基于支持向量机的人脸识别

由于支持向量机(SVM)有着适合处理小样本问题、高维数及泛化性能强等多方面的优势，并且基于核函数主元分析的方法对于非线性问题的特征提取来说较为合理，因此提出了一种基于核函数主元分析(KPCA)与支持向量机(SVM)的人脸识别方法。在使用KPCA方法对人脸图像进行特征提取后，用SVM对特征向量进行分类识别。利用剑桥ORL的人脸数据库进行的实验结果验证了本算法的有效性。

Research of Face Recognition Method Based on SVM

According to the high performance of support vector machine (SVM) in tackling small samplesize, high-dimension and its good generalization, and the more reasonable of kernel-based principal component analysis method in extracting features from nonlinearly input variables as well, this paper proposes a face recognition method based on kernel-based principal component analysis (PCA) and SVM .After extracting the feature from face image by using the KPCA method,it uses the SVM to classify face. The ORL face database is used to test the proposed method. The preparatory experiment result shows that the method is feasible.