将成熟拓扑序列嵌入单阶段检测用于串番茄表型分析

楚秉泉; 吴瑞源; 张海军; 秦浩川; 彭子舜; 朱逢乐; 何勇

doi:10.1631/jzus.B2500647

Your Location：

Home >

Browse articles >

将成熟拓扑序列嵌入单阶段检测用于串番茄表型分析

Scan QR Code

文章被引用时，请邮件提醒。

Submit

Research Article | Updated：2026-05-15

将成熟拓扑序列嵌入单阶段检测用于串番茄表型分析
Embedding of ripening topology into one-stage detection for tomato cluster phenotyping
浙江大学学报（英文版）（B辑：生物医学和生物技术） 2026年27卷第5期页码：466-481
Affiliations：

1.School of Biological and Chemical Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, China
2.College of Mechanical Engineering, Zhejiang University of Technology, Hangzhou 310023, China
3.College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
Funds：

the Science and Technology Program of the Ministry of Agriculture and Rural Affairs of the People’s Republic of China
DOI：10.1631/jzus.B2500647
中图分类号：
CSTR：
收稿：2025-10-14，

修回：2026-03-12，

纸质出版：2026-05-15
Accepted：

楚秉泉,吴瑞源,张海军等.将成熟拓扑序列嵌入单阶段检测用于串番茄表型分析[J].浙江大学学报（英文版）（B辑：生物医学和生物技术）,2026,27(05):466-481.

Bingquan CHU, Ruiyuan WU, Haijun ZHANG, et al. Embedding of ripening topology into one-stage detection for tomato cluster phenotyping[J]. Journal of Zhejiang University-SCIENCE B, 2026, 27(5): 466-481.
楚秉泉,吴瑞源,张海军等.将成熟拓扑序列嵌入单阶段检测用于串番茄表型分析[J].浙江大学学报（英文版）（B辑：生物医学和生物技术）,2026,27(05):466-481. DOI： 10.1631/jzus.B2500647.

Bingquan CHU, Ruiyuan WU, Haijun ZHANG, et al. Embedding of ripening topology into one-stage detection for tomato cluster phenotyping[J]. Journal of Zhejiang University-SCIENCE B, 2026, 27(5): 466-481. DOI： 10.1631/jzus.B2500647.

摘要

番茄成熟度的自动化评估对现代温室作业至关重要，但多变的环境条件为其准确实现带来了持续挑战。为此，本文提出了rank-aware YOLO，一种融合果实簇内自上而下成熟这一生物学先验知识的新型检测框架。该框架通过两项关键创新实现：（1）用于回归果实相对高度的高效位置敏感检测头（efficient position-aware head）；（2）修正空间序列的动态边距感知排序损失（DM-RankLoss）。在包含3500张温室采集图像的数据集上进行评估的结果表明，该模块具有良好的即插即用特性，能将多种YOLO架构的在交并比（IoU）阈值为0.50时的平均精度均值（mAP

）最高提升5.66个百分点。模型有效学习到果实簇的拓扑结构，在归一化高度平均绝对误差（H-MAE）和成对排序准确率上分别达到0.107和84.59%，同时参数量较基线减少超过10%，具备高效部署潜力。可视化分析进一步证实，模型能利用空间上下文信息有效缓解颜色模糊带来的误判。综上，本研究为农业机器人中的原位表型分析提供了一种无需额外传感器的准确且高效的解决方案。

Abstract

The automated assessment of tomato ripeness is vi

tal for modern greenhouse operations

yet challenges remain due to variable environmental conditions. To provide a solution

we propose rank-aware You Only Look Once (YOLO)

a novel detection framework that incorporates the biological prior of top-to-bottom ripening within fruit clusters. This is achieved through two key innovations: an efficient position-aware head for regressing relative height for fruits and a dynamic margin-aware ranking loss (DM-RankLoss) that enforces the correct spatial sequence. Evaluated on a 3500-image dataset from a solar greenhouse

our plug-and-play module could boost the mean average precision (mAP) at intersection over union (IoU) threshold of 0.50 (mAP

) of multiple YOLO architectures by up to 5.66 pecentage points. The model effectively learns the cluster topology

achieving a height-mean absolute error (H-MAE) of 0.107 (normalized) and a pairwise ranking accuracy (PRA) of 84.59%

while it reduces the parameter count by over 10% compared to the baseline for efficient deployment. Visualizations confirm that the model leverages spatial context to resolve color ambiguities. Our work offers a sensor-free

accurate

and efficient solution for in situ phenotyping in agricultural robotics.

关键词

Keywords

references

Albahar M , 2023 . A survey on deep learning and its impact on agriculture: challenges and opportunities . Agriculture , 13 ( 3 ): 540 . https://doi.org/10.3390/agriculture13030540 https://doi.org/10.3390/agriculture13030540

All China Federation of Supply and Marketing Cooperatives , 2021 . Tomato, GH/T 1193-2021 . All China Federation of Supply and Marketing Cooperatives , China .

Banerjee S , Magee L , Wang DK , et al. , 2020 . Semantic segmentation of microscopic neuroanatomical data by combining topological priors with encoder‒decoder deep networks . Nat Mach Intell , 2 ( 10 ): 585 - 594 . https://doi.org/10.1038/s42256-020-0227-9 https://doi.org/10.1038/s42256-020-0227-9

Burges C , Shaked T , Renshaw E , et al. , 2005 . Learning to rank using gradient descent . Proceedings of the 22nd International Conference on Machine Learning. Bonn , Germany , p. 89 - 96 . https://doi.org/10.1145/1102351.1102363 https://doi.org/10.1145/1102351.1102363

Burges CJC , 2010 . From RankNet to LambdaRank to LambdaMART: an overview . Microsoft Research Technical Report , MSR-TR-2010- 82 . Available from: https://www.microsoft.com/en-us/research/publication/from-ranknet-to-lambdarank-to-lambdamart-an-overview https://www.microsoft.com/en-us/research/publication/from-ranknet-to-lambdarank-to-lambdamart-an-overview

Cao WZ , Mirjalili V , Raschka S , 2020 . Rank consistent ordinal regression for neural networks with application to age estimation . Pattern Recogn Lett , 140 : 325 - 331 . https://doi.org/10.1016/j.patrec.2020.11.008 https://doi.org/10.1016/j.patrec.2020.11.008

Chen WB , Liu MC , Zhao CJ , et al. , 2024 . MTD-YOLO: multi-task deep convolutional neural network for cherry tomato fruit bunch maturity detection . Comput Electron Agric , 216 : 108533 . https://doi.org/10.1016/j.compag.2023.108533 https://doi.org/10.1016/j.compag.2023.108533

Chen WJ , Rao Y , Wang FY , et al. , 2024 . MLP-based multimodal tomato detection in complex scenarios: insights from task-specific analysis of feature fusion architectures . Comput Electron Agric , 221 : 108951 . https://doi.org/10.1016/j.compag.2024.108951 https://doi.org/10.1016/j.compag.2024.108951

Chu BQ , Guo ZY , Liu BJ , et al. , 2025 . Fast detection of rice striped stem borer ( Chilo suppressalis ) stress based on UAV sensor and multimodal segmentation method . Plant Growth Regul , 105 ( 4 ): 1057 - 1071 . https://doi.org/10.1007/s10725-025-01320-8 https://doi.org/10.1007/s10725-025-01320-8

Dao T , Fu DY , Ermon S , et al. , 2022 . FlashAttention: fast and memory-efficient exact attention with IO-awareness . arXiv :2205. 14135 . https://doi.org/10.48550/arXiv.2205.14135 https://doi.org/10.48550/arXiv.2205.14135

Deng JK , Guo J , Xue NN , et al. , 2019 . ArcFace: additive angular margin loss for deep face recognition . 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach , CA, USA , p. 4685 - 4694 . https://doi.org/10.1109/CVPR.2019.00482 https://doi.org/10.1109/CVPR.2019.00482

Díaz R , Marathe A , 2019 . Soft labels for ordinal regression . 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach , CA, USA , p. 4733 - 4742 . https://doi.org/10.1109/CVPR.2019.00487 https://doi.org/10.1109/CVPR.2019.00487

Fass E , Shlomi E , Ziv C , et al. , 2025 . Machine learning models based on hyperspectral imaging for pre-harvest tomato fruit quality monitoring . Comput Electron Agric , 229 : 109788 . https://doi.org/10.1016/j.compag.2024.109788 https://doi.org/10.1016/j.compag.2024.109788

Food and Agriculture Organization of the United Nations (FAO) , 2023 . Crops and livestock products . https://www.fao.org/faostat/en/#data/QCL https://www.fao.org/faostat/en/#data/QCL [ Accessed on Oct. 1, 2025 ] .

Fu H , Gong MM , Wang CH , et al. , 2018 . Deep ordinal regression network for monocular depth estimation . 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City , UT, USA , p. 2002 - 2011 . https://doi.org/10.1109/CVPR.2018.00214 https://doi.org/10.1109/CVPR.2018.00214

Gautier H , Rocci A , Buret M , et al. , 2005 . Fruit load or fruit position alters response to temperature and subsequently cherry tomato quality . J Sci Food Agric , 85 ( 6 ): 1009 - 1016 . https://doi.org/10.1002/jsfa.2060 https://doi.org/10.1002/jsfa.2060

Huang YG , Wang YH , Tai Y , et al. , 2020 . CurricularFace: adaptive curriculum learning loss for deep face recognition . 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle , WA, USA , p. 5900 - 5909 . https://doi.org/10.1109/CVPR42600.2020.00594 https://doi.org/10.1109/CVPR42600.2020.00594

Khan Z , Shen Y , Liu H , 2025 . ObjectDetection in agriculture: a comprehensive review of methods, applications, challenges, and future directions . Agriculture , 15 ( 13 ): 1351 . https://doi.org/10.3390/agriculture15131351 https://doi.org/10.3390/agriculture15131351

Krizhevsky A , Sutskever I , Hinton GE , 2017 . ImageNet classification with deep convolutional neural networks . Commun ACM , 60 ( 6 ): 84 - 90 . https://doi.org/10.1145/3065386 https://doi.org/10.1145/3065386

Kumar P , Bhat KM , Shenvi Nadkarni VB , et al. , 2024 . GLiDR: topologically regularized graph generative network for sparse LiDAR point clouds . 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle , WA, USA , p. 15152 - 15161 . https://doi.org/10.1109/CVPR52733.2024.01435 https://doi.org/10.1109/CVPR52733.2024.01435

Li RZ , Ji ZJ , Hu SK , et al. , 2023 . Tomato maturity recognition model based on improved YOLOv5 in greenhouse . Agronomy , 13 ( 2 ): 603 . https://doi.org/10.3390/agronomy13020603 https://doi.org/10.3390/agronomy13020603

Li XX , Chen WB , Wang YQ , et al. , 2023 . Design and experiment of an automatic cherry tomato harvesting system based on cascade vision detection . Trans Chin Soc Agric Eng , 39 ( 1 ): 136 - 145 (in Chinese) . https://doi.org/10.11975/j.issn.1002-6819.202210099 https://doi.org/10.11975/j.issn.1002-6819.202210099

Sampaio GS , Silva LA , Marengoni M , 2021 . 3D reconstruction of non-rigid plants and sensor data fusion for agriculture phenotyping . Sensors , 21 ( 12 ): 4115 . https://doi.org/10.3390/s21124115 https://doi.org/10.3390/s21124115

Su M , Zhou D , Yun YZ , et al. , 2025 . Design and implementation of a high-throughput field phenotyping robot for acquiring multisensor data in wheat . Plant Phenomics , 7 ( 2 ): 100014 . https://doi.org/10.1016/j.plaphe.2025.100014 https://doi.org/10.1016/j.plaphe.2025.100014

Vondrick C , Khosla A , Pirsiavash H , et al. , 2016 . Visualizing object detection features . Int J Comput Vision , 119 ( 2 ): 145 - 158 . https://doi.org/10.1007/s11263-016-0884-7 https://doi.org/10.1007/s11263-016-0884-7

Wang AC , Qian WH , Li A , et al. , 2024 . NVW-YOLOv8s: an improved YOLOv8s network for real-time detection and segmentation of tomato fruits at different ripeness stages . Comput Electron Agric , 219 : 108833 . https://doi.org/10.1016/j.compag.2024.108833 https://doi.org/10.1016/j.compag.2024.108833

Wang XR , Guo LQ , Wang XY , et al. , 2025 . SoftShadow: leveraging soft masks for penumbra-aware shadow removal . 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville , TN, USA , p. 23217 - 23226 . https://doi.org/10.1109/CVPR52734.2025.02162 https://doi.org/10.1109/CVPR52734.2025.02162

Wang Z , Ling YM , Wang XL , et al. , 2022 . An improved Faster R-CNN model for multi-object tomato maturity detection in complex scenarios . Ecol Inform , 72 : 101886 . https://doi.org/10.1016/j.ecoinf.2022.101886 https://doi.org/10.1016/j.ecoinf.2022.101886

Xiao F , Wang HB , Xu YQ , et al. , 2023 . Fruit detection and recognition based on deep learning for automatic harvesting: an overview and review . Agronomy , 13 ( 6 ): 1625 . https://doi.org/10.3390/agronomy13061625 https://doi.org/10.3390/agronomy13061625

Yao J , Ke XB , G XY u et al. , 2025 . Optimized substrate selection for enhanced orchid growth based on high-throughput lysimetric arrays . J Zhejiang Univ-Sci B, online first. https://doi.org/10.1631/jzus.B2500195 https://doi.org/10.1631/jzus.B2500195

Zhang XB , Hu Y , Chen W , et al. , 2021 . 3D brain glioma segmentation in MRI through integrating multiple densely connected 2D convolutional neural networks . J Zhejiang Univ-Sci B (Biomed & Biotechnol) , 22 ( 6 ): 462 - 475 . https://doi.org/10.1631/jzus.B2000381 https://doi.org/10.1631/jzus.B2000381

浏览量

Downloads

CSCD

工具集

关联资源

Secondary manifestations of mitochondrial disorders

水生态系统中内分泌干扰物对水生生物和人类的影响及检测修复技术进展

耐盐植物促生菌Enterobacter sp. Av16和Acinetobacter sp. Av23减轻盐胁迫对白麻（Apocynum pictum）种子萌发和幼苗光合作用的抑制

施耐德膜来源间充质干细胞在上颌窦底提升模型与颅骨缺损模型中的再生潜能

RNA甲基转移酶HENMT1在生物学和疾病中的作用