Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybean

Abstract Background Phenotypic characterization of mature soybean pods is a crucial aspect of breeding programs, yet efficiently obtaining accurate pod phenotypic parameters remains a major challenge. Recent advances in deep learning, particularly in keypoint detection models, have introduced innova...

Full description

Saved in:
Bibliographic Details
Main Authors: Fei Liu, Hang Liu, Qiong Wu, Zhongzhi Han, Shanchen Pang, Shudong Wang, Longgang Zhao
Format: Article
Language:English
Published: BMC 2025-06-01
Series:Plant Methods
Subjects:
Online Access:https://doi.org/10.1186/s13007-025-01399-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849335787294818304
author Fei Liu
Hang Liu
Qiong Wu
Zhongzhi Han
Shanchen Pang
Shudong Wang
Longgang Zhao
author_facet Fei Liu
Hang Liu
Qiong Wu
Zhongzhi Han
Shanchen Pang
Shudong Wang
Longgang Zhao
author_sort Fei Liu
collection DOAJ
description Abstract Background Phenotypic characterization of mature soybean pods is a crucial aspect of breeding programs, yet efficiently obtaining accurate pod phenotypic parameters remains a major challenge. Recent advances in deep learning, particularly in keypoint detection models, have introduced innovative methods for pod phenotype extraction. However, precise identification and analysis of fine-scale phenotypic traits in soybean pods remain challenging in current research. Results We propose Pod-pose, an innovative top-down keypoint detection model for precise soybean pod phenotyping that adapts human pose estimation techniques to plant phenotyping. Specifically, Pod-pose integrates the architectural strengths of various advanced YOLO (You Only Look Once) models through bottleneck structure optimization and positional feature enhancement to achieve superior detection accuracy. Furthermore, we implemented a two-stage detection method augmented with transfer learning, which not only reduces training complexity but also significantly enhances the model's performance. Extensive evaluation of our custom-built dataset demonstrated Pod-Pose's superior performance, with the X variant achieving an Average Precision of 0.912 at an IoU threshold of 0.5 (AP@IoU = 0.5). Notably, four critical pod-related phenotypic traits were successfully quantified: pod length, bending length, curvature, and inflection point width. Conclusions This study establishes Pod-Pose as a viable solution for pod phenotyping, with potential applications in soybean breeding optimization.
format Article
id doaj-art-c2515f2f8a5041008889aa4dc687a561
institution Kabale University
issn 1746-4811
language English
publishDate 2025-06-01
publisher BMC
record_format Article
series Plant Methods
spelling doaj-art-c2515f2f8a5041008889aa4dc687a5612025-08-20T03:45:10ZengBMCPlant Methods1746-48112025-06-0121111710.1186/s13007-025-01399-0Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybeanFei Liu0Hang Liu1Qiong Wu2Zhongzhi Han3Shanchen Pang4Shudong Wang5Longgang Zhao6Qingdao Agricultural UniversityQingdao Agricultural UniversityQingdao Agricultural UniversityQingdao Agricultural UniversityChina University of Petroleum (East China)China University of Petroleum (East China)Qingdao Agricultural UniversityAbstract Background Phenotypic characterization of mature soybean pods is a crucial aspect of breeding programs, yet efficiently obtaining accurate pod phenotypic parameters remains a major challenge. Recent advances in deep learning, particularly in keypoint detection models, have introduced innovative methods for pod phenotype extraction. However, precise identification and analysis of fine-scale phenotypic traits in soybean pods remain challenging in current research. Results We propose Pod-pose, an innovative top-down keypoint detection model for precise soybean pod phenotyping that adapts human pose estimation techniques to plant phenotyping. Specifically, Pod-pose integrates the architectural strengths of various advanced YOLO (You Only Look Once) models through bottleneck structure optimization and positional feature enhancement to achieve superior detection accuracy. Furthermore, we implemented a two-stage detection method augmented with transfer learning, which not only reduces training complexity but also significantly enhances the model's performance. Extensive evaluation of our custom-built dataset demonstrated Pod-Pose's superior performance, with the X variant achieving an Average Precision of 0.912 at an IoU threshold of 0.5 (AP@IoU = 0.5). Notably, four critical pod-related phenotypic traits were successfully quantified: pod length, bending length, curvature, and inflection point width. Conclusions This study establishes Pod-Pose as a viable solution for pod phenotyping, with potential applications in soybean breeding optimization.https://doi.org/10.1186/s13007-025-01399-0Soybean podDeep learningKeypoint detectionDigital breeding
spellingShingle Fei Liu
Hang Liu
Qiong Wu
Zhongzhi Han
Shanchen Pang
Shudong Wang
Longgang Zhao
Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybean
Plant Methods
Soybean pod
Deep learning
Keypoint detection
Digital breeding
title Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybean
title_full Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybean
title_fullStr Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybean
title_full_unstemmed Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybean
title_short Pod-pose : an efficient top-down keypoint detection model for fine-grained pod phenotyping in mature soybean
title_sort pod pose an efficient top down keypoint detection model for fine grained pod phenotyping in mature soybean
topic Soybean pod
Deep learning
Keypoint detection
Digital breeding
url https://doi.org/10.1186/s13007-025-01399-0
work_keys_str_mv AT feiliu podposeanefficienttopdownkeypointdetectionmodelforfinegrainedpodphenotypinginmaturesoybean
AT hangliu podposeanefficienttopdownkeypointdetectionmodelforfinegrainedpodphenotypinginmaturesoybean
AT qiongwu podposeanefficienttopdownkeypointdetectionmodelforfinegrainedpodphenotypinginmaturesoybean
AT zhongzhihan podposeanefficienttopdownkeypointdetectionmodelforfinegrainedpodphenotypinginmaturesoybean
AT shanchenpang podposeanefficienttopdownkeypointdetectionmodelforfinegrainedpodphenotypinginmaturesoybean
AT shudongwang podposeanefficienttopdownkeypointdetectionmodelforfinegrainedpodphenotypinginmaturesoybean
AT longgangzhao podposeanefficienttopdownkeypointdetectionmodelforfinegrainedpodphenotypinginmaturesoybean