基于Hermite特征和核函数判决分析的人脸识别  

Face Recognition Based on Hermite Feature and Kernel Discriminant Analysis

作　　者：刘靖[1] 周激流[1] 

机构地区：[1]四川大学电子信息学院,四川成都610064

出　　处：《光电子．激光》2006年第1期119-123,共5页Journal of Optoelectronics·Laser

基　　金：国家自然科学基金资助项目(60272095)

摘　　要：提出了基于Hermite特征和核函数判决分析(KDA)方法的人脸识别技术,即首先利用多分辨Hermite分析提取人脸图像特征,然后用KDA方法进行分类识别。采用拟Gabor的多分辨Hermite分析能够提取人脸图像中具有空间频率特性、空间位置和方向取向选择特性的判决特征,从而较好地克服由于光照强度、脸部表情变化带来的影响。KDA方法能够提取高维特征空间中的最佳判决特征,本质上对应于原始空间中的非线性最佳判决特征,具有较好的聚类判决分析性能。实验仿真表明了所提方法的有效性。The Hermite-feature-based kernel discriminant analysis（KDA） method for face recognition is presented. Firstly, a multiresolution-directional-oriented Gabor-like Hermite transform is used to derives the facial features that are characterized by spatial frequency, spatial locality,and orientation selectivity to do with the variations due to illumination and facial expression changes. Then the KDA method is used to enhance the face recognition performance. The KDA can extract the features particularly useful in discriminating ability in the high-dimensional space,which are the most discriminating nonlinear features in the input space. It has been demonstrated to be capable of nonlinear discriminant analysis. The feasibility of the method has been successfully tested.

关 键 词：多分辨 HERMITE变换 核函数判决分析(KDA) 人脸识别 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]