一种去除椒盐噪声的自适应模糊中值滤波算法

An adaptive fuzzy median filtering algorithm for salt and pepper noise removal

针对传统中值滤波算法降噪性能不佳以及易造成图像细节模糊的问题,提出了一种自适应模糊中值滤波算法。滤波过程分为噪声检测和噪声去除2 个阶段。噪声检测阶段:采用极值法检测噪声,将图像的像素点分为两类,即疑似噪声点和信号点。通过疑似像素点和与其相邻的已处理的像素点的平均绝对灰度差值定义模糊隶属度函数,利用该函数对疑似噪声点是否为噪声进行模糊分类。噪声去除阶段:信号点保持原值输出,对于疑似噪声点的3 种分类结果,采用模糊加权的中值滤波器进行统一处理。实验结果表明,较于多种传统滤波方法,该算法能更有效地去除椒盐噪声,保护图像细节。

An adaptive fuzzy median filtering algorithm is proposed for the problem that traditional median filtering algorithm has poor processing performance in denoising and easily blurs image details. The process of denoising includes two stages: noise detection and noise removal. In noise detection stage, using the maximum-minimum method to detect noise, pixels in a corrupted image are classified into two categories: suspected noise pixels and signal pixels. The fuzzy membership function is defined by the mean of the absolute gray difference between the suspected pixels and its neighborhood pixels which has been processed, and the suspected pixels are fuzzy classified by this function. In noise removal stage, the signal pixels retain unchanged;For the three classification results of suspected noise points, the fuzzy weighted median filter is used for processing. The experimental results show that compared with several traditional filtering algorithm, the proposed algorithm can reduce salt and pepper noise and preserve image details more effectively.