GPU-based defect detection of spiral submerged arc welds

Xuanming Liu; Weixin Gao

doi:10.1117/12.3034121

13 June 2024 GPU-based defect detection of spiral submerged arc welds

Xuanming Liu, Weixin Gao

Author Affiliations +

Proceedings Volume 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024); 131804K (2024) https://doi.org/10.1117/12.3034121
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 2024, Guangzhou, China

Abstract

Spiral submerged arc welding, as a highly efficient and automated welding method, is widely used in industrial manufacturing. However, weld defects are inevitable during the welding process, affecting the quality of welding. Given the high-speed requirements of factories for the inspection of spiral submerged arc welds, this paper proposes a method for detecting weld defects in spiral submerged arc welding based on Graphics Processing Units (GPU), utilizing the parallel computing capabilities of GPUs to achieve fast and accurate detection of weld defects. This method begins by capturing images of the spiral submerged arc welding seams, followed by the use of image processing technology to preprocess the images and extract weld features. Subsequently, a defect detection model based on deep learning is established, and parallel computing on GPUs is employed to quickly process the weld images and detect defects. Experimental results show that the accuracy of defect detection reaches 95%, indicating that this method can effectively detect various defects in welds with high accuracy and efficiency. It provides a new solution for quality control of spiral submerged arc weld seams.

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

Citation Download Citation

Xuanming Liu and Weixin Gao "GPU-based defect detection of spiral submerged arc welds", Proc. SPIE 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 131804K (13 June 2024); https://doi.org/10.1117/12.3034121

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

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Defect detection

Image processing

Education and training

Neural networks

Convolution

Matrices

Parallel computing

Show All Keywords

Keywords/Phrases

Search In:

Publication Years