基于KPCA和Gabor小波的特征融合人脸识别

Feature Fusion Method for Face Recognition Based on KPCA and Gabor Wavelets

针对大多数人脸识别算法中特征提取只关注一种特征的问题,本文提出了一种基于KPCA和Gabor小波特征融合的人脸识别算法。它是一种用核主成分分析方法 (KPCA)提取人脸高阶非线性全局特征,用Gabor小波提取人脸局部特征,再分别用费谢尔线性判别(FLD)提取特征再加权级联融合的方法。实验表明,该算法不仅计算速度快,识别率高,而且能有效解决小样本空间问题。

Most face recognition algorithms for feature extraction only concentrate on one characteristic.In this paper,we introduce a feature fusion method for face recogniton based on KPCA and Gabor wavelets. It is a method to use the kernel principal component analysis（KPCA） for extracting the higher-order and nonlinear global features of face, use Gabor wavelet to extract local features of face, then use the Fisher Linear Discriminant Analysis （FLD） to extract their fea- tures respectively and combine the two types of features. Experiments showed that the speed of the algorithm is fast, and the accuracy rate is high. Moreover, it can effectively solve the problem of small sample space.