论文部分内容阅读
针对椒盐噪声污染图像的降噪,提出一种开关层次中值滤波新算法。该算法由两部分组成:(1)是开关部分,即判定噪声污染图像中的每个像素点是否为噪声像素点,被判定为噪声的像素点的空间位置信息形成噪声定位图像;(2)是层次中值滤波部分,即对噪声定位图像依照某一准则进行分层处理,最后对噪声污染图像执行层次中值滤波。给出该算法的一种实现,并将该实现与标准中值滤波、自适应中值滤波进行仿真对比实验,并与其它中值滤波算法就指标PSNR进行比较,结果均表明该算法降噪效果优良。
Aiming at the noise reduction of the salt-and-pepper noise pollution image, a new algorithm of switch-level median filtering is proposed. The algorithm consists of two parts: (1) a switch part, that is, determining whether each pixel in the noise pollution image is a noise pixel, and the spatial position information of pixels determined to be noise forms a noise localization image; (2) Is the median filter part of the hierarchy, that is, the noise localization image is layered according to a certain criterion, and finally the level median filtering is performed on the noise pollution image. An implementation of this algorithm is given, and the experiment is compared with standard median filter and adaptive median filter to compare with other median filter algorithms. The results show that the algorithm has the same effect of noise reduction excellent.