为解决在管道自动焊接过程中V型坡口的识别定位问题,提出了一种基于激光视觉传感的局部区域内分步式定位方法.首先建立模板匹配,利用模板匹配方式获取焊缝初始位置区域;然后采用阈值分割和边缘提取获得激光条纹边缘线,通过Shi-Tomasi算法对边缘线进行角点检测,得到边缘线上的亚像素角点位置坐标;最后采用最小二乘法拟合得到精确的边缘直线,对上下边界线求平均值提取激光条纹的中心线,求取直线的交点得到坡口轮廓的拐点信息.通过对实际焊接现场拍摄的50幅同一高度不同位置的图像进行检测,结果表明,分步式定位方法抗干扰性强,精度较高,为完成整个跟踪自动焊接过程奠定了基础.%In order to solve the problem of the identification and positioning of the V-groove in the process of automatic welding of the large pipeline,a multiple-step positioning method based on local area of laser vision sensor is presen-ted. Firstly,the template matching is established to obtain the initial position of the welding seam. Then,the laser stripe edge line is obtained by threshold segmentation and edge extraction;the Shi-Tomasi algorithm is used to detect the corner points of the edge lines,and the sub-pixel corner position coordinates on the edge line are obtained. Final-ly,the least squares method is used to get the exact edge line. The center line of the laser stripe is extracted from the upper and lower boundary lines. The inflection point information of the groove contour is obtained by getting the inter-section of the straight line. Fifty images captured from different positions at the same height are inspected. The results show that the multiple-step positioning method has strong anti-interference ability and high precision,which lays the foundation for tracking in the whole automatic welding process.
展开▼
