一种灰色理论和中值滤波结合的滤波算法

摘要

结合灰色理论和中值滤波算法,提出了一种新的去椒盐噪声算法.算法采用开关策略,如果滤波窗口中心点为噪声点,则根据窗口内非噪声点数目自适应调整窗口大小,并利用非噪声点的灰度值和它们的均值分别作为比较序列和参考序列计算各点的关联度,将窗口各非噪声点灰度值与关联度加权求和,如果结果值为正常灰度值,替换中心噪声点的灰度值,否则,采用窗口内非噪声点的中值替换中心噪声点的灰度值.验证实验表明,和其它几种算法比较,该算法具有更强的去噪能力和边缘保护效果.%A new algorithm for removing salt and pepper noise is proposed based on grey theory and median filtering algorithm.The algorithm uses switching strategy,if the filter window center is noise point,adjusts the window size adaptively according to the number of non-noise,and uses these non-noise points gray and their mean value as comparison sequence and reference sequence to calculate correlation degree,then weighted these non-noise points gray together with their correlation degree,if the result is normal gray value,replaces the center noise point gray,otherwise,uses the median gray of non-noise points in the filter window to replace center noise point gray.The experimental results show that the proposed algorithm has better denoising ability and edge protection effect compared with other algorithms.