基于小波变换和核主成分分析的人脸识别算法

Based on the wavelet transform and the kernel principal component analysis of face recognition algorithm

人脸识别是当前模式识别和图像处理领域的研究热点,属于生物鉴别技术的一部分。一个完整的人脸识别系统主要由以下几个基本环节构成:图像预处理、人脸检测与定位、特征提取分类识别。本文主要针对图像的特征提取分类识别环节进行分析和试验:首先应用哈尔小波变换初步提取人脸图像的特征;再对小波系数运用核主成分分析进行最终的人脸特征提取。

Face recognition is the current pattern recognition and the research focus in the field of image processing, belong to identify part of biological technology. A full face recognition system is mainly composed of the following basic link form： the image preprocessing, face detection and localization, the feature extraction classification and recognition. This article mainly aims at the image feature extraction classification link analysis and test： first application Hal wavelet trans- form preliminary extraction of face image characteristics; Again to wavelet coefficients use kernel principal component analysis of the final face feature extraction.