RT Journal Article
SR Electronic
A1 Shi, Yun
A1 Zhu, Yan-Yan
A1 Wang, Jun-Qi
T1 Surface Defect Detection Method for Welding Robot Workpiece Based on Machine Vision Technology
JF Manufacturing Technology Journal
YR 2023
VO 23
IS 5
SP 691
OP 699
DO 10.21062/mft.2023.100
UL https://journalmt.com/artkey/mft-202305-0019.php
AB With the development of welding technology and the improvement of automation level, welding robots are playing an increasingly important role in industrial production. However, during the welding process, due to factors such as material characteristics, welding parameters, or improper processes, defects may appear on the surface of the workpiece, which may reduce the quality and service life of the workpiece. In order to solve this problem, this article used frequency domain feature extraction and nearest neighbor classifier in workpiece detection algorithms under machine vision technology to extract and classify surface defect images of workpiece, and studied the detection method of welding robot workpiece surface defects. The research results indicated that, under the same other conditions, the accuracy of machine vision technology was over 90% for all five different defect types, while the accuracy of traditional technology was between 75.5% and 84%. The performance of machine vision technology was far superior to traditional technology, indicating that machine vision technology could improve the accuracy of welding robot workpiece surface defect detection methods.