Automotive door panel weld joint recognition based on improved YOLOv5 algorithm

Yimin Sun; Jian Wu; Ying Cong; Mingbo Ma

doi:10.1117/12.2674389

28 April 2023 Automotive door panel weld joint recognition based on improved YOLOv5 algorithm

Yimin Sun, Jian Wu, Ying Cong, Mingbo Ma

Proceedings Volume 12626, International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022); 126260N (2023) https://doi.org/10.1117/12.2674389
Event: International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022), 2022, Zhuhai, China

Abstract

With the rapid development of automation technology, the processing of various kinds of shaped parts has received the attention of the majority of scholars, such as the identification of automotive door panel weld joints. Automotive door plate weld joint features are small targets, interference, and uneven distribution. The previous solder joint recognition method is affected by the small target and many interferences, and these disturbances lead to inaccurate solder joint identification. To address the problem of inaccurate solder joint recognition, a solder joint recognition method based on the improved YOLOv5 algorithm is proposed. Firstly, the dataset of solder joints is constructed, and then the YOLOv5 network structure is improved by adding a tiny target detection layer，which makes the accuracy of small target detection improved. The accuracy of the improved YOLOv5 for solder joint detection is improved from the original 91.7% to 98.9%, which can be extended to such small target application scenarios.

Citation Download Citation

Yimin Sun, Jian Wu, Ying Cong, and Mingbo Ma "Automotive door panel weld joint recognition based on improved YOLOv5 algorithm", Proc. SPIE 12626, International Conference on Signal Processing, Computer Networks, and Communications (SPCNC 2022), 126260N (28 April 2023); https://doi.org/10.1117/12.2674389

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