瑞利分布乘性噪声去除的非局部TV模型  

Nonlocal TV model for multiplicative noise with Rayleigh distribution removal

作　　者：李帅[1] 赵希梅[1] 王国栋[1] 

机构地区：[1]青岛大学信息工程学院,青岛266071

出　　处：《仪器仪表学报》2015年第7期1570-1576,共7页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61303079;61305045)项目资助

摘　　要：基于局部算子的TV模型用于瑞利分布乘性噪声去除时往往存在边缘模糊、纹理模糊等问题。因此,将非局部TV算子应用于传统的局部TV模型,提出一种新模型来去除瑞利分布的乘性噪声,并通过引入辅助变量构造一个逼近的泛函,设计相应的分裂Bregman算法,加快了计算速度。实验结果表明,所提出的非局部TV模型在峰值信噪比、结构相似度和视觉效果方面都优于传统算法,对图像纹理、边缘等特征的保持能力更强。The traditional total variation （TV） models based on local operators for multiplicative noise with Rayleigh distribution removal have some problems, such as smeared edges and smeared textures. The nonloeal TV norm is employed to classical TV model. On this basis, a new nonlocal TV model for multiplicative noise with Rayleigh distribution removal is proposed. By introducing an auxiliary variable, we construct an approximation functional, and designed a fast split Bregrnan algorithm with high computational efficiency. Experimental resuhs show that the proposed model has a superior performance than traditional ones in terms of Peak Signal-to-Noise Ratio（ PSNR）, Structural SIMilarity（SSIM） and subjective visual quality, and the capability of preserving features such as textures and edges is enhanced.

关 键 词：瑞利分布 乘性噪声去噪 非局部TV模型 分裂Bregman算法 特征保持 

分 类 号：TN911.73[电子电信—通信与信息系统] TH166[电子电信—信息与通信工程]