The pod and seed counts are important yield-related traits in soybean.High-precision soybean breeders face the major challenge of accurately phenotyping the number of pods and seeds in a high-throughput manner.Recent ...The pod and seed counts are important yield-related traits in soybean.High-precision soybean breeders face the major challenge of accurately phenotyping the number of pods and seeds in a high-throughput manner.Recent advances in artificial intelligence,especially deep learning(DL)models,have provided new avenues for high-throughput phenotyping of crop traits with increased precision.However,the available DL models are less effective for phenotyping pods that are densely packed and overlap in insitu soybean plants;thus,accurate phenotyping of the number of pods and seeds in soybean plant is an important challenge.To address this challenge,the present study proposed a bottom-up model,DEKR-SPrior(disentangled keypoint regression with structural prior),for insitu soybean pod phenotyping,which considers soybean pods and seeds analogous to human people and joints,respectively.In particular,we designed a novel structural prior(SPrior)module that utilizes cosine similarity to improve feature discrimination,which is important for differentiating closely located seeds from highly similar seeds.To further enhance the accuracy of pod location,we cropped full-sized images into smaller and high-resolution subimages for analysis.The results on our image datasets revealed that DEKR-SPrior outperformed multiple bottom-up models,viz.,Lightweight-Open Pose,OpenPose,HigherH R Net,and DEKR,reducing the mean absolute error from 25.81(in the original DEKR)to 21.11(in the DEKR-SPrior)in pod phenotyping.This paper demonstrated the great potential of DEKR-SPrior for plant phenotyping,and we hope that DEKR-SPrior will help future plant phenotyping.展开更多
基金supported in part by the National Key Research and Development Program of China(2023YFD-1202600)the National Natural Science Foundation of China(62103380)+3 种基金the Research and Development Project from the Department of Science and Technology of Zhejiang Province(2023C01042)Soybean Intelligent Computational Breeding and Application of the Zhejiang Lab(2021PE0AC04)Intelligent Technology and Platform Development for Rice Breeding of the Zhejiang Lab(2021PE0AC05)Fine-grained Semantic Modeling and Cross modal Encoding-Decoding for Multilingual Scene Text Extraction(2022M722911).
文摘The pod and seed counts are important yield-related traits in soybean.High-precision soybean breeders face the major challenge of accurately phenotyping the number of pods and seeds in a high-throughput manner.Recent advances in artificial intelligence,especially deep learning(DL)models,have provided new avenues for high-throughput phenotyping of crop traits with increased precision.However,the available DL models are less effective for phenotyping pods that are densely packed and overlap in insitu soybean plants;thus,accurate phenotyping of the number of pods and seeds in soybean plant is an important challenge.To address this challenge,the present study proposed a bottom-up model,DEKR-SPrior(disentangled keypoint regression with structural prior),for insitu soybean pod phenotyping,which considers soybean pods and seeds analogous to human people and joints,respectively.In particular,we designed a novel structural prior(SPrior)module that utilizes cosine similarity to improve feature discrimination,which is important for differentiating closely located seeds from highly similar seeds.To further enhance the accuracy of pod location,we cropped full-sized images into smaller and high-resolution subimages for analysis.The results on our image datasets revealed that DEKR-SPrior outperformed multiple bottom-up models,viz.,Lightweight-Open Pose,OpenPose,HigherH R Net,and DEKR,reducing the mean absolute error from 25.81(in the original DEKR)to 21.11(in the DEKR-SPrior)in pod phenotyping.This paper demonstrated the great potential of DEKR-SPrior for plant phenotyping,and we hope that DEKR-SPrior will help future plant phenotyping.
