基于尺度噪声水平估计的双树复数小波变换图像去噪方法  

An image de-noising method of dual tree complex wavelet tramsform based on scale noise level estimation

作　　者：许慰玲[1] 方若宇[2] 闫敬文[2] 

机构地区：[1]仲恺农业工程学院信息学院,广东广州510225 [2]汕头大学广东省数字信号与图象处理技术重点实验室,广东汕头515063

出　　处：《仲恺农业工程学院学报》2011年第1期32-35,共4页Journal of Zhongkai University of Agriculture and Engineering

基　　金：国家自然科学基金(40971206)资助项目

摘　　要：利用噪声与图像双树复数小波变换各层系数的关系,提出基于尺度噪声水平估计的双树复数小波变换图像去噪方法并进行了试验.结果表明,该估计方法与图像噪声水平近似线性相关,能很好地反映各层系数的噪声水平.双变量收缩函数阈值去噪法采用该估计比采用全局噪声水平估计去噪效果更好,平均结构相似度明显提高.Utilizing the correlation between the noise and wavelet coefficients of dual tree complex wavelet transform（DTCWT）,a novel DTCWT based scale noise level estimation（DTCWT_SNLE） was proposed.The experiment showed that there existed an approximate linear relationship between DTCWT_SNLE and the noise level.The noise level of coefficients in each layer could be reflected more accurately by DTCWT_SNLE.Comparing with DTCWT based global noise level estimation（DTCWT_GNLE）,the method had better performance in Bivariate shrinkage algorithm and the mean structural similarity increased markedly.

关 键 词：双树复数小波变换 尺度噪声水平估计 双变量收缩函数 阈值去噪 

分 类 号：TN911.73[电子电信—通信与信息系统]