去除图像椒盐噪声的自适应迭代均值滤波算法  

Adaptive Iterative Mean Filtering Algorithm for Removing Salt and Pepper Noise

作　　者：郭慧娟[1] 白文江 岳云霄 林菲 白雪飞[4] GUO Huijuan;BAI Wenjiang;YUE Yunxiao;LIN Fei;BAI Xuefei(Department of Computer Science and Technology,Taiyuan Normal University,Jinzhong 030619,China;Department of Computer Science and Engineering,Taiyuan University,Taiyuan 030032,China;Research Center for Scientific Development in Fenhe River Valley,Taiyuan Normal University,Jinzhong 030619,China;School of Computer and information Technology,Shanxi University,Taiyuan 030006,China)

机构地区：[1]太原师范学院计算机系,山西晋中030619 [2]太原学院计算机科学与工程系,山西太原030032 [3]太原师范学院汾河流域科学发展研究中心,山西晋中030619 [4]山西大学计算机与信息技术学院,山西太原030006

出　　处：《太原学院学报（自然科学版）》2020年第3期23-28,共6页Journal of TaiYuan University:Natural Science Edition

基　　金：国家社科基金青年项目(17CZS003);国家自然科学基金项目(61703252);山西省教育厅2019年度人文社会科学重点研究基地项目(20190123);山西省应用基础研究计划项目(201701D121053)。

摘　　要：针对传统均值滤波算法不能有效去除椒盐噪声的问题,提出了一种自适应迭代均值滤波算法(Adaptive iterative mean filtering,AIMF)。该算法通过窗口内未污染点的个数自适应地由小到大选择滤波窗口(3×3,5×5,7×7),同时对噪声点采用多轮迭代去噪。实验结果表明,在峰值信噪比标准下,本算法较传统滤波算法有明显优势,特别对去除高密度噪声效果显著,即使在噪声污染程度90%时,依然有良好的图像恢复能力。Adaptive iterative mean filtering(AIMF)algorithm is proposed to overcome the drawbacks of classic mean filter.AIMF may automatically select the filtering window(3×3,5×5,7×7),the noisy points were processed iteratively to improve the denoising effect.The experimental results show that under the PSNR standard,the proposed algorithm has obvious advantages over the traditional filtering algorithm,especially for removing high-density noise.Even when the noise pollution level is 90%,it still has good image recovery ability.

关 键 词：中值滤波 均值滤波 椒盐噪声 图像去噪 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]