Defect detection method of lithium battery based on improved YOLOv7

Xiyang Pan; Linsheng Li

doi:10.1117/12.3025951

15 March 2024 Defect detection method of lithium battery based on improved YOLOv7

Xiyang Pan, Linsheng Li

Proceedings Volume 13075, Second International Conference on Physics, Photonics, and Optical Engineering (ICPPOE 2023); 1307520 (2024) https://doi.org/10.1117/12.3025951
Event: Second International Conference on Physics, Photonics, and Optical Engineering (ICPPOE 2023), 2023, Kunming, China

Abstract

For the traditional algorithm to detect lithium battery defects, the missing rate is high and the speed is slow, an improved YOLOv7 algorithm was proposed. Firstly, CBAM attention mechanism is added to feature extraction part, which can enhance network's representation ability. Secondly, in the feature fusion part, ConvNeXt lightweight module was used to replace the original ELAN module to reduce the model's complexity. Finally, SPD module is added before the detection head at the output end to increase focus on smaller goals with surface defects of lithium batteries at the output end. The results show that the optimization algorithm can improve the accuracy and speed of the lithium battery. The proposed algorithm achieves a 92.7% detection accuracy, surpassing the original network by 2.1%.

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

Citation Download Citation

Xiyang Pan and Linsheng Li "Defect detection method of lithium battery based on improved YOLOv7", Proc. SPIE 13075, Second International Conference on Physics, Photonics, and Optical Engineering (ICPPOE 2023), 1307520 (15 March 2024); https://doi.org/10.1117/12.3025951

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