基于排序统计的图像边缘增强滤波

主要研究了如何对受混合噪声污染的模糊图像进行边缘增强滤波，并提出了一种基于排序统计的图像边缘增强滤波算法，该算法将输入样点集划分成两个具有不同灰度值的子集，通过子集的运算值之差判断边缘是否存在。如果边缘存在，则通过判断滤波窗位于边缘的哪一侧来选择某个子集的输出运算值以增强边缘的梯度；如果边缘不存在，则考虑进行噪声的平滑，计算机模拟实验表明，基于排序统计的图像边级增强滤波器在增强边缘的梯度的同时，能

ORDER STATISTICAL EDGE ENHANCING FILTERS FOR NOISY IMAGES

This paper studies mainly how to enhance the edges of noisy fuzzy images corrupted by all kinds of noises.A nonlinear order statistical edge enhancing filter is presented.This algorithm divides the input sample set into two subsets which have different grey levels and decides edges by calculating the difference of the outputs of the two subsets.If there exists an edge,by deciding on which side of the edge,the filter window locates,an output of some subset is chosen to enhance the gradient of the edge.If no edge exists,smoothing noises is taken into account.Computer simulations show that the influence of Gaussian noise and impulsive noise will be eliminated by the order statistical edge enhancing filter while the filter enhances the gradients of edges.This filter will outperform other edge enhancing operators.