去除脉冲噪声的自适应开关中值滤波

Adaptive switching median filter for the removal of impulse noise

为消除图像中的脉冲噪声,提出了自适应开关中值(ASM)滤波算法。该算法采用一种新的噪声检测方法将图像中的像素分为信号点和噪声点两类。对检测出的噪声点统计其个数并由此估算图像中的噪声密度,根据估计的噪声密度自适应确定滤波窗口尺寸,采用改进的中值滤波对检测出的噪声点进行处理;而信号点则保留其灰度值不予处理。对ASM滤波进行仿真实验,结果表明,它能在有效去除噪声的同时很好地保护图像细节,较传统中值滤波及其它改进中值滤波算法有更优的滤波性能。

To remove the impulse noise from the image, the Adaptive Switching Median （ASM） filter is proposed. In the algorithm, a new noise detection scheme is adopted to separate the pixels in the image into signal pixels and noise pixels. The number of noise pixels is computed to estimate the noise density. Based on the noise density, the filtering window size is selected adaptively. The impulse noise is removed using the modified median filtering, while signal pixels hold their gray value and are left unprocessed. Simulation results show that the proposed algorithm can suppress noise very effectively, preserve fine details very well and has better filtering performance than the standard median filter and other improved median filters.