加权变分的图像去噪算法  

Image de-noising algorithms based on weighted variation

作　　者：陈利霞[1,2] 冯象初[1] 王卫卫[1] 宋国乡[1] 

机构地区：[1]西安电子科技大学理学院,陕西西安710071 [2]桂林电子科技大学数学与计算科学学院,广西桂林541004

出　　处：《系统工程与电子技术》2010年第2期392-395,共4页Systems Engineering and Electronics

基　　金：国家自然科学基金(10861005;10501009)资助课题

摘　　要：针对经典的总变分去噪模型边缘信息对噪声敏感且易模糊的缺陷,提出了非线性与线性的加权变分模型。非线性加权变分模型是在总变分模型的正则项中引入权函数,并利用权函数引导扩散,使得新模型在消噪的同时更好地保持图像的纹理特征和边缘信息;线性加权变分模型是对含噪图利用高斯函数进行预处理,再对处理后的图像进行扩散,从而降低计算复杂度。数值实验表明,与经典的总变分模型相比,改进的方法无论是在视觉效果还是峰值信噪比上都有明显的提高。View on the weakness of a classical variational de-noising model in which edge information is sensitive to noises and prone to blur,two improved nonlinear and linear weighted variational algorithms are put forward.In the nonlinear weighted variational model,a weight function is introduced in the regularization term of the classical model to induct diffusion,which gives the result that the new model preserves the texture characteristics and the edge information better while removing noises.In the linear model,the Gaussian function is used to smooth the noised image before diffusion,which reduces the computational complexity.Compared with the classical model,the experimental results show improvements of both the proposed models.

关 键 词：偏微分方程 总变分 图像去噪 权函数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]