He Jingjing, Weng Lin, Xu Xiaogang, Chen Ruochen, Peng Bo, Li Nannan, Xie Zhengchao, Sun Lijian, Han Qiang, He Pengfei, Wang Fangfang, Yu Hui, Bhat Javaid Akhter, Feng Xianzhong
Zhejiang Laboratory, Hangzhou 311100, Zhejiang, China.
School of Computer Science and Technology, Zhejiang Gongshang University, Hangzhou 310012, Zhejiang, China.
Plant Phenomics. 2024 Jun 27;6:0198. doi: 10.34133/plantphenomics.0198. eCollection 2024.
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 in situ 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 in situ 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-OpenPose, OpenPose, HigherHRNet, 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.
豆荚和种子数量是大豆中与产量相关的重要性状。高精度大豆育种者面临着以高通量方式准确测定豆荚和种子数量的重大挑战。人工智能的最新进展，特别是深度学习（DL）模型，为提高作物性状高通量表型分析的精度提供了新途径。然而，现有的DL模型在对原位大豆植株中密集排列且相互重叠的豆荚进行表型分析时效果较差；因此，准确测定大豆植株中豆荚和种子的数量是一项重要挑战。为应对这一挑战，本研究提出了一种自底向上的模型DEKR-SPrior（具有结构先验的解缠关键点回归）用于原位大豆豆荚表型分析，该模型分别将大豆豆荚和种子类比为人体和关节。具体而言，我们设计了一种新颖的结构先验（SPrior）模块，利用余弦相似度来提高特征辨别能力，这对于区分位置相近的种子与高度相似的种子非常重要。为进一步提高豆荚定位的准确性，我们将全尺寸图像裁剪成更小的高分辨率子图像进行分析。我们图像数据集的结果表明，DEKR-SPrior优于多个自底向上的模型，即Lightweight-OpenPose、OpenPose、HigherHRNet和DEKR，在豆荚表型分析中将平均绝对误差从25.81（原始DEKR）降低到21.11（DEKR-SPrior）。本文证明了DEKR-SPrior在植物表型分析方面的巨大潜力，我们希望DEKR-SPrior将有助于未来的植物表型分析。
