| 基于分块Contourlet变换的图像盲分离 |
| |
| 引用本文： | 彭伟伟,蒋兴浩.基于分块Contourlet变换的图像盲分离[J].信息安全与通信保密,2008(6):119-122. |
| |
| 作者姓名： | 彭伟伟 蒋兴浩 |
| |
| 作者单位： | 上海交通大学信息安全工程学院,上海,200030 |
| |
| 摘 要： | 文章研究了多尺度Contourlet变换并将它与独立分量分析ICA结合起来,在传统的小波ICA基础上提出了FRITICA方法。利用Contourlet变换比小波变换具有更好的稀疏性以及图像进行分块处理后某些图像块的星座图比其他图像块的星座图方向性更好的特点,提出一种基于分块Contourlet变换的独立分量分析方法,进而提高叠加混合图像的盲分离的效果与速度。
|
| 关 键 词： | 图像分离 独立分量分析 Contourlet变换 |
| 文章编号： | 1009-8054(2008)06-0119-04 |
| 修稿时间： | 2008年5月5日 |
Block-wise Contourlet Transform-based Blind Image Separation Algorithm |
| |
| PENG Wei-wei,JIANG Xing-hao.Block-wise Contourlet Transform-based Blind Image Separation Algorithm[J].China Information Security,2008(6):119-122. |
| |
| Authors: | PENG Wei-wei JIANG Xing-hao |
| |
| Affiliation: | (School of Information Security Engineering, Shanghai Jiao Tong University, Shanghai 200030, China) |
| |
| Abstract: | Contourlet transform has been studied, and modified ICA algorithms are described. Based on the characteristics of good sparsity and capturing effectively the smooth contours in natural images for Contourlet transform, a novel independent component analysis algorithm is proposed by using the block-wise Contourlet transform in this thesis. Experimental resuits show that the proposed algorithm is able to achieve a better performance and spends less time in image separation. |
| |
| Keywords: | image separation independent component Analysis（ICA） contourlet transform |
| 本文献已被 维普 万方数据 等数据库收录！ |
