多小波构造及在图像处理中的实现  

Multi-wavelet Construction and implementation in image processing

作　　者：谭小容[1] 李鹏[1] Tan Xiaorong;Li Peng(School of Electronic Engineering Jiujiang University,Jiujiang Jiangxi 332005,China;School of Electronic Engineering Jiujiang University,Jiujiang Jiangxi 332005,China)

机构地区：[1]九江学院电子工程学院,江西九江332005

出　　处：《电子技术（上海）》2018年第10期81-83,75,共4页Electronic Technology

基　　金：江西省卫生计生委科技计划项目(20183016);九江学院科技项目(2016KJ11);九江学院科技项目(2014SKYB009)

摘　　要：为了构造简单有效的多小波并使其能在数字图像处理中得以实现,提出由初等旋转矩阵构造与已知紧支撑对称正交多小波具有相似性质的多小波方法。该方法分析了多小波的矩阵滤波器组与多尺度函数、多小波函数以及多小波变换的关系,推导出多小波的矩阵滤波器组参数化表达式,计算出不同旋转角度下的滤波器值并在数字图像中实现多小波变换。实验结果表明,这种多小波构造方法算法简单,还原图像的信噪比相比于已知多小波的还原图像增加了近二十个分贝值。In order to construct a simple and effective multi wavelet and make it possible to realize the digital image processing, a multi wavelet method is proposed, which is constructed from the elementary rotation matrix and the known compactly supported symmetric orthogonal multi wavelets. In this method, the relationship between multi wavelet matrix filter banks and multi wavelet transform is analyzed, and the matrix filter group expression of multi wavelet is derived, and the multi wavelet transform is realized in digital image. The experimental results show that the algorithm is simple and the PSNR of the reduced image increases by nearly twenty dB compared to the restored image of the known multi wavelet.

关 键 词：初等旋转矩阵 构造 多小波变换 矩阵滤波器组 数字图像 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]