基于改进随机Hough变换的混合圆/椭圆快速检测方法

Hybrid Circle and Ellipse Fast Detection Using Improved Randomized Hough Transform

针对随机Hough变换 (RHT)在复杂图像中检测圆时随机采样所造成的大量无效采样和无效累积问题 ,提出了一种新的基于连续曲线段的RHT(CRHT)混合圆 椭圆检测方法 ,它首先利用连续的边缘曲线段来优先决定随机采样点的数据 ,再利用不准确的梯度方向信息来决定是按照圆还是椭圆对采样到的几个点进行参数累积 ,一方面它较好地解决了标准RHT检测中无效采样概率大、无效积累时间多的问题 ,另一方面也解决了圆与椭圆的混合检测问题。实验结果表明本文的算法具有较高的计算速度 。

To solve the problem of a large amount of invalid sample and accumulations yielded by random sampling when randomized Hough transform(RHT) is used to detect circles in complex images processing, an new improved CRHT is presented. It uses not only gradient direction information but also uses continue-curve image edge to determine whether the parameter should be used and accumulated or not. The problem of invalid sampling and accumulations and hybrid circle/ellipse detection is well solved. The experiment results show that our method has higher speed, smaller storage and better detection performance.