微型机与应用Issue(20)：44-47,4.
基于NSCT的独立分量分析方法
Independent component analysis based on NSCT
张军龙 1刘立程 1俞坤游1
作者信息
- 1. 广东工业大学 信息工程学院，广东 广州 510006
- 折叠
摘要
Abstract
In order to improve the separation performance of the independent component analysis ( ICA ) algorithm , a new method based on nonsubsampled contourlet transform ( NSCT ) is proposed in this paper . Firstly , the mixed images are decomposed using NSCT , which increases the non-Gaussianity in the NSCT domain , then the separated matrix can be obtained in the transform domain directly . Theory analysis and simulation experiments all show that the new method can achieve superior performance in both convergence rate and separation accuracy .关键词
独立分量分析/非下采样轮廓波变换/固定点算法/峭度/性能指数Key words
ICA/NSCT/FastICA/kurtosis/performance index分类
信息技术与安全科学引用本文复制引用
张军龙,刘立程,俞坤游..基于NSCT的独立分量分析方法[J].微型机与应用,2013,(20):44-47,4.
