模糊随机Hough变换算法

提出一种新颖的将模糊理论与随机Hough变换相结合实现参数曲线自动检测的方法，即模糊随机Hough变换．与其他随机Hough变换方法相比，该方法可以处理背景噪声较强、曲线有一定交叠和缺损的图像，检测出的曲线精确度更高，能够有效地避免检出错误曲线的情况．为了叙述的简便，仅以直线和圆为实例，但可以推广到任意曲线的检测．

A New Improvement of Randomized Hough Transform Based on Fuzzy Theory

A new curve detection approach, to detect shapes in noisy data, called the Fuzzy Randomized Hough Transform (FRHT) was heuristically proposed, inspired by the efforts of using fuzzy and random techniques for curve detection. FRHT finds shapes by approximately fitting the data points, which avoids the spurious shapes detected when using the Randomized Hough transform (RHT). An efficient implementation of the algorithm was found in detecting lines and circles, which was tested by experiments on images with various kinds of strong noise.