Accurate and fast implementation of soybean pod counting and localization from high-resolution image

Zhenghong Yu; Yangxu Wang; Jianxiong Ye; Shengjie Liufu; Dunlu Lu; Xiuli Zhu; Zhongming Yang; Qingji Tan

doi:10.3389/fpls.2024.1320109

Accurate and fast implementation of soybean pod counting and localization from high-resolution image

Front Plant Sci. 2024 Feb 20:15:1320109. doi: 10.3389/fpls.2024.1320109. eCollection 2024.

Authors

Zhenghong Yu^#¹, Yangxu Wang^#^{1

2}, Jianxiong Ye^{1

3}, Shengjie Liufu⁴, Dunlu Lu¹, Xiuli Zhu¹, Zhongming Yang¹, Qingji Tan⁵

Affiliations

¹ College of Robotics, Guangdong Polytechnic of Science and Technology, Zhuhai, China.
² Department of Network Technology, Guangzhou Institute of Software Engineering, Conghua, China.
³ School of Electronics and Information Engineering, Wuyi University, Jiangmen, China.
⁴ College of Business, Guangzhou College of Technology and Business, Foshan, China.
⁵ School of Mechanical Engineering, Guangxi University, Nanning, China.

^# Contributed equally.

Abstract

Introduction: Soybean pod count is one of the crucial indicators of soybean yield. Nevertheless, due to the challenges associated with counting pods, such as crowded and uneven pod distribution, existing pod counting models prioritize accuracy over efficiency, which does not meet the requirements for lightweight and real-time tasks.

Methods: To address this goal, we have designed a deep convolutional network called PodNet. It employs a lightweight encoder and an efficient decoder that effectively decodes both shallow and deep information, alleviating the indirect interactions caused by information loss and degradation between non-adjacent levels.

Results: We utilized a high-resolution dataset of soybean pods from field harvesting to evaluate the model's generalization ability. Through experimental comparisons between manual counting and model yield estimation, we confirmed the effectiveness of the PodNet model. The experimental results indicate that PodNet achieves an R² of 0.95 for the prediction of soybean pod quantities compared to ground truth, with only 2.48M parameters, which is an order of magnitude lower than the current SOTA model YOLO POD, and the FPS is much higher than YOLO POD.

Discussion: Compared to advanced computer vision methods, PodNet significantly enhances efficiency with almost no sacrifice in accuracy. Its lightweight architecture and high FPS make it suitable for real-time applications, providing a new solution for counting and locating dense objects.

Keywords: computer vision; convolutional network; counting and locating; dense objects; soybean pod.

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was supported in part by 2022 key scientific research project of ordinary universities in Guangdong Province under Grant 2022ZDZX4075, in part by 2022 Guangdong province ordinary universities characteristic innovation project under Grant 2022KTSCX251, in part by the Collaborative Intelligent Robot Production & Education Integrates Innovative Application Platform Based on the Industrial Internet under Grant 2020CJPT004, in part by 2020 Guangdong Rural Science and Technology Mission Project under Grant KTP20200153, in part by the Engineering Research Centre for Intelligent equipment manufacturing under Grant 2021GCZX018, in part by GDPST&DOBOT Collaborative Innovation Centre under Grant K01057060 and in part by Innovation Project of Guangxi Graduate Education under Grant YCSW2022081.