一种概率密度的中值滤波算法

提出了一种新的图象噪声消除算法,该方法可以快速有效地去除图象中的脉冲噪声,并保留图象的细节.该算法主要通过利用象素点的值在滤波窗口出现的概率密度信息,进行噪声的判定及滤波过程的介绍,对本算法进行了分析仿真,结果表明该算法去除噪声与保留细节的效果好,且运算速度较快.

A New Median Filtering Algorithm Based on Probability Density

In this paper we introduce a novel image denoising algorithm,the efficient signals are reserved when the noises are removed through the algorithm.The filters are based on local probability density.First,all the pixels are separated into signal pixels and noise pixels according to the decision criterion given in the following;then,noise pixels are replaced with the median value of their neighborhood in the input image.The decision criterion :if a pixel value is the extreme(max or min) of its neighborhood,it is a noise pixel;else where,it is a signal pixel.If a pixel value is higher or lower than the other value of its neighborhood,we may consider that it had been contaminated with noise.Other wise,we consider that it represents an effective signal.Experimental results show that the assumption fits the facts quit well.