首页 | 本学科首页   官方微博 | 高级检索  
     

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
引用本文:许慰玲,方若宇,闫敬文. 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[J]. 仲恺农业技术学院学报, 2011, 24(1): 32-35
作者姓名:许慰玲  方若宇  闫敬文
作者单位: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
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.

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

An image de-noising method of dual tree complex wavelet tramsform based on scale noise level estimation
XU Wei-ling,FANG Ruo-yu,YAN Jing-wen. An image de-noising method of dual tree complex wavelet tramsform based on scale noise level estimation[J]. Journal of Zhongkai Agrotechnical College, 2011, 24(1): 32-35
Authors:XU Wei-ling  FANG Ruo-yu  YAN Jing-wen
Affiliation:XU Wei-ling1,FANG Ruo-yu2,YAN Jing-wen2(1.Information College,Zhongkai University of Agriculture and Engineering,Guangzhou 510225,China,2.Guangdong Provincial Key Laboratory of Digital Signal and Image Processing Technology,Shantou University,Shantou 515063,China)
Abstract: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 Bivaria...
Keywords:dual tree complex wavelet transform  scale noise level estimation  bivariate shrinkage function  threshold de-noising  
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号