一种用Hough变换检测圆的快速算法*

为了克服传统Hough变换检测圆时耗时巨大的缺陷,给出了一种新的基于Hough变换检测圆的快速算法.新算法与传统的方法相比具有以下特点:计算量少,提高了检测的速度;保留了传统Hough变换识别率高、抗噪性强、对不完整边缘具有鲁棒性等所有优点;不需要任何特殊的限定条件.实验表明,新的快速算法可以快速进行目标识别,在实时目标识别系统中具有良好的表现.

New fast algorithm of Hough transform detection of circles

This paper proposed a new fast algorithm for the detection of circles, which improved the problem of huge time consuming when using the traditional Hough transform. The new algorithm had three advantages: low time consuming; hold all the advantages of the traditional Hough transform such as high detection rate, robust to noise and fragmentary boundaries; did not need any special qualification. Experimental results demonstrate that the new fast algorithm can be used for fast target recognition, which has a good performance in real time system.