基于多小波变换的图像去噪新方法  

A New Image De-noising Method Based on Multiwavelet Transform

作　　者：孙伟峰[1,2] 何俊华[1] 

机构地区：[1]中国科学院西安光学精密机械研究所,西安710119 [2]中国科学院研究生院,北京100049

出　　处：《光子学报》2008年第11期2360-2364,共5页Acta Photonica Sinica

基　　金：国防科技重点实验室基金(5144830105ZK1801)资助

摘　　要：离散多小波在图像去噪和图像融合方面有突出的效果.利用GHM多小波,对图像去噪中阈值的选取给出了具体的方法,对单小波与多小波去噪方法进行了比较.结果表明,多小波在图像去噪的效果上明显优于单小波.Comparing with wavelet, discrete multiwavelet has a lot of advantages in image processing, especially in image de-noising and fusion. The method of choosing threshold while using multiwavelet in image de-noising was presented used GHM multiwavelet. A comparision was made between wavelet and multiwavelet in image de-noising. The results show that multiwavelet performs better than wavelet.

关 键 词：多小波 分解 重构 去噪 闽值 信噪比 

分 类 号：TN911[电子电信—通信与信息系统]