| 基于快速曲波变换和独立分量分析的有噪图像盲分离算法 |
| |
| 引用本文： | 毕杨,肖军.基于快速曲波变换和独立分量分析的有噪图像盲分离算法[J].自动化技术与应用,2010,29(1):53-56. |
| |
| 作者姓名： | 毕杨 肖军 |
| |
| 作者单位： | 1. 西安航空技术高等专科学校电气系,陕西,西安,710077;西北工业大学航海学院,陕西,西安,710072
2. 西安航空技术高等专科学校电气系,陕西,西安,710077 |
| |
| 摘 要： | 曲波（Curvelet）变换有两种数值实现方法，一种是基于非均匀采样的快速Fourier变换即usfft算法；另一种是基于特殊选择的Fourier采样的卷绕即wrapping算法。将这两种实现方法针对图像去噪性能进行了比较，并且提出将曲波变换的wrapping算法和FastICA算法结合起来，对含有噪声的混合图像进行了盲分离。仿真研究结果表明，对于含有加性观测噪声的混合图像，该方法有更好的去噪分离效果，并且运算速度显著提高。
|
| 关 键 词： | 快速离散曲波变换 usfft算法 wrapping算法 FastlCA算法 图像盲分离 |
Noisy Image Blind Separation Algorithm Based on The Fast Curvelet Transform and Independent Component Analysis |
| |
| BI Yang,XIAO Jun.Noisy Image Blind Separation Algorithm Based on The Fast Curvelet Transform and Independent Component Analysis[J].Techniques of Automation and Applications,2010,29(1):53-56. |
| |
| Authors: | BI Yang XIAO Jun |
| |
| Affiliation: | 1. Department of Electrical Engineering, Xi＇an aerotechnical college, Xi＇an 710077 China; 2. Institute of Marine, Northwestern polytechnical university, Xi＇an 710072 China ) |
| |
| Abstract: | The There are two digital implementations: for curvelet transform the first digital transformation is based on unequally-spaced fast Fourier transform while the second is based on the wrapping of specially selected Fourier samples.The realization of these two methods for image denoising performance is compared,and a new method of noisy image blind separation is proposed using Fast Discrete Curvelet Transforms via wrapping algorithm and the algorithm of Fast ICA.The simulation results show that this method h... |
| |
| Keywords: | fast discrete curvelet transform(FDCT) usfft algorithm wrapping algorithm the algorithm of fast ICA image blind separation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
