摘要
焊缝跟踪图像处理是对视觉传感器采集到的图像进行处理,抑制非必要信息,改善图像质量,将焊缝特征信息提取的过程。首先简述了视觉系统的典型组成,其次分析了图像处理技术在的焊缝跟踪方面的研究现状,包括图像预处理、焊缝特征提取和基于深度学习的图像处理,最后总结了基于主动视觉的焊缝跟踪技术中存在的不足和发展方向。
Weld-tracking image processing involves collecting the image by visual sensor, suppressing unnecessary information, improving image quality, and extracting seam feature information. First, this paper briefly describes the typical composition of a vision system, then analyzes the research status of image processing technology in weld tracking, including image processing, weld feature extraction and image processing based on deep learning, and finally summarizes the shortcomings and developmental direction of weld tracking technology based on active vision.
作者
叶冬旭
魏昕
蒙禹舟
YE Dongxu;WEI Xin;MENG Yuzhou(School of Electromechanical Engineering,Guangdong University of Technology,Guangzhou 510006,China)
出处
《激光杂志》
CAS
北大核心
2022年第11期6-10,共5页
Laser Journal
基金
广州市产学研重大专项资助项目(No.201604016115)。
关键词
焊缝跟踪
视觉系统
图像处理
深度学习
weld-tracking
vision system
image processing
deep learning
