基于灰色关联度的图像混合噪声的自适应滤波算法  

An Adaptive Filter Approach for Removal of Mixed Noise Based on Grey Relevancy

作　　者：黄春艳[1] 张云鹏[1] 黄红艳[2] 

机构地区：[1]华北水利水电学院,河南郑州450011 [2]辽宁石油化工大学,辽宁抚顺113001

出　　处：《微电子学与计算机》2010年第2期126-128,共3页Microelectronics & Computer

基　　金：河南省教育厅自然科学研究项目(2008B120004);华北水利水电学院青年科研基金(HSQJ2009017)

摘　　要：利用中值滤波和灰色关联度的特点,提出基于中值滤波和灰色关联度相结合的混合噪声图像滤波算法.算法选取加窗混合噪声图像的中值,采用灰色关联度自适应地计算各像素的加权系数,通过加权得到结果.实验结果表明,该算法有较好的自适应性,不但能够有效去除含有高斯噪声和脉冲噪声的图像噪声,而且能较好地保护图像的细节信息,提高图像的去噪效果和清晰度.Using the characteristic of median filter and grey relevancy, an algorithm for mixed noise filtering is presented based on median filtering and grey relevancy. A median pixel is determined in the filter window, and the coefficient of the proposed method can be varied adaptively based on degree of the grey relevancy, then the values of the processed pixel can be calculated. Experiment results show that the proposed method can remove mixed Gaussian and impulse noise efficiently preserve the integrity of edge and detail information and remove mixed noise effectively.

关 键 词：灰色关联度 混合噪声 自适应滤波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]