摘要：

提出一种基于图嵌入正则化的人脸线性判别分析方法。构造非监督最优类可分准则，基于图嵌入理论，求解该最优类可分准则下的最优投影向量，在非监督的图嵌入框架下利用样本局部类别信息提高人脸识别率，降低矩阵计算复杂度。在典型的人脸数据库上的实验证明了该方法的有效性。

关键词: 人脸识别, 图嵌入, 正则化, 线性判别分析

Abstract:

This paper proposes a new Linear Discriminant Analysis(LDA) for face recognition based on the graph embedding and regularization. The unsupervised optimal class separate criterion is built, and it proposes a method to get this projection vector on the graph embedding frame. It can get a huge save of computational consumption and utilize the class information of samples more effectively. Experimental results demonstrate the effectiveness and efficiency of this method.

Key words: face recognition, graph embedding, regularization, Linear Discriminant Analysis(LDA)

中图分类号: 

• TP274+.3