| 基于Curvelet稀疏表示的图像盲分离初始化 |
| |
| 引用本文： | 王军华,方勇.基于Curvelet稀疏表示的图像盲分离初始化[J].应用科学学报,2009,27(2):161-166. |
| |
| 作者姓名： | 王军华 方勇 |
| |
| 作者单位： | 1. 上海大学通信与信息工程学院，上海200072
2. 上海大学特种光纤与光接入网重点实验室，上海200072 |
| |
| 基金项目： | 高等学校博士学科点专项科研基金，上海市重点学科建设项目 |
| |
| 摘 要： | 针对盲分离初始化问题提出一种基于Curvelet稀疏表示的图像盲分离初始化方法. 该方法充分利用信号Curvelet变换的稀疏特性，选取稀疏性最好的高频系数组，采用聚类方法估计聚轴中心，寻求混合矩阵估计值，实现对盲分离学习算法的初始化. 实验结果表明，该初始化方法能避免盲分离算法在收敛时陷入局部最小，加快收敛，并提高分离精度.
|
| 关 键 词： | 盲源分离 稀疏表示 Curvelet变换 初始化 |
| 收稿时间： | 2008-07-22 |
| 修稿时间： | 2008-10-18 |
Initialization Algorithm for Blind Image Separation Based on Curvelet Sparse Representation |
| |
| WANG Jun-hua,FANG Yong.Initialization Algorithm for Blind Image Separation Based on Curvelet Sparse Representation[J].Journal of Applied Sciences,2009,27(2):161-166. |
| |
| Authors: | WANG Jun-hua FANG Yong |
| |
| Institution: | 1. School of Communication and Information Engineering, Shanghai University, Shanghai 200072 China;
2. Key Laboratory of Special Optical Fibers and Optical Access Network, Shanghai University, Shanghai 200072, China |
| |
| Abstract: | A new initialization algorithm for blind separation of images is proposed based on curvelet sparse representation. A mixed matrix can be estimated by estimating the center of received signals. This method can
improve convergence and effectively avoid falling into local minima. Simulation results show that the proposed
algorithm can achieve better performance for blind source separation of images. |
| |
| Keywords: | blind source separation sparse representation Curvelet transform initialization |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《应用科学学报》浏览原始摘要信息 |
| 点击此处可从《应用科学学报》下载免费的PDF全文 |
