现代电子技术Issue(9)：51-53,56,4.
一种多子窗口中值加权的高斯噪声滤波算法
Gauss noise filtering algorithm using multi-subwindow median weighting
摘要
Abstract
In order to remove the Gauss noise in an image effectively, the filtering algorithm based on multi⁃subwindow me⁃dian weighting is proposed. The 5×5 filtering windows are applied to the algorithm, and the windows are divided into nine 3×3 sub⁃window areas. The median pixels of every subwindow are found out, then the absolute value of difference between every me⁃dian pixel and its median point is calculated. The weight is calculated by the method of normalization with the average of the ab⁃solute values. The weighted calculation for median of all subwindows is carried out, whose result is taken as the filtering output of center pixel in the filtering window. The experiment results indicate that the proposed algorithm has good filtering performance on Gauss noise images, and keeps the image details well. Its effect is superior to the traditional filtering algorithms.关键词
多子窗口/中值加权/均值滤波/峰值信噪比Key words
multi-subwindow/median weighting/mean filtering/PSNR分类
信息技术与安全科学引用本文复制引用
沈德海,侯建,鄂旭,张龙昌..一种多子窗口中值加权的高斯噪声滤波算法[J].现代电子技术,2015,(9):51-53,56,4.基金项目
国家自然科学基金项目基于博弈论的高效稳定聚类算法研究（61473045）；辽宁省高等学校实验室项目（L2012397）；博士后基金项目（2012M520158）；辽宁省“百千万人才工程”资助项目（2012921058）；教育厅科研一般项目 （）
