针对靶场视觉测量相机标定中圆形标志在复杂背景图像上的提取和定位问题,研究了两种不同的圆形标志提取方法,一种是基于感兴趣区域的交互式提取方法,另一种是基于图像灰度和几何形态学的提取方法。对两种方法的圆形标志提取过程进行了研究,并分别采用两种方法对视觉测量图像中的圆形标志进行提取实验,对两种方法的边缘提取效果和定位结果进行了对比,分析了两种方法的优势和不足。圆形标志的提取结果表明,第一种方法能够平滑地提取出圆形标志的边缘,但受圆形标志的成像情况影响较大,第二种方法边缘提取效果受圆形标志的成像情况影响较小,但提取出的边缘为锯齿形,部分圆形标志的像素点被误判为背景,与圆形标志的实际成像有一定误差。

To solve the problem of extracting and locating the circular marks from complex background in the image of range visual measurement,two extraction methods for circular marks are studied.The first method is the interactive extraction method based on the region of interest and the second is the extraction method based on image gray scale and geometric morphology.In this paper,the two methods are introduced at first,and then the methods are adopted to extract the circular marks from the image of visual measurement.The edge extraction and positioning results are compared,and the advantages and disadvantages of the two methods are analyzed.The result shows that the first method can extract the smooth edge of the circular mark,but this method is affected by the image of the circular mark.The second method is less affected by the image of the circular mark,but the extracted edge is serrated,and part of the circular mark pixels are misidentified as background pixels,which is not in conformity with the actual situation.