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.
出处
《计算机应用研究》
CSCD
北大核心
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)
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
face recognition
two-dimensional principal component analysis
two-dimensional linear discriminate analysis
classifier combination
