Abstract: A method of face recognition based on the whole and local singular value decomposition is presented.It discusses the pre-process of face image，the theory and realization process of singular value decomposition and support vector machine.Gets the whole and local features of the image by utilizing the whole and local singular value decomposition.Then utilizing support vector machine to classify.The result of experiment shows that the method is effective.

Key words: face recognition, feature extraction, singular value decomposition, support vector machine

摘要： 提出了一种基于整体与局部奇异值分解相结合的人脸识别算法。文章叙述了人脸图像的预处理、奇异值分解以及支持向量机的原理及实现过程。运用整体与局部奇异值分解，分别获得图像的整体与局部特征，然后采用支持向量机进行分类识别，实验验证了该方法的有效性。

关键词: 人脸识别, 特征提取, 奇异值分解, 支持向量机