摘要： 针对现有算法对高密度椒盐噪声滤波不足的问题，提出一种新的高密度椒盐噪声滤波算法。通过噪声检测将含噪图像的像素分为信号点和噪声点，对每一个椒盐像素，计算以该像素为中心的窗口内非椒盐像素中值，信号点则保持其灰度值不变直接输出，重复以上过程，直到没有噪声点被替换。实验结果表明，该算法能在有效去除椒盐噪声的同时保护图像细节，较传统中值滤波及其改进算法有更好的滤波性能。

关键词: 高密度噪声, 椒盐噪声, 中值滤波, 噪声检测, 噪声滤波

Abstract: Aiming at drawbacks of the poor de-noising capacity for high-density noised images, a novel filtering algorithm of high density salt and pepper noise is proposed in this paper. The algorithm is adopted to separate the pixels in the image into signal pixels and noise pixels. It calculates the median value of the non salt and pepper pixels in the window centered at the current salt and pepper pixel, while signal pixels hold their gray values and be output unprocessed. The steps above are repeated until there is no noise replaced. Experimental results show that the algorithm can suppress noise effectively and preserve the image details. Contrasted to the traditional median filtering algorithm and its improved algorithms, it has better filtering performance.

Key words: high density noise, salt and pepper noise, median filtering, noise detection, noise filtering

中图分类号: 

• TP391.41