摘要
以机械加工零件中常见的带孔平面工件为研究对象,采用透视可变形模板匹配的识别方法,Blob分析与Hough检测边缘相结合的定位方法。对于如何得到图像对中工件圆特征中心的同名点像素坐标,提出了一种新思路:首先在模板图像上绘制ROI轮廓,将工件的局部特征分割成单独的区域,这种分块化缩小图像定义域的方式,可以加快识别速度,接着在左右图像中投影变换出搜索区域,然后采用上述的定位算法得到圆心坐标,最后对目标点进行三维重建。通过对工件厚度测量的实验结果表明,Z轴重建精度为±0.108 mm,在视觉测量领域具有较好的应用前景。
Taking the common plane workpiece with holes in machining parts as the research object,this paper adopts the identification method of perspective deformable template matching,and the positioning method combining Blob analysis and Hough detection edge.A new idea is proposed for how to obtain the pixel coordinates of the point with the same name at the center of the workpiece circular feature in the image pair:first,draw a ROI on the template image to divide the local features of the workpiece into separate regions.The method can speed up the recognition speed,and then project and transform the search area in the left and right images,and then use the above positioning algorithm to obtain the coordinates of the center of the circle,and finally carry out three-dimensional reconstruction of the target point.The experimental results of workpiece thickness measurement show that the reconstruction accuracy of Z axis is±0.108 mm,which has a good application prospect in the field of robot grasping.
作者
何敏军
张晓玲
张稀柳
陈胜
HE Minjun;ZHANG Xiaoling;ZHANG Xiliu;CHEN Sheng(School of Electrical Information Engineering,Jiangsu University of Technology,Changzhou Jiangsu 213001,China;Shanghai Ximing Vision Technology Co.,Ltd,Wuxi Jiangsu 214000,China)
出处
《激光杂志》
CAS
北大核心
2023年第7期199-204,共6页
Laser Journal
基金
江苏理工学院研究生实践创新项目(No.XSJCX21_57)。
