双目极线约束下的储罐环焊缝定位算法  

Binocular Localization Algorithm for Circular Seam with Epipolar Constraint

作　　者：金东升 曹文田 张英杰[1] JIN Dongsheng;CAO Wentian;ZHANG Yingjie(School of Mechanical Engineering,Xi′an Jiaotong University,Xi′an 710049,China;Xi′an Shuangtian Electromechanical Equipment Co.,Ltd.,Xi′an 710075,China)

机构地区：[1]西安交通大学机械工程学院,西安710049 [2]西安双田机电设备有限公司,西安710075

出　　处：《机械科学与技术》2025年第1期1-8,共8页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金面上项目(51875445)。

摘　　要：精准定位焊缝轨迹是实现自动化焊接的前提,该文依托双目视觉系统,以储罐环焊缝为焊接目标,研究了基于双目极线约束的焊缝定位算法。首先分析了双目系统成像模型以及三角测量原理,为后续实现焊缝的三维定位提供了理论依据。随后研究了基于灰度的焊缝特征提取算法,结合边缘检测、阈值分割、轮廓提取以及骨架提取等方法有效地提取出焊缝特征。并基于极线约束实现了左右图像焊缝特征的匹配,简化了匹配流程并降低了计算量,提高了匹配效率。实验结果表明:对焊缝特征点的三角测量平均误差为0.140106,能够满足实际工程需要。Accurate positioning of weld seams is a prerequisite for achieving automated welding.This paper takes the circular seam of storage tank as the welding target and investigates the weld seam positioning algorithm based on epipolar constraint,based on binocular vision system.Firstly,the imaging model of the binocular system and the principle of triangulation measurement are analyzed,providing a theoretical basis for the subsequent three-dimensional positioning of weld seams.Then,the weld seam feature extraction algorithm based on grayscale is studied,which effectively extracts weld seam features by combining edge detection,threshold segmentation,contour extraction and skeleton extraction.Furthermore,the epipolar constraint is used to match the weld seam features of the left and right images,simplifying the matching process and reducing computational costs,thereby improving matching efficiency.The experimental results show that the average error of triangulation measurement for weld seam points is 0.140106,which meets the requirements of practical engineering.

关 键 词：双目视觉 焊缝提取 视觉定位 极线约束 三角测量 

分 类 号：TG156[金属学及工艺—热处理]