基于多小波变换的图像去噪.pdfVIP

第 24 卷第 12期 计算机应用与软件 V o l24 No. 12 2007 年 12 月 Com puter A pplications and Sofwt are Dec. 2007 1, 2 1 2 董卫军 樊养余 周明全 1 7 10072 ) 2 7 10069 ) 小波分析是目前国际上最新的时间频率分析工具, 是信号去 的强有力处理工具小波变换可以将交织在 一起的混合 信号分解成不同频率的块信号多小波所拥有的对称性正交性有限支撑等重要特性弥补了单小波的不足提出了 一种新的图像 去 方法, 该方法以多小波变换为基础实验证明, 去 效果良好 多小波 图像去 小波变换 IMAGE DENOSING METHOD BASED ON MULTIWAVELET TRANSFORM 1, 2 1 2 Dong W e ijun F an Y angyu Zhou M ingquan 1 (S chool of E lectron ics and Inf orm ation; N orthw estern P olytechn ica l Un iv ersity; X ian 710072, Shaan ,i Ch ina) 2 (Colleg e of Inf orm ation Science and Technology, Northw est Un iv ersity, X ian 710069, Shaan i, Ch ina) Abstract W ave le t ana ly sis, the latest tmi efrequency ana lysis too l internationa lly, is a pow erful tool for mi age denos ing. T hrough w ave le t transform, signals are separated into b lock signa lw ith d ifferen t frequency. M ultiw ave let has som e very mi portant properties such as o rthogona l ity, symm etry, and f inite support, w hich m ake up fo r the shortcom ing of scalar w ave let. A new mi age denosing me thod based on m ultiw ave le t transform is proposed. The expe rmi en tal results show that the new m ethod can ach ieve good denos ing effec t. K eywords M u ltiw av elet Im age denosing W ave le t transform , , 0 , , , . , [ 1~ 4] , ; [ 5~ 7] , ,