基于极值的椒盐噪声滤波改进算法

为了在有效滤除椒盐噪声的同时更好地保护图像细节，提出一种基于极值的椒盐噪声滤波改进算法.算法首先进行噪声检测，将灰度值为0和255附近的像素点，且不构成5像素或以上直线的点作为噪声点，其余点作为信号点；然后进行噪声滤波，为了保护图像中的边缘、细节或细线，信号点不做任何处理，而对噪声点使用梯度法进行处理.Matlab仿真实验结果表明，新算法不仅能有效滤除椒盐噪声，在保护细节方面也取得了优于传统中值滤波算法的效果.

Improved Algorithm Based on the Extreme Value of Salt and Pepper Noise Filtering

In order to effectively remove salt and pepper noise and protect image details, this paper proposes an improved algorithm based on the extreme value of salt and pepper noise filtering. Firstly, the algorithm detects noise, each pixel is classified to be possible noise pixel and signal pixel, and the value of possible noise pixels is near 0 and 255, and does not constitute 5 pixels or more straight points, the remaining points is as the signal point. Then, the noise filtering is taken, in order to protect the image edge details, or thread, signal pixel without any treatment, and the noise pixels using the gradient method for processing. The experimental results show that the new algorithm not only can effectively remove salt and pepper noise, and effect made in detail preservation is superior to traditional median filtering algorithm.