基于激光视觉的多层多道焊接坡口特征点提取方法  

Method for Extracting Multi-Pass Welding Groove Feature Point Based on Laser Vision

作　　者：顾帆[1] 陈华斌[1] 何银水[1] 陈善本[1] 

机构地区：[1]上海交通大学材料科学与工程学院,上海200240

出　　处：《上海交通大学学报》2016年第S1期62-65,共4页Journal of Shanghai Jiaotong University

摘　　要：多层多道自适应焊接一直是实现中厚板焊接自动化的难点.文章研究了多层多道焊焊接坡口特征点提取方法,通过对焊接坡口激光条纹图像特征分析,首先采用中值滤波去除图像噪声,利用阈值分割和重心法提取激光条纹中心线,然后经过中心线拟合直线求交点和逐列搜索对比获取焊接坡口轮廓的四个拐点,为后续多层多道自适应焊接提供了可行性技术途径.Adaptive multi-pass welding has always been the difficulties for realizing welding automation.This paper studies the multi-pass welding groove feature point extraction method,with analyzing laser welding groove stripe image feature,first through median filtering to remove image noise,extracting the laser stripe centerline by threshold segmentation and center of gravity method,and then fitting a straight line and the center line of the intersection by searching column points,obtaining four corners welding groove profile.As a result,this paper provided the feasibility of technical approach for the subsequent adaptive multi-pass welding.

关 键 词：多层多道焊接 特征提取 图像处理 激光视觉 

分 类 号：TG409[金属学及工艺—焊接]