光笔式视觉坐标测量中控制点光斑图像的识别

Control points' light spot image recognition in light-pen vision coordinate measurement

介绍了一种光笔式单摄像机便携三维坐标视觉测量系统。在分析该系统成像特点的基础上,提出了一种控制点椭圆形光斑图像的数字识别方法。常用的方法是在CCD摄像机镜头上加滤光片来隔离背景干扰亮斑图像,该方法以光笔上被测控制点在摄像机CCD像平面上的光斑图像轮廓所占像素个数为基准,剔除过大和过小的图像轮廓,进而采用基于椭圆线轮廓度误差的评定方法实现控制点椭圆形光斑图像的识别。给出了该方法的具体算法,实验结果表明该方法简单方便,易于实现,检测速度较快,适用于圆形被测控制点的视觉检测系统,具有一定的理论意义和工程实用价值。

A light-pen single-camera 3D coordinate vision measurement system is introduced in this paper. By analysising characteristics of the system image, a digital method for recognizing the elliptical light spot images of the measured control points is proposed. On the basis of the pixel number of a measured control point＇ elliptical light spot image contour, too large and too small light spot images can be eliminated. Then, using the method based on the evaluation for elliptical profile error, the measured control points＇ elliptical light spot image can be recognized. The algorithm for this method is given, and the experimental results show that this algorithm is simple, fast, easy to realize and practical to the vision inspection system with circular measured control points. Compared with the common method, which applies the filter mounted on the CCD camera lens to segregate the noise bright spot image in the image background,it has certain theoretical significance and practical project value.