基于2D-PCA和2D-LDA的人脸识别方法  

Face recognition method based on 2D-PCA and 2D-LDA

作　　者：温福喜[1] 刘宏伟[1] 

机构地区：[1]西安电子科技大学雷达信号处理国家重点实验室,西安710071

出　　处：《计算机应用研究》2007年第8期201-203,共3页Application Research of Computers

基　　金：国家自然科学基金资助项目(60302009);陕西省自然科学基金资助项目(2005F35)

摘　　要：提出了基于2D-PCA、2D-LDA两种特征采用融合分类器的人脸识别方法。首先提取人脸图像的2D-PCA和2D-LDA特征,对不同特征在决策层对分类器进行融合。在ORL人脸库上的试验结果表明,分类器决策层融合方法在识别性能上优于2D-PCA和2D-LDA,更具有鲁棒性。A face recognition technique based on 2D-PCA and 2D-LDA using combining classifier was presented. First the original face images＇ 2D-PCA and 2D-LDA features were extracted, then, the decision level combination of the classifier was applied for different features. A series of experiments were performed on face image databases： ORL human face databases. The experimental result indicates that the recognition performance of classifier combination in decision level is superior to that of 2D-PCA and 2D-LDA, and is more robust.

关 键 词：人脸识别 二维主分量分析 二维线性可分性分析 分类器融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]