基于机器视觉的钢轨对接焊高精度检测方法  

High-precision Detection Method for Rail Butt Welding Based on Machine Vision

作　　者：陈庚 张伟[1] 李存荣[1] 邹振东 任勇涛 CHEN Geng;ZHANG Wei;LI Cunrong;ZOU Zhendong;REN Yongtao(School of Mechanical and Electronic Engineering,Wuhan University of Technology,Wuhan 430070,China)

机构地区：[1]武汉理工大学机电工程学院

出　　处：《铁道标准设计》2020年第3期75-80,共6页Railway Standard Design

基　　金：国家自然科学基金项目(71171154);湖北省科技支撑计划项目(2015BAA063)

摘　　要：为提高闪光对接焊设备焊接过程的焊接精度,提出一种基于机器视觉和图像处理技术结合的钢轨对接焊高精度定位检测方法。该方法采用黑白面阵CMOS工业相机、线激光器作为主要检测硬件,其中线激光器发射的激光分别垂直投射在与轨道轴线平行的轨顶面和轨侧面上,通过对工业相机所采集的图像进行特征提取,针对多种不同特征进行模式识别,从而获得钢轨顶部和侧面高度差。通过重复性与可靠性试验,表明该方法满足工程实际检测要求。与传统的手工检测方法相比,本方法具有更高的精度,且受外部环境的影响较小,满足工况环境条件下的焊接精度要求,在复杂环境下同样可获得较好的检测效果。In order to improve the welding precision of the flash butt welding equipment, a high precision positioning detection method based on the combination of machine vision and image processing technology is proposed. The black and white area array industrial camera and line laser are used as the main detection hardware in this method, the laser light emitted by the line laser is vertically projected on the top surface and the rail side of the rail parallel to the track axis, and the image acquired by the industrial camera is extracted to perform pattern recognition for a plurality of different features, thereby obtaining the difference between the top and the side height of the rail. The repeatability and reliability test shows that the method meets the actual testing requirements of the project. Compared with the traditional manual detection method, this method has higher precision and is less affected by the external environment. It meets the welding precision requirements under working conditions and also obtains better detection effects in complex environments.

关 键 词：机器视觉 对接焊 图像处理 检测精度 钢轨 

分 类 号：U216.3[交通运输工程—道路与铁道工程]