SPP-extractor:Automatic phenotype extraction for densely grown soybean plants

作     者：Wan Zhou Yijie Chen Weihao Li Cong Zhang Yajun Xiong Wei Zhan Lan Huang Jun Wang Lijuan Qiu Wan Zhou;Yijie Chen;Weihao Li;Cong Zhang;Yajun Xiong;Wei Zhan;Lan Huang;Jun Wang;Lijuan Qiu

作者机构：College of Computer ScienceYangtze UniversityJingzhou 434023HubeiChina College of AgricultureYangtze UniversityJingzhou 434025HubeiChina National Key Facility for Gene Resources and Genetic Improvement/Institute of Crop SciencesChinese Academy of Agricultural SciencesBeijing 100081China MARA Key Laboratory of Sustainable Crop Production in the Middle Reaches of the Yangtze River(Co-construction by Ministry and Province)College of AgricultureYangtze UniversityJingzhou 434025HubeiChina 

出 版 物：《作物学报:英文版》 (The Crop Journal)

年 卷 期：2023年第11卷第5期

页      面：1569-1578页

核心收录：

学科分类：0710[理学-生物学] 0401[教育学-教育学] 04[教育学] 0901[农学-作物学] 0902[农学-园艺学] 

基　　金：supported by the National Natural Science Foundation of China(62276032,32072016) the Agricultural Science and Technology Innovation Program(ASTIP)of Chinese Academy of Agricultural Sciences 

主　　题：Soybean phenotype Branch length Computer vision A*algorithm Phenotype acquisition 

摘      要：Automatic collecting of phenotypic information from plants has become a trend in breeding and smart *** mature soybean plants at the harvesting stage,which are dense and overlapping,we have proposed the SPP-extractor(soybean plant phenotype extractor)algorithm to acquire phenotypic ***,to address the mutual occultation of pods,we augmented the standard YOLOv5s model for target detection with an additional attention *** resulting model could accurately identify pods and stems and could count the entire pod set of a plant in a single ***,considering that mature branches are usually bent and covered with pods,we designed a branch recognition and measurement module combining image processing,target detection,semantic segmentation,and heuristic *** results on real plants showed that SPP-extractor achieved respective R^(2) scores of 0.93–0.99 for four phenotypic traits,based on regression on manual measurements.