基于邻域信息与像素概率的椒盐噪声滤波算法  

Salt and Pepper Noise Filtering Algorithm Based on Neighborhood Information and Pixel Probability

作　　者：余思琛[1] 成春晟[1] 万天中 YU Sichen;CHENG Chunsheng;WAN Tianzhong(The 28th Research Institute of China Electronics Technology Group Corporation,Nanjing 210007,China)

机构地区：[1]中国电子科技集团公司第二十八研究所,江苏南京210007

出　　处：《电子质量》2025年第2期1-6,共6页Electronics Quality

摘　　要：为了滤除高密度椒盐噪声,且避免使用较大滤波窗口,提出了一种基于邻域信息与像素概率的椒盐噪声滤波算法。首先,利用极值法检测出椒盐噪声点;其后,利用邻域信息自适应选择滤波器窗口大小,将已滤波噪声点视为有效信息,避免噪声密度高时扩大滤波窗口;最后,根据窗口内像素概率,选取概率最大的像素修复椒盐噪声点。对比实验表明,即使在高密度噪声情形下,该算法也具有较高的信噪比与较低的均方误差,能够更好地滤除数字图像中的椒盐噪声。In order to filter out high-density salt and pepper noise and avoid using a large filter window,a salt and pepper noise filtering algorithm based on neighborhood information and pixel probability are proposed.Firstly,the extreme value method is used to detect the salt and pepper noise points.Then,the neighborhood information is used to adaptively select the filter window size,and the filtered noise points are regarded as effective information to avoid the expansion of the filter window when density of noise is high.Finally,according to the pixel probability in the window,the pixel with the highest probability is selected to repair the salt and pepper noise point.Comparative experiments show that even in the case of high-density noise,the algorithm has a higher signal-to-noise ratio and lower mean square error,and can better filter out salt and pepper noise in digital images.

关 键 词：椒盐噪声 邻域信息 像素概率 自适应滤波器 滤波算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]