基于Hough变换的圆检测算法的改进  

Improvement of circle detection algorithm based on Hough transform

作　　者：龚昕[1] 张楠 GONG Xin;ZHANG Nan(Chinese People’s Liberation Army 92493,Huludao 125000,Liaoning Province,China;The 32nd Research Institute,China Electronics Technology Group Corporation,Shanghai 201808,China)

机构地区：[1]中国人民解放军92493部队,辽宁葫芦岛125000 [2]中国电子科技集团公司第三十二研究所,上海201808

出　　处：《信息技术》2020年第6期89-93,98,共6页Information Technology

摘　　要：圆形的检测是图像处理领域的一个重要问题,是各种图像分析得以顺利进行的基础。传统的圆形检测方法为利用Hough变换将二维像素点映射到高维的Hough空间,然后以一定的半径步长对得到的可能的圆心做检测分析,存在抗干扰力差,计算量大,提取准确率低的问题。为此提出在Hough算法的基础上利用圆形本身的几何特性,由圆上点确定圆心与半径,减少计算量;同时引入了聚类分析的思想,合并相似度高的圆形,有效减少了提取结果中由于边缘模糊而造成的多个假圆的问题。经实验验证该算法可以在降低计算量的同时提高Hough变换的准确率。The detection of circles is an important issue in the field of image processing,and is the basis for the smooth implementation of various image analysis.The traditional circle detection method uses Hough transform to map 2D pixels to high-dimensional Hough space,and then detects and analyzes the possible circle center with a certain radius step,which is with poor anti-interference,large amount of calculation,and low accuracy.Therefore,it is proposed to use the geometric characteristics of the circle itself based on the Hough algorithm to determine the center and radius of the circle from the points on the circle to reduce the amount of calculation.At the same time,the idea of cluster analysis is introduced to merge circles with high similarity,which effectively reduces the problem of multiple false circles due to blurred edges in the extraction results.The experiment proves that the algorithm can reduce the amount of calculation and improve the accuracy of Hough transform.

关 键 词：图像处理 HOUGH变换 圆检测 几何分析 聚类算法 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]