基于面结构光的拼接焊缝三维位置提取方法  

3D position extraction method of splicing weld based on surface structured light

作　　者：邱飞达 熊新炎 董静[1] QIU Fei-da;XIONG Xin-yan;DONG Jing(College of Light Industry,Harbin University of Commerce,Harbin 150028,China)

机构地区：[1]哈尔滨商业大学轻工学院,哈尔滨150028

出　　处：《哈尔滨商业大学学报（自然科学版）》2022年第3期325-331,共7页Journal of Harbin University of Commerce：Natural Sciences Edition

摘　　要：针对视觉传感器对大尺寸工件中复杂拼接焊缝提取时存在的精度不高、稳定性不强等问题,通过研究点云重建与二维焊缝识别的方法,提出了一种基于面结构光的工件拼接焊缝三维位置提取方法.该提取方法采用安装于机械手末端、与其同步运动的面结构光视觉传感器对工件表面进行扫描,在测量位置投射一组正弦结构光并由相机进行同步采集,进而恢复出焊缝表面的三维信息并生成深度图.通过对深度图像进行图像处理,识别提取焊缝位置与其起止点位置的二维像素坐标并转化为三维位置坐标,传递给机器人进行焊接.结果表明,该方法能够识别出多种复杂的拼接焊缝,且识别率较高,速度较快,稳定性较强,精度较高,对环境的抗干扰能力强,并能够重建出工件表面与焊缝位置的三维图像,从而为拼接焊缝的自动化焊接提供起止点数据.Aiming at the problems of low accuracy and low stability in the extraction of complex stitching welds from large size workpiece by visual sensors, a 3D location extraction method of stitching welds from large size workpiece based on surface structured light was proposed by studying the method of point cloud reconstruction and two-dimensional weld recognition. In this extraction method, the surface of the workpiece was scanned by a surface structured light vision sensor which was installed at the end of the manipulator and moved synchronously with it. A group of sinusoidal structured light was projected at the measured position and synchronously collected by the camera, so as to recover the three-dimensional information of the weld surface and generate a depth map. Through the image processing of the depth image, the 2D pixel coordinates of the weld position and its starting and ending points were identified and extracted and converted into 3D position coordinates, which were transmitted to the robot for welding. Results showed that the extraction method to identify a variety of complex splicewelds, and the recognition rate was higher, faster, stronger stability, high precision, strong anti-jamming capability, to the environment and be able to rebuild on the surface of the work and the position of weld three-dimensional images, thus provided load-point data splicing welding seam automatic welding.

关 键 词：面结构光 三维重建 图像处理 焊缝识别 自动化焊接 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]