| 基于PCA与ICA的人脸识别算法研究 |
| |
| 引用本文： | 王展青,刘小双,张桂林,王仲君.基于PCA与ICA的人脸识别算法研究[J].华中师范大学学报(自然科学版),2007,41(3):373-376. |
| |
| 作者姓名： | 王展青 刘小双 张桂林 王仲君 |
| |
| 作者单位： | 华中科技大学,图像识别与人工智能研究所,武汉,430074;武汉理工大学,理学院,数学系,武汉,430070;武汉理工大学,理学院,数学系,武汉,430070;华中科技大学,图像识别与人工智能研究所,武汉,430074 |
| |
| 摘 要： | ICA是一种基于数据高阶统计信息的有效的数据独立特征提取技术,它能够更好地表示人脸的局部特征,ICA是PCA从二阶统计分析向高阶统计分析的拓展.本文提出了一种加权融合这两种技术的人脸特征提取算法,并结合不同的相似性度量进行了人脸识别实验.结果表明,该方法比用一种单独的特征提取方式识别率要高.
|
| 关 键 词： | 人脸识别 特征抽取 主分量分析 独立成分分析 |
| 文章编号： | 1000-1190（2007）03-0373-04 |
| 修稿时间： | 2007-03-21 |
Face recognition based on PCA and ICA |
| |
| WANG Zhanqing,LIU Xiaoshuang,ZHANG Guilin,WANG Zhongjun.Face recognition based on PCA and ICA[J].Journal of Central China Normal University(Natural Sciences),2007,41(3):373-376. |
| |
| Authors: | WANG Zhanqing LIU Xiaoshuang ZHANG Guilin WANG Zhongjun |
| |
| Institution: | 1. Institute for Pattern Recognition and Artificial Intelligence, Huazhong University of Science and Technology, Wuhan 430074; 2. Department of Mathematics, College of Science, Wuhan University of Technology, Wuhan 430070 |
| |
| Abstract: | ICA, which is a sufficient method of extracting independent features based on the higher statistic information of data,can express the local facial feature better. ICA is the extension of PCA from second-order statistic analysis to higher statistic analysis. A facial feature extraction algorithm combined these two methods by being weighted is presented in this article, and face recognition experiments are performed with different similar measurement. The result indicates that the proposed method is better than those that use one method only. |
| |
| Keywords: | face recognition feature extraction PCA ICA |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《华中师范大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《华中师范大学学报(自然科学版)》下载免费的PDF全文 |
|
