AUTHOR=Zhu Rongsheng , Wang Xueying , Yan Zhuangzhuang , Qiao Yinglin , Tian Huilin , Hu Zhenbang , Zhang Zhanguo , Li Yang , Zhao Hongjie , Xin Dawei , Chen Qingshan 

TITLE=Exploring Soybean Flower and Pod Variation Patterns During Reproductive Period Based on Fusion Deep Learning

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.922030

DOI=10.3389/fpls.2022.922030

ISSN=1664-462X

ABSTRACT=The soybean flower and pod drop are important factors in soybean yield, and the use of computer vision techniques to obtain the phenotypes of flower and pod in bulk, as well as in a quick and accurate manner, is a key aspect of the study of soybean flower and pod drop rate. This paper compared a variety of deep learning algorithms for identifying and counting soybean flower and pod, and found that the Faster R-CNN model had the best performance. Furthermore, the Faster R-CNN model was further improved and optimized based on the characteristics of soybean flower and pod. The accuracy of the final model for identifying flower and pod was increased to 94.36% and 91%, respectively. Afterwards, a fusion model for soybean pod recognition and counting was proposed based on the Faster R-CNN model, where the coefficient of determination R^2 between counts of soybean flower and pod by fusion model and manual counts reached 0.965 and 0.98, respectively. The above results show that the fusion model is a robust recognition and counting algorithm that can reduce labor intensity and improve efficiency. Its application will greatly facilitate the study of the variable patterns of soybean flower and pod during the reproductive period. Finally, based on the fusion model, we explored the variable patterns of soybean flower and pod during the reproductive period, the spatial distribution patterns of soybean flower and pod , and soybean flower and pod drop patterns.