双边加权滤波算法-中国图象图形学报.DOC

中图法分类号：TP301.6 文献标识码：A 文章编号：1006-8961(200 ) - - 论文引用格式： 沿边局部灰度差分椒盐噪声滤波 孔勇奇1,潘志庚2,3 （1.浙江商业职业技术学院, 杭州 310053； 2. 杭州师范大学 数字媒体与人机交互研究中心 摘 要：为了在减小高密度椒盐噪声漏检的同时最大限度提升滤波效果,提出一种新算法。该算法首先利用椒盐灰度极值，抽取准噪声点；然后对准噪声点执行局部窗口沿边检测，判断疑似噪声集合；最后，只对疑似噪声集合中像素点进行就近双边加权灰度替换滤波。实验数据显示，不同噪声密度环境下，本文算法在现有滤波算法的基础上，将峰值信噪比参数PSNR分别提高0.2到1.6分贝。实验结果表明，该算法的综合性能优于其他相关的 Salt and pepper noise filtering algorithm based on local border gray-scale difference Kong Yongqi1, Pan Zhigeng2,3 (1. Institute of information technology, Zhejiang Vocational College of Commerce, Hangzhou 310053; 2. Digital Media and HCI Research Center, Hangzhou Normal University, Hangzhou 310036; 3.Digital Entertainment and Animation Research Center, Zhejiang University, Hangzhou 310058) Abstract: In order to reduce the detecting failure in high density salt and pepper noise and improve the filtering performance, a new local adaptive gray similarity bilateral filter algorithm is proposed in this paper. Firstly, the algorithm extracts quasi-noise through gray-scale extremism of salt and pepper gray value, and then detects local window border for every quasi-noise point to get the possible noise collection, and finally, only the pixels in the quasi-noise collection need to be filter with nearby bilateral weighted gray-scale filtering. Experimental results demonstrate that the new algorithm has increased the parameters for peak signal-to-noise ratio(PSNR) by 0.2 to 1.6 DB in different noise density. Experimental results demonstrate that the new algorithm has the most integrated optimal performance with comparison to other filtering algorithms. Key words: Gray-scale Difference；Based on Local Border；Principle of the nearest distance；salt and pepper noise 0 引 言 噪声滤波是图像处理的一个重要方面，滤波效果的好坏直接影响到后期图像处理的质量，椒盐噪声是一种在图像产生、信道传输、解码处理等过程中产生的典型图像噪声[1,2]。针对椒盐噪声的滤波处理，研究人员提出了许多算法。其中，标准中值算法(TM)[3]是一种最早被提出的椒盐噪声滤波算法，它针对污染图像，选取一定长度并包含噪声点的矩形区域作为滤波窗口，对滤波窗口内像素点的灰度值进行排序，采用排序后的中值进行噪声点的灰度值替换。该算法在低噪声密度情况下，表现出一定滤波效果，但是随着噪声密度的增加，污染图像引入较多噪声点，此时TM算法的滤波质量