基于微根管图像的作物根系分割和表型信息提取  

Crop root segmentation and phenotypic information extraction based on images of minirhizotron

作　　者：郑一力[1,2,3] 张振翔 邢达 刘卫平 ZHENG Yili;ZHANG Zhenxiang;XING Da;LIU Weiping(School of Technology,Beijing Forestry University,Beijing 100083,China;Key Laboratory of National Forestry and Grassland Administration on Forestry Equipment and Automation,Beijing 100083,China;State Key Laboratory of Efficient Production of Forest Resources,Beijing 100083,China)

机构地区：[1]北京林业大学工学院,北京100083 [2]林业装备与自动化国家林业和草原局重点实验室,北京100083 [3]林木资源高效生产全国重点实验室,北京100083

出　　处：《农业工程学报》2024年第18期110-119,共10页Transactions of the Chinese Society of Agricultural Engineering

基　　金：国家自然科学青年基金项目(32101590);北京林业大学“5·5工程”科研创新团队项目(BLRC2023C05)。

摘　　要：微根管法采集的作物根系图像具有复杂的土壤背景和较小的根系占比,当深度学习的感受野较小或多尺度特征融合不充分时,会导致根系边缘处的像素被错分为土壤。同时,微根管法的图像采集周期长且在初期很难采集到大量有效样本,限制了根系提取模型的快速部署。为提升根系表型测算精度和优化提取模型部署策略,该研究设计了一种原位自动根系成像系统以实时获取作物的微根管图像,构建全尺度跳跃特征融合机制,使用感受野丰富的U^(2)-Net模型对微根管图像中的根系像素进行有效分类。结合数据增强以及迁移学习微调训练,实现对目标种类根系提取模型的快速部署。试验结果表明,使用加入全尺度跳跃特征融合机制的改进U^(2)-Net模型对蒜苗根系分割的F1得分和交并比IoU分别为86.54%和76.28%,相比改进前、U-Net、SegNet和DeeplabV3+_Resnet50模型,F1得分分别提高0.66、5.51、8.67和2.84个百分点;交并比分别提高1.02、8.18、12.52和4.31个百分点。迁移学习微调训练相比混合训练,模型的F1得分和交并比分别提高了2.89和4.45个百分点。改进U^(2)-Net模型分割图像的根系长度、面积和平均直径与手动标注结果的决定系数R^(2)分别为0.965、0.966、0.830。研究结果可为提升基于微根管图像的根系表型测算精度和根系提取模型的快速部署提供参考。The images of crop roots collected by the minirhizotron method have a complex soil background,and roots occupy a relatively small proportion.Common semantic segmentation models used for root extraction may misclassify pixels at the edge of roots as soil when the receptive field is small or multi-scale feature fusion is inadequate.Moreover,the long image acquisition cycle and initial difficulty in collecting a large number of valid samples with the minirhizotron method hindered the rapid deployment of root extraction models.To improve the accuracy of root phenotyping measurements and optimize extraction model deployment strategies,this study devised an in-situ automatic root imaging system to acquire minirhizotron images of crops in real time.A full-scale skip feature fusion mechanism is constructed for the U^(2)-Net model with rich receptive fields for the effective classification of root pixels in minirhizotron images.Integrating data augmentation and fine-tuning method of transfer learning methods to achieve rapid deployment of root extraction models for target species.The full-scale skip feature fusion mechanism involved fusing the output features of the upper encoder and lower decoder of the U^(2)-Net model across all scales,thereby serving as input features for a certain decoder layer and effectively retaining more feature information to enhance the decoder's information restoration capability.In terms of model deployment,this study compared fine tuning method of transfer learning with mixed training to address the issue of model training with limited samples.Experimental materials included self-developed in-situ automatic root imaging systems for collecting garlic sprout root system images and the publicly available minirhizotron dataset PRMI(plant root minirhizotron imagery).The experimental design included performance metric comparisons on the PRMI(plant root minirhizotron imagery)dataset,followed by analysis and validation on garlic sprout data.The control group comprised pre-improvement and post-improve

关 键 词：图像分割 迁移学习 微根管 U2-Net 蒜苗 根系表型 

分 类 号：S24[农业科学—农业电气化与自动化] TP391.4[农业科学—农业工程]