一种基于Hough变换的快速圆检测算法  

A Quick Circle Detection Algorithm Based on Hough Transformation

作　　者：王敏 童水光[1] 陈玉辉 从飞云[1] 

机构地区：[1]浙江大学机械工程学院,浙江杭州310027

出　　处：《机械工程与自动化》2018年第1期152-154,共3页Mechanical Engineering & Automation

摘　　要：为加速轮毂轮廓圆的检测速度,提高轮毂自动化加工的效率,提出一种基于Hough变换的改进算法。该算法利用圆的先验几何特性,把采样点到三维参数空间的映射降低至二维,求得映射后距离各拟合直线最近的点,将其作为圆心点完成对轮廓圆的特征提取。实验结果表明:该算法不仅保留了传统Hough变换的优点,而且通过降低参数空间维数,使无效点带来的计算消耗大大降低,提高了检测效率,具有较高的应用价值。In order to accelerate the circle detection and improve the automatic processing efficiency of wheel hubs,this paper proposed an improved algorithm based on Hough transform.This algorithm mapped the sampling points into two-dimensional（2 D）parameter spaces instead of three-dimensional（3 D）ones in the accordance of the priori geometric characteristic of circle.Then,it obtained a point that was the nearest one from the fitting straight lines and regarded the point as a center of a circle then extracted the features of the circle.The experimental results indicate that the algorithm not only retains the advantages of traditional Hough transform,but also considerably reduces the calculation consumption of invalid points by reducing the dimension of the parameter space.In this case,it improves the detective efficiency and presents good application value.

关 键 词：HOUGH变换 圆检测 对称性 机器视觉 图像处理 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]