关键词: 椒盐噪声, 噪声检测, 动态窗口, 自适应策略, 滤波算法

Abstract: Aiming to solve the problem that median filtering algorithm has poor processing capacity on high-density noise image and delicate texture image, a dynamic window-based adaptive median filter algorithm was proposed. According to the associated level between noisepoint information and its surrounding, the new algorithm adjusted the noisepoint filter value, which can deal with the details of the images better. The adaptive strategies of the new algorithm strengthen the denoising performance of the filtering algorithm, which is good at dealing with any density noise. The simulation analysis shows that the new algorithm can effectively improve the peak signal to noise ratio of the image , and the denoising effect is more satisfactory than other methods.

Key words: salt-and-pepper noise, noise detection, dynamic window, adaptive strategy, filter algorithm