Research on improved YOLOv5-based pest recognition

Mengjie Xing; Shukun Cao; Zhenyu Tang; Guo Zhao

doi:10.1117/12.3025578

28 February 2024 Research on improved YOLOv5-based pest recognition

Mengjie Xing, Shukun Cao, Zhenyu Tang, Guo Zhao

Proceedings Volume 13071, International Conference on Mechatronic Engineering and Artificial Intelligence (MEAI 2023); 1307136 (2024) https://doi.org/10.1117/12.3025578
Event: International Conference on Mechatronic Engineering and Artificial Intelligence (MEAI 2023), 2023, Shenyang, China

Abstract

Previous means of controlling pests and diseases were relatively outdated and did not improve the yield and efficiency of farmland very much. In order to be able to monitor the degree of pests and diseases of planted crops such as wheat and corn in real time and to increase robustness, this paper proposes a pest detection model YOLOv5s-ECA based on the improved YOLOv5 algorithm, which introduces an attention mechanism, ECA, into the backbone network of YOLOv5s in order to achieve the enhancement of the network's capability of extracting image features while there is no increase in the model's parameters and volume. Thus, the purpose of improving the accuracy of detecting the target is achieved. In order to verify the performance of the model, we built a pest and disease dataset, and the training results show that YOLOv5-ECA improves the detection mAP by 3.1% and Precision by 4.1%, and the detection results are better than other detection algorithms.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Mengjie Xing, Shukun Cao, Zhenyu Tang, and Guo Zhao "Research on improved YOLOv5-based pest recognition", Proc. SPIE 13071, International Conference on Mechatronic Engineering and Artificial Intelligence (MEAI 2023), 1307136 (28 February 2024); https://doi.org/10.1117/12.3025578

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available