基于模糊随机广义霍夫变换的目标匹配

Object matching with fuzzy randomized generalized Hough transform

为了提高工业检测中图像匹配精度和速度,提出了一种用于二维目标匹配的新算法——模糊随机广义霍夫变换(FRGHT)。此算法结合了模糊推理系统(FIS)和随机广义霍夫变换(RGHT)。模糊推理系统引入模糊集合概念,计算待配准图像中边缘点对配准参数的投票,从而可以抑制噪声,解决扭曲问题,提高了匹配精度;随机抽取待配准图像中边缘点进行投票,实现了多对一的映射,从而减少了内存需求,提高计算速度。实验表明,该方法计算速度快,匹配精度高,不受噪声污染、扭曲、遮挡、混乱等情况的影响。

A new algorithm called Fuzzy Randomized Generalized Hough Transform （FRGHT） was proposed to improve the industrial detection accuracy and the speed of image matching in this paper. This algorithm combined Fuzzy Inference System （FIS） and Random Generalized Hough Transform （RGHT）, in which fuzzy sets of FIS were used to compute the votes of edge points of reference image for registration parameters, can effectively solve the problem of noise and distortion and improves the matching accuracy; and the random sampling giving a many-to-one mapping reduces the memory requirements and improves the matching speed. The experiments demonstrate that the proposed algorithm exhibits faster speed and higher accuracy than RGHT and Fuzzy GHT （FGHT）, moreover it is robust to the serious noise pollution, distortion, occlusions, clutter, etc.