金属加工（热加工）Issue(10)：12-19,8.
基于机器视觉的焊接缺陷检测方法研究
Research on welding defect detection method based on machine vision
摘要
Abstract
In order to improve welding quality and production efficiency,an automatic welding defect detection method based on machine vision is proposed in this paper.The method realizes intelligent detection and quantitative evaluation of welding defects by constructing a system architecture including image acquisition,preprocessing and defect identification.Firstly,in the image processing,the median filter and Gaussian filter algorithm are comprehensively used to effectively suppress the noise,and the discernability of the weld and its defect features is significantly enhanced through the steps of gray-scale,binarization and morphological processing.Secondly,combined with edge detection technology and quantitative analysis of defect area,the common defect types such as pores,cracks and non-fusion in welds are accurately identified,providing reliable data support for welding quality control.Finally,the experimental results show that the method has a high detection accuracy for typical welding defects such as cracks,pores and non-fusion,and the average error between the measured results and the measured values is controlled within 0.1cm,which fully meets the practical application needs of industrial production.关键词
机器视觉/焊接缺陷检测/图像处理/缺陷特征提取/缺陷面积测量Key words
machine vision/welding defect detection/image processing/defect feature extraction/defect area measurement引用本文复制引用
张立明,黄怀德,张海军,吴小刚,张富富,刘新华..基于机器视觉的焊接缺陷检测方法研究[J].金属加工（热加工）,2025,(10):12-19,8.基金项目
国家自然科学基金(U24A20116). （U24A20116）
