基于实例分割技术的草莓叶龄及冠幅表型快速提取方法  

Fast Extracting Method for Strawberry Leaf Age and Canopy Width Based on Instance Segmentation Technology

作　　者：樊江川 王源桥[2,3] 苟文博 蔡双泽 郭新宇 赵春江[2] FAN Jiangchuan;WANG Yuanqiao;GOU Wenbo;CAI Shuangze;GUO Xinyu;ZHAO Chunjiang(College of Information and Electrical Engineering,China Agricultural University,Beijing 100083,China;Beijing Key Laboratory of Digital Plant,Beijing Research Center for Information Technology in Agriculture,China National Engineering Research Center for Information Technology in Agriculture(NERCITA),Beijing 100097,China;College of Information Engineering,Northwest A&F University,Yangling 712100 Shaanxi,China;Beijing PAIDE Science and Technology Development Co.,Ltd.,Beijing 100097,China)

机构地区：[1]中国农业大学信息与电气工程学院,北京100083 [2]国家农业信息化工程技术研究中心/北京市农林科学院信息技术研究中心/数字植物北京市重点实验室,北京100097 [3]西北农林科技大学信息工程学院,陕西杨陵712100 [4]北京派得伟业科技发展有限公司,北京100097

出　　处：《智慧农业（中英文）》2024年第2期95-106,共12页Smart Agriculture

基　　金：北京市科技新星计划(Z211100002121065);北京市科技新星计划交叉合作课题(Z20220484202);“十四五”国家重点研发计划项目(2022YFD2002302-02)。

摘　　要：[目的/意义]为解决高通量草莓叶龄及冠幅提取问题,提出一种基于移动式表型平台和实例分割技术的高通量表型提取方法。[方法]利用小型移动式表型平台对温室内盆栽草莓植株的俯拍图像进行数据获取,并利用改进型Mask R-CNN(Convolutional Neural Network)模型对图像进行处理,以此获取草莓植株叶龄信息。首先利用带有分组注意力模块的Split-Attention Networks(ResNeSt)骨干网络替换原有网络,从而提高图像特征信息提取精度和执行效率。在训练时,利用Mosaic方法对草莓图像进行数据增强,并且使用二元交叉熵损失函数对原本的交叉熵分类损失函数进行优化,以达到更好的植株与叶片的检测准确度。在此基础上,对训练结果进行后处理,利用标定比值对冠幅进行计算。[结果和讨论]该方法能够在ResNeSt-101骨干网络下,实现80.1%的掩膜准确率和89.6%的检测框准确率,并且能够以99.3%的植株检测正确率和98.0%的叶片数量检出率实现高通量的草莓叶龄估算工作。而模型推理后草莓植株南北和东西向冠幅测试值与真实值相比误差均低于5%的约占98.1%。[结论]该方法有着较高的鲁棒性,能够为智慧农业下高通量植物表型获取与解析工作提供技术支持。[Objective]There's a growing demand among plant cultivators and breeders for efficient methods to acquire plant phenotypic traits at high throughput,facilitating the establishment of mappings from phenotypes to genotypes.By integrating mobile phenotyping platforms with improved instance segmentation techniques,researchers have achieved a significant advancement in the automation and accuracy of phenotypic data extraction.Addressing the need for rapid extraction of leaf age and canopy width phenotypes in strawberry plants cultivated in controlled environments,this study introduces a novel high-throughput phenotyping extraction approach leveraging a mobile phenotyping platform and instance segmentation technology.[Methods]Data acquisition was conducted using a compact mobile phenotyping platform equipped with an array of sensors,including an RGB sensor,and edge control computers,capable of capturing overhead images of potted strawberry plants in greenhouses.Targeted adjustments to the network structure were made to develop an enhanced convolutional neural network(Mask R-CNN)model for processing strawberry plant image data and rapidly extracting plant phenotypic information.The model initially employed a split-attention networks(ResNeSt)backbone with a group attention module,replacing the original network to improve the precision and efficiency of image feature extraction.During training,the model adopted the Mosaic method,suitable for instance segmentation data augmentation,to expand the dataset of strawberry images.Additionally,it optimized the original cross-entropy classification loss function with a binary cross-entropy loss function to achieve better detection accuracy of plants and leaves.Based on this,the improved Mask R-CNN description involves post-processing of training results.It utilized the positional relationship between leaf and plant masks to statistically count the number of leaves.Additionally,it employed segmentation masks and image calibration against true values to calculate the canopy width of

关 键 词：移动式表型平台 实例分割 草莓表型 叶龄统计 冠幅 Mask R-CNN ResNeSt 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] S22[自动化与计算机技术—控制科学与工程]