摘要
焊接技术被广泛应用于航空航天、机械、核能、船舶及石油化工等领域。为保证焊缝质量,提升焊接件的可靠性,对焊接件进行无损检测是一个不可或缺的环节。随着计算机技术的发展,焊缝缺陷的识别方法已成为目前的研究热点。基于机器视觉的焊缝缺陷识别方法进行了综述,分别介绍了焊缝缺陷图像的预处理、基于机器学习的缺陷检测以及基于深度学习的缺陷识别,指出了未来的研究方向,为该领域进一步深入研究提供参考。
Welding technology is widely used in aerospace,machinery,nuclear energy,shipbuilding and petrochemical fields.In order to ensure the weld quality and improve the reliability of welding parts,nondestructive testing of welding parts is an indispensable link.With the development of computer technology,the recognition method of weld defects has become a research hotspot.Identification method of weld defects based on machine vision is reviewed.The preprocessing of weld defect image,defect detection based on machine learning and defect recognition based on deep learning are introduced respectively.The future research direction is pointed out,provide some reference for further research in this field.
作者
张思
石峰
ZHANG Si;SHI Feng(Sinopec Engineering Quality Monitoring Co.,Ltd. , Beijing 100020 , China;Luoyang Xinlong Engineering Testing Co.,Ltd. , Luoyang 471012 , China)
出处
《河南化工》
CAS
2022年第6期15-19,共5页
Henan Chemical Industry
关键词
焊接缺陷
检测
预处理
机器学习
深度学习
welding defects
testing
pretreatment
machine learning
deep learning