Efficient randomized Hough transform for circle detection using novel probability sampling and feature points

Abstract

This paper presents an efficient randomized Hough transform algorithm for circle detection. It optimizes the methods for determining sample points and finding candidate circles. Due to these two optimizations, sampling validity is improved and many false circles are prevented from being regarded as candidate circles. Experimental results demonstrate that the proposed algorithm, which features a strong robustness and high resolution, can dramatically speed up circle detection as compared to other algorithms. It can also be applied to detect ellipses.