一种新的图像椒盐噪声的非线性滤波算法

为了在滤除椒盐噪声的同时能很好地保持图像的边缘细节，提出了一种新颖的图像椒盐噪声非线性滤波算法。利用局部统计信息，先将图像像素点分为信号点和可能的噪声点两类。然后将可能的噪声点进一步细分为边缘点、噪声点和信号点：利用方向信息、均方差来判断是否为边缘点，利用自适应阈值的方法来判断是否为噪声点，并且对边缘点和噪声点采取不同的方法进行滤波。经过仿真实验并与其它滤波算法进行比较表明，文中的算法具有更好的效果。

A New Nonlinear Algorithm for Removal of Salt-and-Pepper Noise from Images

In order to achieve very good salt-and-pepper noise removal as well as preserving image edge details,a new nonlinear algorithm is proposed in this paper.Utilizing the statistical information in a local window of the image,each pixel is classified to be signal pixel and possible noise pixel.Then the possibel noise pixels are subdivided into edge pixels,noise pixels and signal pixels: an edge pixel is judged according to direction information and average variance,a noise pixel is judged according to an adaptive threshold.Different filtering schemes are employed for edge pixels and noise pixels.The simulation results show that this algorithm outperforms is better than other algorithms.