apples, whereas amodal masks encompass both visible and occluded apple regions. Notably, this dataset is the first public resource to incorporate on-tree fruit amodal masks. This pioneering inclusion addresses a critical gap in existing datasets, enabling the development of robust automatic fruit sizing methods and accurate fruit visibility estimation, particularly in the presence of partial occlusions. Besides the fruit segmentation masks, the dataset also includes the fruit size (calliper) ground truth for each annotated apple. The second sub-set comprises 2731 RGB-D images capturing five Elstar apple trees at four distinct growth stages. This sub-set includes mean diameter information for each tree at every growth stage and serves as a valuable resource for evaluating fruit sizing methods trained with the first sub-set. The present data was employed in the research paper titled "Looking behind occlusions: a study on amodal segmentation for robust on-tree apple fruit size estimation" [1].
Why it matches plant phenotyping methodsリンゴ果実のRGB-D画像、アノテーション、サイズ正解値を含む公開データセットで、果実サイズ推定法の開発・評価を直接支援するため、植物フェノタイピング手法のデータ資源として中心的です。
abstractenabling the development of robust automatic fruit sizing methods and accurate fruit visibility estimation
Reproduction assets foundThe article is a Data in Brief describing the AmodalAppleSize_RGB-D dataset (RGB-D apple tree images with modal/amodal segmentation masks and fruit size ground truth), publicly deposited in Dataverse (CORA) with DOI 10.34810/data916 and a direct URL. This is the paper's own phenotyping data (images, annotations, callipDataset · publicData accessibility
Repository name: Dataverse
Data identification number: https://doi.org/10.34810/data916 [2]
Direct URL to data: https://dataverse.csuc.cat/dataset.xhtml?persistentId=doi:10.34810/data916Open asset ↗Dataverse · doi:10.34810/data916lines:43-67Code / dataset availability confirmedOpenAlex · checked 13 Sept 2026
Rapid and precise tree Diameter at Breast Height (DBH) measurement is pivotal in forest inventories. While the recent advancements in LiDAR and Structure from Motion (SFM) technologies have paved the way for automated DBH measurements, the significant equipment costs and the complexity of operational procedures continue to constrain the ubiquitous adoption of these technologies for real-time DBH assessments. In this research, we introduce KAN-Forest, a real-time DBH measurement and key point localization algorithm utilizing RGB-D (Red, Green, Blue-Depth) imaging technology. Firstly, we improved the YOLOv5-seg segmentation module with a Channel and Spatial Attention (CBAM) module, augmenting its efficiency in extracting the tree’s edge features in intricate forest scenarios. Subsequently, we devised an image processing algorithm for real-time key point localization and DBH measurement, leveraging historical data to fine-tune current frame assessments. This system facilitates real-time image data upload via wireless LAN for immediate host computer processing. We validated our approach on seven sample plots, achieving bbAP50 and segAP50 scores of: 90.0%(+3.0%), 90.9%(+0.9%), respectively with the improved YOLOv5-seg model. The method exhibited a DBH estimation RMSE of 17.61∼54.96 mm (R2=0.937), and secured 78% valid DBH samples at a 59 FPS. Our system stands as a cost-effective, portable, and user-friendly alternative to conventional forest survey techniques, maintaining accuracy in real-time measurements compared to SFM- and LiDAR-based algorithms. The integration of WLAN and its inherent scalability facilitates deployment on Unmanned Ground Vehicles (UGVs) to improve the efficiency of forest inventory. We have shared the algorithms and datasets on Github for peer evaluations.
Why it matches plant phenotyping methodsRGB-D画像とアルゴリズムを用いて樹木のDBHという明示的な形態形質をリアルタイム推定し、精度検証とシステム実装を行った研究であり、植物フェノタイピング手法が中心である。
abstractwe introduce KAN-Forest, a real-time DBH measurement and key point localization algorithm utilizing RGB-D (Red, Green, Blue-Depth) imaging technology.
Reproduction assets foundThe authors explicitly state that the code used in this DBH measurement research is publicly available on GitHub (KAN-Forest repository), making it a paper-specific, public, actionable code asset. The phenotype/trait datasets (DBH measurements and forest images) are only available upon request from the corresponding作者,Code · publicwe have made
the code used in this research available on GitHub at: https://github.com/CharmingZh/KAN-ForestOpen asset ↗CharmingZh/KAN-Forestpdf-page:27 lines:1-59Code / dataset availability confirmedEurope PMC · OpenAlex · checked 15 Sept 2026
The measurement of fruit size is of great interest to estimate the yield and predict the harvest resources in advance. This work proposes a novel technique for in-field apple detection and measurement based on Deep Neural Networks. The proposed framework was trained with RGB-D data and consists of an end-to-end multitask Deep Neural Network architecture specifically designed to perform the following tasks: 1) detection and segmentation of each fruit from its surroundings; 2) estimation of the diameter of each detected fruit. The methodology was tested with a total of 15,335 annotated apples at different growth stages, with diameters varying from 27 mm to 95 mm. Fruit detection results reported an F1-score for apple detection of 0.88 and a mean absolute error of diameter estimation of 5.64 mm. These are state-of-the-art results with the additional advantages of: a) using an end-to-end multitask trainable network; b) an efficient and fast inference speed; and c) being based on RGB-D data which can be acquired with affordable depth cameras. On the contrary, the main disadvantage is the need of annotating a large amount of data with fruit masks and diameter ground truth to train the model. Finally, a fruit visibility analysis showed an improvement in the prediction when limiting the measurement to apples above 65% of visibility (mean absolute error of 5.09 mm). This suggests that future works should develop a method for automatically identifying the most visible apples and discard the prediction of highly occluded fruits.
Why it matches plant phenotyping methodsRGB-D画像と深層学習を用いて果実の検出・セグメンテーションおよび直径推定法を開発し、アノテーションデータで性能評価しているため、果実形質の取得手法が中心である。
abstractThis work proposes a novel technique for in-field apple detection and measurement based on Deep Neural Networks.
Reproduction assets foundThe authors explicitly state that the code for their multitask Mask R-CNN diameter-regression network was made publicly available together with the annotated RGB-D apple dataset (masks, diameter ground truth, spherical mask projections) at the GRAP-UdL publication page. This is a paper-specific, public, actionable codeCode · publice, which goes from
14 14 (default pooling resolution) to 28 28. After the
deconvolution, the data is flattened and fed to a linear layer
that predicts the diameter for that mask.
The developed network was implemented in the Pytorch
framework and the code has been made publicly available
jointly with the presented dataset at http://www.grap.udl.cat/en/publications/papple_rgb-d-size-dataset/.2.2.3. Network training and inference details
a) Weight initialisation: Mask ReCNN has a set of weight
initialisations pre-trained with different backbones on
ImageNet (Deng et al., 2009). In our case, the used
weights were pre-trained with a ResNet50 backbone.
However, during the course of this projecOpen asset ↗pdf-raw-page:6 lines:1-143Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Improving soybean ( Glycine max L. (Merr.)) yield is crucial for strengthening national food security. Predicting soybean yield is essential to maximize the potential of crop varieties. Non-destructive methods are needed to estimate yield before crop maturity. Various approaches, including the pod-count method, have been used to predict soybean yield, but they often face issues with the crop background color. To address this challenge, we explored the application of a depth camera to real-time filtering of RGB images, aiming to enhance the performance of the pod-counting classification model. Additionally, this study aimed to compare object detection models (YOLOV7 and YOLOv7-E6E) and select the most suitable deep learning (DL) model for counting soybean pods. After identifying the best architecture, we conducted a comparative analysis of the model's performance by training the DL model with and without background removal from images. Results demonstrated that removing the background using a depth camera improved YOLOv7's pod detection performance by 10.2% precision, 16.4% recall, 13.8% mAP@50, and 17.7% mAP@0.5:0.95 score compared to when the background was present. Using a depth camera and the YOLOv7 algorithm for pod detection and counting yielded a mAP@0.5 of 93.4% and mAP@0.5:0.95 of 83.9%. These results indicated a significant improvement in the DL model's performance when the background was segmented, and a reasonably larger dataset was used to train YOLOv7.
Why it matches plant phenotyping methods深度カメラと物体検出モデルを用いてダイズ莢数という植物形質を非破壊推定する手法を開発・比較・評価しており、表現型取得が中心的である。
abstractwe explored the application of a depth camera to real-time filtering of RGB images, aiming to enhance the performance of the pod-counting classification model.
Reproduction assets foundThe paper's Data Availability Statement points to an authors' public GitHub repository containing the datasets generated and analyzed (soybean depth-camera images and pod-count segmentation data). Other URLs (labelImg, scikit-learn, CC license) are generic tools/licenses, not paper-specific assets.Dataset · publicThe datasets generated and analyzed for this study can be found in the Github repository Soybean pod count depth segmentation project 2022 accessible at https://github.com/jithin8mathew/soybean_pod_count_Depth_segmentation_project (accessed on 28 June 2023).Open asset ↗jithin8mathew/soybean_pod_count_Depth_segmentation_projectlines:183-198Code / dataset availability confirmedarXiv · OpenAlex · checked 13 Sept 2026
Artificial intelligence applications enable farmers to optimize crop growth and production while reducing costs and environmental impact. Computer vision-based algorithms in particular, are commonly used for fruit segmentation, enabling in-depth analysis of the harvest quality and accurate yield estimation. In this paper, we propose TomatoDIFF, a novel diffusion-based model for semantic segmentation of on-plant tomatoes. When evaluated against other competitive methods, our model demonstrates state-of-the-art (SOTA) performance, even in challenging environments with highly occluded fruits. Additionally, we introduce Tomatopia, a new, large and challenging dataset of greenhouse tomatoes. The dataset comprises high-resolution RGB-D images and pixel-level annotations of the fruits.
Why it matches plant phenotyping methods植物上のトマト果実を画像からセグメンテーションする手法を開発・比較し、RGB-D画像と画素アノテーションのデータセットも提供しており、植物器官の状態・位置推定に関わる方法が中心である。
abstractwe propose TomatoDIFF, a novel diffusion-based model for semantic segmentation of on-plant tomatoes
Reproduction assets foundThe paper introduces TomatoDIFF and the Tomatopia dataset, with explicit public availability of source code and dataset at the authors' GitHub repository. It also trains/evaluates on the public Kaggle 'Tomato dataset' (andrewmvd/tomato-detection), which is a paper-specific public image dataset used directly in the phenCode · publicThe source code of TomatoDIFF and Tomatopia are available at https://github.com/MIvanovska/TomatoDIFF .Open asset ↗MIvanovska/TomatoDIFFlines:1-44Code / dataset availability confirmedOpenAlex · checked 15 Sept 2026
The detection and sizing of fruits with computer vision methods is of interest because it provides relevant information to improve the management of orchard farming. However, the presence of partially occluded fruits limits the performance of existing methods, making reliable fruit sizing a challenging task. While previous fruit segmentation works limit segmentation to the visible region of fruits (known as modal segmentation), in this work we propose an amodal segmentation algorithm to predict the complete shape, which includes its visible and occluded regions. To do so, an end-to-end convolutional neural network (CNN) for simultaneous modal and amodal instance segmentation was implemented. The predicted amodal masks were used to estimate the fruit diameters in pixels. Modal masks were used to identify the visible region and measure the distance between the apples and the camera using the depth image. Finally, the fruit diameters in millimetres (mm) were computed by applying the pinhole camera model. The method was developed with a Fuji apple dataset consisting of 3925 RGB-D images acquired at different growth stages with a total of 15,335 annotated apples, and was subsequently tested in a case study to measure the diameter of Elstar apples at different growth stages. Fruit detection results showed an F1-score of 0.86 and the fruit diameter results reported a mean absolute error (MAE) of 4.5 mm and R2 = 0.80 irrespective of fruit visibility. Besides the diameter estimation, modal and amodal masks were used to automatically determine the percentage of visibility of measured apples. This feature was used as a confidence value, improving the diameter estimation to MAE = 2.93 mm and R2 = 0.91 when limiting the size estimation to fruits detected with a visibility higher than 60%. The main advantages of the present methodology are its robustness for measuring partially occluded fruits and the capability to determine the visibility percentage. The main limitation is that depth images were generated by means of photogrammetry methods, which limits the efficiency of data acquisition. To overcome this limitation, future works should consider the use of commercial RGB-D sensors. The code and the dataset used to evaluate the method have been made publicly available at https://github.com/GRAP-UdL-AT/Amodal_Fruit_Sizing.
Why it matches plant phenotyping methods果実の遮蔽に頑健な画像ベースのアモーダル分割と、リンゴ果径という植物形質の推定手法を開発・検証しており、方法が研究の中心である。
abstractThe predicted amodal masks were used to estimate the fruit diameters in pixels.
Reproduction assets foundThe paper's apple amodal segmentation dataset (RGB-D images, modal/amodal masks, calliper-measured diameters) and the authors' analysis code are both explicitly stated to be publicly available at the authors' GitHub repository GRAP-UdL-AT/Amodal_Fruit_Sizing.Dataset · publictain data from both maturity stages, of different fruit size and with
different fruit visibilities. The dataset split was performed randomly,
obtaining in each partition a similar distribution of diameters (Fig. 4.b)
and apples visibilities (Fig. 4.d) than in the original dataset. The dataset
has been made publicly available at https://github.com/GRAP-UdL-AT/Amodal_Fruit_Sizing.The data used for the case study was acquired in an Elstar apple
orchard located in Randwijk (the Netherlands). Five different trees were
imaged at four different dates (Table 1), obtaining data at different
growth stages: BBCH75, BBCH77, BBCH78 and BBCH85 (Fig. 2b). To
have a complete representation of trees, images Open asset ↗GRAP-UdL-AT/Amodal_Fruit_Sizingpdf-raw-page:3 lines:1-74Code · publicft, Supervision.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Data availability
The code and the dataset used to evaluate the method have been
made publicly available at https://github.com/GRAP-UdL-AT/Amodal_Fruit_Sizing.Acknowledgements
This work was partly funded by the Departament de Recerca i Uni
versitats de la Generalitat de Catalunya (grant 2021 LLAV 00088), the
Spanish Ministry of Science, Innovation and Universities (grants
RTI2018-094222-B-I00 [PAgFRUIT project], PID2021-126648OB-I00
[PAgPROTECT project] and PID2020-117142GOpen asset ↗GRAP-UdL-AT/Amodal_Fruit_Sizingpdf-raw-page:12 lines:1-75Code / dataset availability confirmedEurope PMC · checked 13 Sept 2026
In most of the countries, grapes are considered as a cash crop. Currently huge research is going on in development of automated grape harvesting systems. Speedy and reliable grape bunch detection is prime need for various deep learning based automated systems which deals with object detection and object segmentation tasks. But currently very few datasets are available on grape bunches in vineyard, because of which there is restriction to the research in this area. In comparison to the vineyard in outside countries, Indian vineyard structure is more complex, so it becomes hard to work in real-time. To overcome these problems and to make vineyard dataset for suitable for Indian vineyard scenarios, this paper proposed four different datasets on grape bunches in vineyard. For creating all datasets in GrapesNet, natural environmental conditions have been considered. GrapesNet includes total 11000+ images of grape bunches. Necessary data for weight prediction of grape cluster is also provided with dataset like height, width and real weight of cluster present in image. Proposed datasets can be used for prime tasks like grape bunch detection, grape bunch segmentation, and grape bunch weight estimation etc. of future generation automated vineyard harvesting technologies.
Why it matches plant phenotyping methodsブドウ果房画像データセットを構築し、果房の検出・セグメンテーションに加えて重量推定用の寸法と実重量を提供することが中心で、再利用可能な植物表現型データセットに該当する。
abstractthis paper proposed four different datasets on grape bunches in vineyard.
Reproduction assets foundThe paper is a data descriptor for GrapesNet, a public Mendeley Data repository of Indian vineyard RGB/RGB-D grape bunch image datasets with ground-truth cluster height, width, and weight measurements used for phenotyping tasks (detection, segmentation, weight estimation). The dataset is the paper's core asset and is aDataset · publicRepository name: GrapesNet: Indian Grape Clusters RGB & RGB-D Image Datasets
Data identification number (DOI): 10.17632/mhzmzd5cwx.1
Direct URL to data: https://data.mendeley.com/datasets/mhzmzd5cwx/1Open asset ↗10.17632/mhzmzd5cwx.1lines:1-95Code / dataset availability confirmedEurope PMC · Crossref · checked 7 Sept 2026
Recent studies indicate that food demand will increase by 35-56% over the period 2010-2050 due to population increase, economic development, and urbanization. Greenhouse systems allow for the sustainable intensification of food production with demonstrated high crop production per cultivation area. Breakthroughs in resource-efficient fresh food production merging horticultural and AI expertise take place with the international competition "Autonomous Greenhouse Challenge". This paper describes and analyzes the results of the third edition of this competition. The competition's goal is the realization of the highest net profit in fully autonomous lettuce production. Two cultivation cycles were conducted in six high-tech greenhouse compartments with operational greenhouse decision-making realized at a distance and individually by algorithms of international participating teams. Algorithms were developed based on time series sensor data of the greenhouse climate and crop images. High crop yield and quality, short growing cycles, and low use of resources such as energy for heating, electricity for artificial light, and CO 2 were decisive in realizing the competition's goal. The results highlight the importance of plant spacing and the moment of harvest decisions in promoting high crop growth rates while optimizing greenhouse occupation and resource use. In this paper, images taken with depth cameras (RealSense) for each greenhouse were used by computer vision algorithms (Deepabv3+ implemented in detectron2 v0.6) in deciding optimum plant spacing and the moment of harvest. The resulting plant height and coverage could be accurately estimated with an R 2 of 0.976, and a mIoU of 98.2, respectively. These two traits were used to develop a light loss and harvest indicator to support remote decision-making. The light loss indicator could be used as a decision tool for timely spacing. Several traits were combined for the harvest indicator, ultimately resulting in a fresh weight estimation with a mean absolute error of 22 g. The proposed non-invasively estimated indicators presented in this article are promising traits to be used towards full autonomation of a dynamic commercial lettuce growing environment. Computer vision algorithms act as a catalyst in remote and non-invasive sensing of crop parameters, decisive for automated, objective, standardized, and data-driven decision making. However, spectral indexes describing lettuces growth and larger datasets than the currently accessible are crucial to address existing shortcomings between academic and industrial production systems that have been encountered in this work.
Why it matches plant phenotyping methods深度カメラ画像とコンピュータビジョンによりレタスの草丈・被覆率・収量関連形質を推定し、精度評価と自動意思決定指標への応用を行っており、植物表現型取得法が中心である。
abstractimages taken with depth cameras (RealSense) for each greenhouse were used by computer vision algorithms (Deepabv3+ implemented in detectron2 v0.6) in deciding optimum plant spacing and the moment of harvest.
Reproduction assets foundThe paper's complete challenge dataset (climate time-series and annotated lettuce crop images used for the computer vision phenotyping) is published open access on 4TU.ResearchData, cited both in the Data Availability Statement and in reference 56.Dataset · public3rd Autonomous Greenhouse Challenge-Real Challenge Data Climate and Images Dataset: 4TU.ResearchData 2023 Available online: https://data.4tu.nl/articles/dataset/3rd_Autonomous_Greenhouse_Challenge_Online_Challenge_Lettuce_Images/15023088Open asset ↗4TU.ResearchData · 15023088lines:853-968Code / dataset availability confirmedEurope PMC · OpenAlex · checked 15 Sept 2026
Smart farming (SF) applications rely on robust and accurate computer vision systems. An important computer vision task in agriculture is semantic segmentation, which aims to classify each pixel of an image and can be used for selective weed removal. State-of-the-art implementations use convolutional neural networks (CNN) that are trained on large image datasets. In agriculture, publicly available RGB image datasets are scarce and often lack detailed ground-truth information. In contrast to agriculture, other research areas feature RGB-D datasets that combine color (RGB) with additional distance (D) information. Such results show that including distance as an additional modality can improve model performance further. Therefore, we introduce WE3DS as the first RGB-D image dataset for multi-class plant species semantic segmentation in crop farming. It contains 2568 RGB-D images (color image and distance map) and corresponding hand-annotated ground-truth masks. Images were taken under natural light conditions using an RGB-D sensor consisting of two RGB cameras in a stereo setup. Further, we provide a benchmark for RGB-D semantic segmentation on the WE3DS dataset and compare it with a solely RGB-based model. Our trained models achieve up to 70.7% mean Intersection over Union (mIoU) for discriminating between soil, seven crop species, and ten weed species. Finally, our work confirms the finding that additional distance information improves segmentation quality.
Why it matches plant phenotyping methods植物種の画素単位セグメンテーション用RGB-Dデータセットとベンチマークを構築し、植物識別・分離という表現型取得ワークフローを中心的に評価しているため。
abstractwe introduce WE3DS as the first RGB-D image dataset for multi-class plant species semantic segmentation in crop farming.
Reproduction assets foundThe paper's WE3DS RGB-D image dataset (2568 annotated images) and the authors' modified ESANet analysis code are publicly deposited on Zenodo (DOI 10.5281/zenodo.7457983), as stated in the experiments section. The MDPI supplementary file contains only tables (species list, depth accuracy, confusion matrices), not the影像Dataset · public024 × 512
20.6
27.0
37.7 †
34.2
11.5
RGB
1024 × 512
52.4
22.2
39.2 †
38.4
11.5
RGB-D
1024 × 512
59.1
19.2
85.8
55.3
18.5
D
1280 × 960
48.5
11.3
154.1
37.0
27.1
RGB
1280 × 960
70.1
11.0
156.1
46.8
27.1
RGB-D
1280 × 960
70.7
8.6
240.3
66.6
43.4
Information on the dataset and modified code of the ESANet can be found on our website https://doi.org/10.5281/zenodo.7457983 (accessed on 18 December 2022).
4.4. ResultsOpen asset ↗Zenodo · 10.5281/zenodo.7457983lines:69-146Code / dataset availability confirmedOpenAlex · checked 13 Sept 2026
Estimation of fruit size on-tree is useful for yield estimation, harvest timing and market planning. Automation of measurement of fruit size on-tree is possible using RGB-depth (RGB-D) cameras, if partly occluded fruit can be removed from consideration. An RGB-D Time of Flight camera was used in an imaging system that can be driven through an orchard. Three approaches were compared, being: (i) refined bounding box dimensions of a YOLO object detector; (ii) bounding box dimensions of an instance segmentation model (Mask R-CNN) applied to canopy images, and (iii) instance segmentation applied to extracted bounding boxes from a YOLO detection model. YOLO versions 3, 4 and 7 and their tiny variants were compared to an in-house variant, MangoYOLO, for this application, with YOLO v4-tiny adopted. Criteria developed to exclude occluded fruit by filtering based on depth, mask size, ellipse to mask area ratio and difference between refined bounding box height and ellipse major axis. The lowest root mean square error (RMSE) of 4.7 mm and 5.1 mm on the lineal length dimensions of a population (n = 104) of Honey Gold and Keitt varieties of mango fruit, respectively, and the lowest fruit exclusion rate was achieved using method (ii), while the RMSE on estimated fruit weight was 113 g on a population weight range between 180 and 1130 g. An example use is provided, with the method applied to video of an orchard row to produce a weight frequency distribution related to packing tray size.
Why it matches plant phenotyping methodsRGB-D画像と物体検出・インスタンスセグメンテーションを用いて樹上マンゴー果実のサイズ・重量を推定し、複数手法を比較検証しているため、植物表現型取得法が研究の中心である。
abstractThree approaches were compared, being: (i) refined bounding box dimensions of a YOLO object detector; (ii) bounding box dimensions of an instance segmentation model (Mask R-CNN) applied to canopy images, and (iii) instance segmentation applied to extracted bounding boxes from a YOLO detection model.
Reproduction assets foundThe paper publicly releases the RGB-D image datasets (Dataset-B and Dataset-C) used for training/testing the Mask R-CNN and YOLO-based mango fruit sizing models via a DOI deposit. No author analysis code or trained model checkpoints are explicitly deposited; the GitHub links cited are third-party frameworks (Darknet, MDataset · publicAll images in Dataset B and Dataset C used in
this study are available at https://doi.org/10.25946/21655628 (accessed on 15 October 2022).Open asset ↗10.25946/21655628pdf-page:4 lines:1-58Code / dataset availability confirmedEurope PMC · OpenAlex · checked 13 Sept 2026
Growth traits, such as fresh weight, diameter, and leaf area, are pivotal indicators of growth status and the basis for the quality evaluation of lettuce. The time-consuming, laborious and inefficient method of manually measuring the traits of lettuce is still the mainstream. In this study, a three-stage multi-branch self-correcting trait estimation network (TMSCNet) for RGB and depth images of lettuce was proposed. The TMSCNet consisted of five models, of which two master models were used to preliminarily estimate the fresh weight (FW), dry weight (DW), height (H), diameter (D), and leaf area (LA) of lettuce, and three auxiliary models realized the automatic correction of the preliminary estimation results. To compare the performance, typical convolutional neural networks (CNNs) widely adopted in botany research were used. The results showed that the estimated values of the TMSCNet fitted the measurements well, with coefficient of determination ( R 2 ) values of 0.9514, 0.9696, 0.9129, 0.8481, and 0.9495, normalized root mean square error (NRMSE) values of 15.63, 11.80, 11.40, 10.18, and 14.65% and normalized mean squared error (NMSE) value of 0.0826, which was superior to compared methods. Compared with previous studies on the estimation of lettuce traits, the performance of the TMSCNet was still better. The proposed method not only fully considered the correlation between different traits and designed a novel self-correcting structure based on this but also studied more lettuce traits than previous studies. The results indicated that the TMSCNet is an effective method to estimate the lettuce traits and will be extended to the high-throughput situation. Code is available at https://github.com/lxsfight/TMSCNet.git.
Why it matches plant phenotyping methodsRGB・深度画像からレタスの複数形質を推定する新規ネットワークを開発し、既存手法と性能比較しており、植物フェノタイピング手法が研究の中心である。
abstracta three-stage multi-branch self-correcting trait estimation network (TMSCNet) for RGB and depth images of lettuce was proposed
Reproduction assets foundThe paper uses the public Autonomous Greenhouses Challenge 3 dataset (RGB/depth lettuce images with FW/DW/H/D/LA measurements) and states author code availability on GitHub.Code · publicCode is available at https://github.com/lxsfight/TMSCNet.git .Open asset ↗github.com/lxsfight/TMSCNetlines:1-41Code / dataset availability confirmedEurope PMC · OpenAlex · checked 8 Sept 2026
Fresh weight is a widely used growth indicator for quantifying crop growth. Traditional fresh weight measurement methods are time-consuming, laborious, and destructive. Non-destructive measurement of crop fresh weight is urgently needed in plant factories with high environment controllability. In this study, we proposed a multi-modal fusion based deep learning model for automatic estimation of lettuce shoot fresh weight by utilizing RGB-D images. The model combined geometric traits from empirical feature extraction and deep neural features from CNN. A lettuce leaf segmentation network based on U-Net was trained for extracting leaf boundary and geometric traits. A multi-branch regression network was performed to estimate fresh weight by fusing color, depth, and geometric features. The leaf segmentation model reported a reliable performance with a mIoU of 0.982 and an accuracy of 0.998. A total of 10 geometric traits were defined to describe the structure of the lettuce canopy from segmented images. The fresh weight estimation results showed that the proposed multi-modal fusion model significantly improved the accuracy of lettuce shoot fresh weight in different growth periods compared with baseline models. The model yielded a root mean square error (RMSE) of 25.3 g and a coefficient of determination ( R 2 ) of 0.938 over the entire lettuce growth period. The experiment results demonstrated that the multi-modal fusion method could improve the fresh weight estimation performance by leveraging the advantages of empirical geometric traits and deep neural features simultaneously.
Why it matches plant phenotyping methodsRGB-D画像からレタスの生体重を非破壊推定する画像解析・深層学習手法の開発が研究の中心であり、植物表現型取得法に該当する。
abstractA lettuce leaf segmentation network based on U-Net was trained for extracting leaf boundary and geometric traits.
Reproduction assets foundThe paper's phenotyping inputs (top-view RGB and aligned depth images of 388 lettuces with destructively measured traits) come from the publicly available 3rd Autonomous Greenhouse Challenge Online Challenge Lettuce Images dataset, with an explicit public URL in the data availability statement. No author analysis code,Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://data.4tu.nl/articles/dataset/3rd_Autonomous_Greenhouse_Challenge_Online_Challenge_Lettuce_Images/15023088 .Open asset ↗data.4tu.nl · 15023088lines:657-691Code / dataset availability confirmedEurope PMC · OpenAlex · checked 13 Sept 2026
Growth indices can quantify crop productivity and establish optimal environmental, nutritional, and irrigation control strategies. A convolutional neural network (CNN)-based model is presented for estimating various growth indices (i.e., fresh weight, dry weight, height, leaf area, and diameter) of four varieties of greenhouse lettuce using red, green, blue, and depth (RGB-D) data obtained using a stereo camera. Data from an online autonomous greenhouse challenge (Wageningen University, June 2021) were employed in this study. The data were collected using an Intel RealSense D415 camera. The developed model has a two-stage CNN architecture based on ResNet50V2 layers. The developed model provided coefficients of determination from 0.88 to 0.95, with normalized root mean square errors of 6.09%, 6.30%, 7.65%, 7.92%, and 5.62% for fresh weight, dry weight, height, diameter, and leaf area, respectively, on unknown lettuce images. Using red, green, blue (RGB) and depth data employed in the CNN improved the determination accuracy for all five lettuce growth indices due to the ability of the stereo camera to extract height information on lettuce. The average time for processing each lettuce image using the developed CNN model run on a Jetson SUB mini-PC with a Jetson Xavier NX was 0.83 s, indicating the potential for the model in fast real-time sensing of lettuce growth indices.
Why it matches plant phenotyping methodsRGB-D画像とCNNを用いてレタスの複数の生育形質を推定する手法を開発・検証しており、表現型取得が研究の中心である。
abstractA convolutional neural network (CNN)-based model is presented for estimating various growth indices (i.e., fresh weight, dry weight, height, leaf area, and diameter) of four varieties of greenhouse lettuce using red, green, blue, and depth (RGB-D) data obtained using a stereo camera.
Reproduction assets foundThe paper's phenotyping inputs (388 RGB-D lettuce image pairs with destructive growth-index measurements from the Third Autonomous Greenhouse Challenge) are a third-party public dataset explicitly stated to be publicly available at 4TU.ResearchData, with the DOI 10.4121/15023088.v1 cited in the text and figure captionsDataset · publicThe dataset is available in online: https://doi.org/10.4121/15023088.v1 [ 30 ].Open asset ↗10.4121/15023088.v1lines:518-697Code / dataset availability confirmedCrossref · Europe PMC · OpenAlex · checked 8 Sept 2026
Plant breeders, scientists, and commercial producers commonly use growth rate as an integrated signal of crop productivity and stress. Plant growth monitoring is often done destructively via growth rate estimation by harvesting plants at different growth stages and simply weighing each individual plant. Within plant breeding and research applications, and more recently in commercial applications, non-destructive growth monitoring is done using computer vision to segment plants in images from the background, either in 2D or 3D, and relating these image-based features to destructive biomass measurements. Recent advancements in machine learning have improved image-based localization and detection of plants, but such techniques are not well suited to make biomass predictions when there is significant self-occlusion or occlusion from neighboring plants, such as those encountered under leafy green production in controlled environment agriculture. To enable prediction of plant biomass under occluded growing conditions, we develop an end-to-end deep learning approach that directly predicts lettuce plant biomass from color and depth image data as provided by a low cost and commercially available sensor. We test the performance of the proposed deep neural network for lettuce production, observing a mean prediction error of 7.3% on a comprehensive test dataset of 864 individuals and substantially outperforming previous work on plant biomass estimation. The modeling approach is robust to the busy and occluded scenes often found in commercial leafy green production and requires only measured mass values for training. We then demonstrate that this level of prediction accuracy allows for rapid, non-destructive detection of changes in biomass accumulation due to experimentally induced stress induction in as little as 2 days. Using this method growers may observe and react to changes in plant-environment interactions in near real time. Moreover, we expect that such a sensitive technique for non-destructive biomass estimation will enable novel research and breeding of improved productivity and yield in response to stress.
Why it matches plant phenotyping methodsRGB-D画像と深層学習を用いて、遮蔽下のレタス個体バイオマスを非破壊推定する手法を開発・評価しており、植物表現型取得が研究の中心です。
abstractwe develop an end-to-end deep learning approach that directly predicts lettuce plant biomass from color and depth image data as provided by a low cost and commercially available sensor.
Reproduction assets foundThe article provides an authors' public GitHub repository containing the analysis code for the RGB-D deep learning biomass estimation pipeline. The raw image/biomass dataset is only available on request (no public deposit).Code · publicCode available at https://github.com/NicoBux/Plant-Biomass-Monitoring .Open asset ↗NicoBux/Plant-Biomass-Monitoringlines:466-524Code / dataset availability confirmedarXiv · checked 13 Sept 2026
Obtaining 3D sensor data of complete plants or plant parts (e.g., the crop or fruit) is difficult due to their complex structure and a high degree of occlusion. However, especially for the estimation of the position and size of fruits, it is necessary to avoid occlusions as much as possible and acquire sensor information of the relevant parts. Global viewpoint planners exist that suggest a series of viewpoints to cover the regions of interest up to a certain degree, but they usually prioritize global coverage and do not emphasize the avoidance of local occlusions. On the other hand, there are approaches that aim at avoiding local occlusions, but they cannot be used in larger environments since they only reach a local maximum of coverage. In this paper, we therefore propose to combine a local, gradient-based method with global viewpoint planning to enable local occlusion avoidance while still being able to cover large areas. Our simulated experiments with a robotic arm equipped with a camera array as well as an RGB-D camera show that this combination leads to a significantly increased coverage of the regions of interest compared to just applying global coverage planning.
Why it matches plant phenotyping methods果実の位置・サイズ推定に必要な3Dセンサデータ取得を対象に、局所遮蔽回避と大域的視点計画を組み合わせる視点計画法を開発・評価しており、植物表現型取得が中心である。
abstractespecially for the estimation of the position and size of fruits, it is necessary to avoid occlusions as much as possible and acquire sensor information of the relevant parts
Reproduction assets foundThe paper's authors explicitly state that the source code of their combined local/global viewpoint planning system (used for fruit ROI coverage experiments) is publicly available on GitHub. OctoMap is a generic third-party library and is excluded.Code · publicThe source code of our system is available on GitHub 1 1
1
https://github.com/Eruvae/roi_viewpoint_planner .Open asset ↗Eruvae/roi_viewpoint_plannerlines:1-105Code / dataset availability confirmedEurope PMC · OpenAlex · checked 14 Sept 2026
The growth and the harvestability of a broccoli crop is monitored by the size of the broccoli head. This size estimation is currently done by humans, and this is inconsistent and expensive. The goal of our work was to develop a software algorithm that can estimate the size of field-grown broccoli heads based on RGB-Depth (RGB-D) images. For the algorithm to be successful, the problem of occlusion must be solved, which is the partial visibility of the broccoli head due to overlapping leaves. This partial visibility causes sizing errors. In this research, we studied the use of deep-learning algorithms to deal with occlusions. We specifically applied the Occlusion Region-based Convolutional Neural Network (ORCNN) that segmented both the visible and the amodal region of the broccoli head (which is the visible and the occluded region combined). We hypothesised that ORCNN, with its amodal segmentation, can improve the size estimation of occluded broccoli heads. The ORCNN sizing method was compared with a Mask R–CNN sizing method that only used the visible broccoli region to estimate the size. The sizing performance of both methods was evaluated on a test set of 487 broccoli images with systematic levels of leaf occlusion. With a mean sizing error of 6.4 mm, ORCNN outperformed Mask R–CNN, which had a mean sizing error of 10.7 mm. Furthermore, ORCNN had a significantly lower absolute sizing error on 161 heavily occluded broccoli heads with an occlusion rate between 50% and 90%. Our software and data set are available on https://git.wur.nl/blok012/sizecnn.
Why it matches plant phenotyping methodsRGB-D画像と深層学習によりブロッコリー頭部サイズを推定する手法を開発し、Mask R-CNNと比較検証しているため、植物形質取得法が研究の中心です。
abstractThe goal of our work was to develop a software algorithm that can estimate the size of field-grown broccoli heads based on RGB-Depth (RGB-D) images.
The use of 3D sensors combined with appropriate data processing and analysis has provided tools to optimise agricultural management through the application of precision agriculture. The recent development of low-cost RGB-Depth cameras has presented an opportunity to introduce 3D sensors into the agricultural community. However, due to the sensitivity of these sensors to highly illuminated environments, it is necessary to know under which conditions RGB-D sensors are capable of operating. This work presents a methodology to evaluate the performance of RGB-D sensors under different lighting and distance conditions, considering both geometrical and spectral (colour and NIR) features. The methodology was applied to evaluate the performance of the Microsoft Kinect v2 sensor in an apple orchard. The results show that sensor resolution and precision decreased significantly under middle to high ambient illuminance (>2000 lx). However, this effect was minimised when measurements were conducted closer to the target. In contrast, illuminance levels below 50 lx affected the quality of colour data and may require the use of artificial lighting. The methodology was useful for characterizing sensor performance throughout the full range of ambient conditions in commercial orchards. Although Kinect v2 was originally developed for indoor conditions, it performed well under a range of outdoor conditions.
Why it matches plant phenotyping methodsRGB-Dセンサーの性能を、果樹キャノピーの3D形状・色・NIR特徴の取得という植物フェノタイピング用途で、照明・距離条件下で評価する方法論が中心である。
abstractThis work presents a methodology to evaluate the performance of RGB-D sensors under different lighting and distance conditions, considering both geometrical and spectral (colour and NIR) features.
Reproduction assets foundThe paper's Kinect Evaluation in Orchard conditions (KEvOr) dataset of RGB/NIR/point-cloud captures from an apple orchard is publicly deposited on Zenodo, and the authors' MATLAB analysis code for the sensor evaluation is publicly available on GitHub. Both are paper-specific, public, and actionable.Code · publicA MATLAB® (R2020a, Math Works Inc., Natick, MA, USA) code was developed to analyse all
the data and provide the sensor evaluation results. This code has been made publicly available at
https://github.com/GRAP-UdL-AT/RGBD_sensors_evaluation_in_Orchards [35].Open asset ↗GitHub · GRAP-UdL-AT/RGBD_sensors_evaluation_in_Orchardspdf-page:6 lines:1-60Code / dataset availability confirmedarXiv · checked 13 Sept 2026
Modern agricultural applications require knowledge about the position and size of fruits on plants. However, occlusions from leaves typically make obtaining this information difficult. We present a novel viewpoint planning approach that builds up an octree of plants with labeled regions of interest (ROIs), i.e., fruits. Our method uses this octree to sample viewpoint candidates that increase the information around the fruit regions and evaluates them using a heuristic utility function that takes into account the expected information gain. Our system automatically switches between ROI targeted sampling and exploration sampling, which considers general frontier voxels, depending on the estimated utility. When the plants have been sufficiently covered with the RGB-D sensor, our system clusters the ROI voxels and estimates the position and size of the detected fruits. We evaluated our approach in simulated scenarios and compared the resulting fruit estimations with the ground truth. The results demonstrate that our combined approach outperforms a sampling method that does not explicitly consider the ROIs to generate viewpoints in terms of the number of discovered ROI cells. Furthermore, we show the real-world applicability by testing our framework on a robotic arm equipped with an RGB-D camera installed on an automated pipe-rail trolley in a capsicum glasshouse.
Why it matches plant phenotyping methods果実の位置・サイズという植物器官形質をRGB-Dセンサで取得・推定する視点計画法を開発し、シミュレーションと実環境で検証しており、フェノタイピング手法が中心である。
abstractWe present a novel viewpoint planning approach that builds up an octree of plants with labeled regions of interest (ROIs), i.e., fruits.
Reproduction assets foundThe paper's viewpoint-planning system source code and the simulated capsicum plant environments used in the experiments are publicly available on GitHub. OctoMap is a generic third-party library, not a paper-specific asset.Code · publicThe source code of our system is available on GitHub 1 1
1
https://github.com/Eruvae/roi_viewpoint_planner .Open asset ↗Eruvae/roi_viewpoint_plannerlines:73-113Code / dataset availability confirmedOpenAlex · Europe PMC · Crossref · checked 9 Sept 2026
BACKGROUND: Assessing the impact of the environment on plant performance requires growing plants under controlled environmental conditions. Plant phenotypes are a product of genotype × environment (G × E), and the Enviratron at Iowa State University is a facility for testing under controlled conditions the effects of the environment on plant growth and development. Crop plants (including maize) can be grown to maturity in the Enviratron, and the performance of plants under different environmental conditions can be monitored 24 h per day, 7 days per week throughout the growth cycle. RESULTS: The Enviratron is an array of custom-designed plant growth chambers that simulate different environmental conditions coupled with precise sensor-based phenotypic measurements carried out by a robotic rover. The rover has workflow instructions to periodically visit plants growing in the different chambers where it measures various growth and physiological parameters. The rover consists of an unmanned ground vehicle, an industrial robotic arm and an array of sensors including RGB, visible and near infrared (VNIR) hyperspectral, thermal, and time-of-flight (ToF) cameras, laser profilometer and pulse-amplitude modulated (PAM) fluorometer. The sensors are autonomously positioned for detecting leaves in the plant canopy, collecting various physiological measurements based on computer vision algorithms and planning motion via "eye-in-hand" movement control of the robotic arm. In particular, the automated leaf probing function that allows the precise placement of sensor probes on leaf surfaces presents a unique advantage of the Enviratron system over other types of plant phenotyping systems. CONCLUSIONS: The Enviratron offers a new level of control over plant growth parameters and optimizes positioning and timing of sensor-based phenotypic measurements. Plant phenotypes in the Enviratron are measured in situ-in that the rover takes sensors to the plants rather than moving plants to the sensors.
Why it matches plant phenotyping methodsロボット rover、複数センサー、コンピュータビジョン、葉面プロービングを統合した植物フェノタイピング基盤の開発・実証が中心である。
abstractThe Enviratron is an array of custom-designed plant growth chambers that simulate different environmental conditions coupled with precise sensor-based phenotypic measurements carried out by a robotic rover.
Reproduction assets foundThe paper describes the Enviratron phenotyping facility and explicitly states that a Git repository containing the code developed to operate and support the Enviratron is publicly available on GitLab. This is an authors' public code asset directly tied to this paper's phenotyping system. No phenotype datasets or image/Code · publicA git repository containing code developed to operate and support the Enviratron is available online at https://gitlab.com/dill_picl/enviratron .Open asset ↗dill_picl/enviratronlines:165-207Code / dataset availability confirmedEurope PMC · checked 11 Sept 2026
Dissecting the genetic basis of complex traits is aided by frequent and nondestructive measurements. Advances in range imaging technologies enable the rapid acquisition of three-dimensional (3D) data from an imaged scene. A depth camera was used to acquire images of sorghum (Sorghum bicolor), an important grain, forage, and bioenergy crop, at multiple developmental time points from a greenhouse-grown recombinant inbred line population. A semiautomated software pipeline was developed and used to generate segmented, 3D plant reconstructions from the images. Automated measurements made from 3D plant reconstructions identified quantitative trait loci for standard measures of shoot architecture, such as shoot height, leaf angle, and leaf length, and for novel composite traits, such as shoot compactness. The phenotypic variability associated with some of the quantitative trait loci displayed differences in temporal prevalence; for example, alleles closely linked with the sorghum Dwarf3 gene, an auxin transporter and pleiotropic regulator of both leaf inclination angle and shoot height, influence leaf angle prior to an effect on shoot height. Furthermore, variability in composite phenotypes that measure overall shoot architecture, such as shoot compactness, is regulated by loci underlying component phenotypes like leaf angle. As such, depth imaging is an economical and rapid method to acquire shoot architecture phenotypes in agriculturally important plants like sorghum to study the genetic basis of complex traits.
Why it matches plant phenotyping methods深度画像から3D植物再構成と形質自動計測を行う半自動パイプラインを開発し、ソルガムの草型形質を取得する方法が研究の中心である。
abstractA semiautomated software pipeline was developed and used to generate segmented, 3D plant reconstructions from the images.
Reproduction assets foundThe authors explicitly deposit their image acquisition/processing and QTL mapping code (C++, Bash, Python, R) plus genotype/phenotype data on GitHub, and per-plant depth images, RGB images, and segmented meshes on Dryad. Both are paper-specific, public, and actionable.Code · publicThe C++, Bash, and Python code written for image acquisition and processing, the R code written for QTL mapping, the genotype and phenotype data, and the full multiple- QTL models for each phenotype-by-time point combination can be found on GitHub at https://github.com/MulletLab/SorghumReconstructionAndPhenotyping .Open asset ↗MulletLab/SorghumReconstructionAndPhenotypinglines:240-292Dataset · publicFor each imaged plant, its depth images, a single RGB image, and the segmented mesh can be found at the Dryad Digital Repository ( http://dx.doi.org/10.5061/dryad.9vs26 ).Open asset ↗Dryad Digital Repository · 10.5061/dryad.9vs26lines:240-292Code / dataset availability confirmedbioRxiv · Europe PMC · Crossref · checked 11 Sept 2026
Dissecting the genetic basis of complex traits is aided by frequent and non-destructive measurements. Advances in range imaging technologies enable the rapid acquisition of three-dimensional (3D) data from an imaged scene. A depth camera was used to acquire images of Sorghum bicolor, an important grain, forage, and bioenergy crop, at multiple developmental timepoints from a greenhouse-grown recombinant inbred line population. A semi-automated software pipeline was developed and used to generate segmented, 3D plant reconstructions from the images. Automated measurements made from 3D plant reconstructions identified quantitative trait loci (QTL) for standard measures of shoot architecture such as shoot height, leaf angle and leaf length, and for novel composite traits such as shoot compactness. The phenotypic variability associated with some of the QTL displayed differences in temporal prevalence; for example, alleles closely linked with the sorghum Dwarf3 gene, an auxin transporter and pleiotropic regulator of both leaf inclination angle and shoot height, influence leaf angle prior to an effect on shoot height. Furthermore, variability in composite phenotypes that measure overall shoot architecture, such as shoot compactness, is regulated by loci underlying component phenotypes like leaf angle. As such, depth imaging is an economical and rapid method to acquire shoot architecture phenotypes in agriculturally important plants like sorghum to study the genetic basis of complex traits.
Why it matches plant phenotyping methods深度画像から3D植物再構成と形質自動抽出を行う半自動パイプラインを開発し、ソルガムのシュート構造形質を取得・評価しており、表現型取得法が研究の中心です。
abstractA semi-automated software pipeline was developed and used to generate segmented, 3D plant reconstructions from the images.
Reproduction assets foundThe paper explicitly deposits its authors' image acquisition/processing and QTL mapping code on GitHub, and its per-plant depth images, RGB images, and segmented meshes on the Dryad repository. Both are paper-specific, public, and actionable.Code · publicThe C++, Bash, and Python code written for image acquisition and processing, the R code written for
QTL mapping, the genotype and phenotype data, and the full multiple-QTL models for each phenotype
by timepoint combination can be found on GitHub at
https://github.com/MulletLab/SorghumReconstructionAndPhenotyping.Open asset ↗MulletLab/SorghumReconstructionAndPhenotypingpdf-page:8 lines:1-43