基于核主元分析和支持向量机的人脸识别

核主元分析(KPCA，Kernel Principal Components Analysis)具有能较好地提取非线性特征的优势；支持向量机(SVM，Support Vector Machine)具有较好的非线性映射能力，且泛化能力强。结合核主元分析与支持向量机的特点，提出了一种基于核主元分析与支持向量机的人脸识别方法。该方法首先利用核主元分析对人脸图像进行特征提取，然后依据支持向量机与最近邻准则对所提取的核主元特征进行分类识别。基于ORL(Olivetti Research Laboratory)人脸数据库的实验结果表明了该方法的有效性。

Face recognition method based on KPCA and SVM

Kernel principal components analysis (KPCA) has the advantage of extracting nonlinear features. Nonlinear mapping and generalization are the strong capabilities of support vector machine (SVM). By integrating the characteristics of KPCA and SVM, a face recognition method based on these two algorithms is presented. Firstly KPCA is used to extract the features of human face image, and then SVM combined with the nearest distance rule is used for classification, which depends on the kernel principal components extracted. Experimental results on ORL (Olivetti Research Laboratory) face database verifies the efficiency of the method.