CMRD-Net: a deep learning-based Cnaphalocrocis medinalis damage symptom rotated detection framework for in-field survey

Frontiers in Plant Science

Plant Science

Wei Dong¹  Jie Zhang²  Jianming Du²  Hongbo Chen³  Tianjiao Chen³  Rujing Wang⁴  Meng Zhang⁵ 

[1] Agricultural Economy and Information Research Institute, Anhui Academy of Agricultural Sciences, Hefei, China;Institute of Intelligent Machines, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, China;Institute of Intelligent Machines, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, China;Science Island Branch, University of Science and Technology of China, Hefei, China;Institute of Intelligent Machines, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, China;Science Island Branch, University of Science and Technology of China, Hefei, China;Institutes of Physical Science and Information Technology, Anhui University, Hefei, China;Jingxian Plant Protection Station, Jingxian Plantation Technology Extension Center, Xuancheng, China;

关键词: Cnaphalocrocis medinalis;    damage symptom;    deep learning;    rotated object detection;    horizontal object detection;   

DOI  :  10.3389/fpls.2023.1180716

 received in 2023-03-06, accepted in 2023-05-03,  发布年份 2023

来源: Frontiers

【 摘 要 】

The damage symptoms of Cnaphalocrocis medinalis (C.medinalis) is an important evaluation index for pest prevention and control. However, due to various shapes, arbitrary-oriented directions and heavy overlaps of C.medinalis damage symptoms under complex field conditions, generic object detection methods based on horizontal bounding box cannot achieve satisfactory results. To address this problem, we develop a Cnaphalocrocis medinalis damage symptom rotated detection framework called CMRD-Net. It mainly consists of a Horizontal-to-Rotated region proposal network (H2R-RPN) and a Rotated-to-Rotated region convolutional neural network (R2R-RCNN). First, the H2R-RPN is utilized to extract rotated region proposals, combined with adaptive positive sample selection that solves the hard definition of positive samples caused by oriented instances. Second, the R2R-RCNN performs feature alignment based on rotated proposals, and exploits oriented-aligned features to detect the damage symptoms. The experimental results on our constructed dataset show that our proposed method outperforms those state-of-the-art rotated object detection algorithms achieving 73.7% average precision (AP). Additionally, the results demonstrate that our method is more suitable than horizontal detection methods for in-field survey of C.medinalis.

【 授权许可】

Unknown   
Copyright © 2023 Chen, Wang, Du, Chen, Zhang, Dong and Zhang