1.中国农业大学工学院,北京,100083
2.中国农业科学院油料作物研究所/农业农村部油料作物生物学与遗传育种重点实验室/农业农村部生物毒素检测重点实验室/农业农村部油料产品质量安全风险评估实验室/农业农村部油料及制品质量检验测试中心,湖北 武汉,430062
3.湖北大学,湖北 武汉,430062
赵亚(1996- ),女,博士研究生,研究方向为计算机视觉与图像处理,E-mail: zhaoya1109@163.com;§
刘兆东(1999- ),男,硕士研究生,研究方向为计算机视觉与图像处理,E-mail: 420717273@qq.com
王伟(1975- ),男,教授,研究方向为图像与信息处理,E-mail: playerww@cau.edu.cn;
收稿:2026-03-06,
纸质出版:2026-08-25
移动端阅览
赵亚,刘兆东,张文等.基于智能手机与深度学习的田间花生根瘤表型高通量数字化解析方法[J].中国油料作物学报,2026,48(04):236-247.
ZHAO Ya,LIU Zhao-dong,ZHANG Wen,et al.High-throughput digital phenotyping of peanut nodules in field using smartphones and deep learning[J].Chinese Journal of Oil Crop Sciences,2026,48(04):236-247.
赵亚,刘兆东,张文等.基于智能手机与深度学习的田间花生根瘤表型高通量数字化解析方法[J].中国油料作物学报,2026,48(04):236-247. DOI: 10.19802/j.issn.1007-9084.2026074.
ZHAO Ya,LIU Zhao-dong,ZHANG Wen,et al.High-throughput digital phenotyping of peanut nodules in field using smartphones and deep learning[J].Chinese Journal of Oil Crop Sciences,2026,48(04):236-247. DOI: 10.19802/j.issn.1007-9084.2026074.
针对花生根瘤微小密集、易粘连遮挡、田间成像条件复杂以及不同手机设备间图像差异较大等问题,建立了基于智能手机的花生根瘤表型田间快速解析方法。通过搭建封闭式白光定标采集装置,集成标准色卡与刻度标尺,实现根系图像颜色与尺度的标准化采集,从源头提高田间图像获取的一致性与可比性。提出适用于复杂田间背景的轻量化根瘤检测模型STAC-YOLO11n,可在手机端实现根瘤图像的快速分析与计数。结果表明,方法在复杂环境下具有较好的根瘤检出能力和计数稳定性,准确率为86.85%,且在不同手机设备的泛化能力较好。该研究为花生根瘤田间高通量、低成本、可复现表型获取提供了技术支撑,也为深入解析根瘤固氮机制、精准育种提供了新路径。
For high-throughput
low-cost
and reproducible field phenotyping of peanut nodules
and for a new avenue in deep investigation of nodulation and nitrogen fixation mechanisms on precision breeding
this study developed a smartphone-based rapid field phenotyping method. It is to address the challenges associated with peanut nodule phenotyping in field
including small size nodules and dense distribution
frequent adhesion and occlusion
complexity of field imaging conditions
and substantial image variability across smartphone devices. A closed white-light calibrated imaging device integrating a standard color card and scale ruler was constructed to enable standardized acquisition of root images in terms of both color and scale
thereby improving the consistency and comparability of field images at the source. In addition
a lightweight nodule detection model
STAC-YOLO11n
was proposed for complex field backgrounds
enabling rapid image analysis and nodule counting on smartphones. The results demonstrated that the proposed method achieved robust nodule detection and stable counting performance under complex environmental conditions
with an accuracy of 86.85%
and exhibited good cross-device generalization across different smartphone platforms.
Lin J S , Bjørk P K , Kolte M V , et al . Zinc mediates control of nitrogen fixation via transcription factor filamentation [J]. Nature , 2024 , 631 ( 8019 ): 164 - 169 . DOI: 10.1038/s41586-024-07607-6 http://dx.doi.org/10.1038/s41586-024-07607-6 .
Zubrod M . Counting the nodules that count, relationships between seed nitrogen and root nodules [J]. Nat Sci Educ , 2022 , 51 ( 2 ): e20088 . DOI: 10.1002/nse2.20088 http://dx.doi.org/10.1002/nse2.20088 .
Wang X Y , Yu N P , Sun Y Z , et al . An efficient and low-cost deep learning-based method for counting and sizing soybean nodules [J]. Agronomy , 2024 , 14 ( 9 ): 2041 . DOI: 10.3390/agronomy14092041 http://dx.doi.org/10.3390/agronomy14092041 .
Dias P A S , Santos Melo P G , Santos Pereira H , et al . Automated and estimation methods to count nodule number in common bean [J]. Aust J Crop Sci , 2017 , 11 ( 8 ): 968 - 973 . DOI: 10.21475/ajcs.17.11.08.pne452 http://dx.doi.org/10.21475/ajcs.17.11.08.pne452 .
李培武 , 岳晓凤 , 张奇 , 等 . 一种促进豆科作物增加根瘤数量与根瘤固氮酶活性的微生物菌剂及其应用 [P]. CN 113980854A , 2022-01-28 .
赵亚 , 张文 , 王督 , 等 . ARC微生物菌剂缓解黑暗胁迫下花生根瘤功能衰退的效应 [J]. 中国农业科学 , 2025 , 58 ( 22 ): 4617 - 4627 . DOI: 10.3864/j.issn.0578-1752.2025.22.005 http://dx.doi.org/10.3864/j.issn.0578-1752.2025.22.005 .
周扬 , 岳晓凤 , 唐晓倩 , 等 . 黄曲霉毒素阻控与花生超级结瘤耦合效应研究初报 [J]. 中国油料作物学报 , 2021 , 43 ( 6 ): 947 - 960 . DOI: 10.19802/j.issn.1007-9084.2021300 http://dx.doi.org/10.19802/j.issn.1007-9084.2021300 .
Das Choudhury S , Samal A , Awada T . Leveraging image analysis for high-throughput plant phenotyping [J]. Front Plant Sci , 2019 , 10 : 508 . DOI: 10.3389/fpls.2019.00508 http://dx.doi.org/10.3389/fpls.2019.00508 .
Woo D , Ghimire A , Jeong S , et al . Soybean root image dataset and its deep learning application for nodule segmentation [J]. Comput Electron Agric , 2023 , 215 : 108465 . DOI: 10.1016/j.compag.2023.108465 http://dx.doi.org/10.1016/j.compag.2023.108465 .
Hui F , Xie Z W , Li H G , et al . Image-based root phenotyping for field-grown crops: an example under maize/soybean intercropping [J]. J Integr Agric , 2022 , 21 ( 6 ): 1606 - 1619 . DOI: 10.1016/s2095-3119(20)63571-7 http://dx.doi.org/10.1016/s2095-3119(20)63571-7 .
Vikman P $ , Vessey J K . Ontogenetic changes in root nodule subpopulations of common bean ( Phaseolus vulgaris L.): Ⅲ. Nodule formation, growth and degradation [J]. J Exp Bot , 1993 , 44 ( 3 ): 579 - 586 . DOI: 10.1093/jxb/44.3.579 http://dx.doi.org/10.1093/jxb/44.3.579 .
Downie H F , Adu M O , Schmidt S , et al . Challenges and opportunities for quantifying roots and rhizosphere interactions through imaging and image analysis [J]. Plant Cell Environ , 2015 , 38 ( 7 ): 1213 - 1232 . DOI: 10.1111/pce.12448 http://dx.doi.org/10.1111/pce.12448 .
Barbedo J G A . Method for automatic counting root nodules using digital images [C]// 2012 12th International Conference on Computational Science and Its Applications . June 18 - 21 , 2012 . Salvador, Bahia, Brazil. IEEE , 2012: 159 - 161 . DOI: 10.1109/iccsa.2012.39 http://dx.doi.org/10.1109/iccsa.2012.39 .
Novikova T I , Gordienko N I . Study of the root nodules in some species of the Papilionaceae subfamily by scanning electron microscopy [J]. Tsitologiia , 2001 , 43 ( 2 ): 107 - 113 .
Smith A G , Han E , Petersen J , et al . RootPainter: deep learning segmentation of biological images with corrective annotation [J]. New Phytol , 2022 , 236 ( 2 ): 774 - 791 . DOI: 10.1111/nph.18387 http://dx.doi.org/10.1111/nph.18387 .
Chung Y S , Lee U , Heo S , et al . Image-based machine learning characterizes root nodule in soybean exposed to silicon [J]. Front Plant Sci , 2020 , 11 : 520161 . DOI: 10.3389/fpls.2020.520161 http://dx.doi.org/10.3389/fpls.2020.520161 .
Jubery T Z , Carley C N , Singh A , et al . Using machine learning to develop a fully automated soybean nodule acquisition pipeline (SNAP) [J]. Plant Phenomics , 2021 , 2021 : 9834746 . DOI: 10.34133/2021/9834746 http://dx.doi.org/10.34133/2021/9834746 .
Zhao Y , Zhang W , Zhang L X , et al . Development of an enhanced hybrid attention YOLOv8s small object detection method for phenotypic analysis of root nodules [J]. Artif Intell Agric , 2026 , 16 ( 1 ): 12 - 43 . DOI: 10.1016/j.aiia.2025.07.001 http://dx.doi.org/10.1016/j.aiia.2025.07.001 .
Liu L B , Yu L J , Wu D , et al . PocketMaize: an Android-smartphone application for maize plant phenotyping [J]. Front Plant Sci , 2021 , 12 : 770217 . DOI: 10.3389/fpls.2021.770217 http://dx.doi.org/10.3389/fpls.2021.770217 .
Cheung V , Westland S , Connah D , et al . A comparative study of the characterisation of colour cameras by means of neural networks and polynomial transforms [J]. Color Technol , 2004 , 120 ( 1 ): 19 - 25 . DOI: 10.1111/j.1478-4408.2004.tb00201.x http://dx.doi.org/10.1111/j.1478-4408.2004.tb00201.x .
Zhang X , Du B Y , Jia Y X , et al . MSD-YOLO11n: an improved small target detection model for high precision UAV aerial imagery [J]. J King Saud Univ Comput Inf Sci , 2025 , 37 ( 5 ): 100 . DOI: 10.1007/s44443-025-00116-0 http://dx.doi.org/10.1007/s44443-025-00116-0 .
Xiao Y , Guo Y , Pang Q H , et al . STar-DETR: a lightweight real-time detection transformer for space targets in optical sensor systems [J]. Sensors , 2025 , 25 ( 4 ): 1146 . DOI: 10.3390/s25041146 http://dx.doi.org/10.3390/s25041146 .
Liu Y J , Zhou S J , Liu X Y , et al . Unbiased faster R-CNN for single-source domain generalized object detection [C]// 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . June 16 - 22 , 2024 . Seattle, WA, USA . IEEE , 2024: 28838 - 28847 . DOI: 10.1109/cvpr52733.2024.02724 http://dx.doi.org/10.1109/cvpr52733.2024.02724 .
Zhang T Y , Suen C Y . A fast parallel algorithm for thinning digital patterns [J]. Commun ACM , 1984 , 27 ( 3 ): 236 - 239 . DOI: 10.1145/357994.358023 http://dx.doi.org/10.1145/357994.358023 .
Zafar A , Aftab D , Qureshi R , et al . Single stage adaptive multi-attention network for image restoration [J]. IEEE Trans Image Process , 2024 , 33 : 2924 - 2935 . DOI: 10.1109/tip.2024.3384838 http://dx.doi.org/10.1109/tip.2024.3384838 .
Kirillov A , He K M , Girshick R , et al . Panoptic segmentation [C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . June 15 - 20 , 2019 . LongBeach, CA, USA. IEEE, 2019 . DOI: 10.1109/cvpr.2019.00963 http://dx.doi.org/10.1109/cvpr.2019.00963 .
张灵西 . 基于拓扑结构约束和特征增强的医学影像标志点定位算法 [J]. 计算机系统应用 , 2022 , 31 ( 9 ): 173 - 182 . DOI: 10.15888/j.cnki.csa.008700 http://dx.doi.org/10.15888/j.cnki.csa.008700 .
刘建伟 , 刘俊文 , 罗雄麟 . 深度学习中注意力机制研究进展 [J]. 工程科学学报 , 2021 , 43 ( 11 ): 1499 - 1511 . DOI: 10.13374/j.issn2095-9389.2021.01.30.005 http://dx.doi.org/10.13374/j.issn2095-9389.2021.01.30.005 .
Wyburd M K , Dinsdale N K , Jenkinson M , et al . Anatomically plausible segmentations: Explicitly preserving topology through prior deformations [J]. Med Image Anal , 2024 , 97 : 103222 . DOI: 10.1016/j.media.2024.103222 http://dx.doi.org/10.1016/j.media.2024.103222 .
孙婧 , 苏剑波 . 先验信息驱动的跨模态通用特征空间构建与分析 [J]. 电子学报 , 2025 , 53 ( 8 ): 2614 - 2623 . DOI: 10.12263/DZXB.C250022 http://dx.doi.org/10.12263/DZXB.C250022 .
郭文豪 , 郝斌 , 张飞 . 改进YOLOv11的轻量化水下生物目标检测算法 [J/OL]. 电子测量与仪器学报 , 1 - 17 [ 2026-03-19 ]. https://link.cnki.net/urlid/11.2488.TN.20260226.1623.046 https://link.cnki.net/urlid/11.2488.TN.20260226.1623.046 .
dos Santos C F G , Arrais R R , Santos da Silva J V , et al . ISP meets deep learning: a survey on deep learning methods for image signal processing [J]. ACM Comput Surv , 2025 , 57 ( 5 ): 1 - 44 . DOI: 10.1145/3708516 http://dx.doi.org/10.1145/3708516 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621