基于小波子空间集成的人脸识别  

Face recognition based on ensemble of wavelet subspaces

作　　者：翟俊海[1] 翟梦尧[1] 张素芳 王熙照[1] 

机构地区：[1]河北省机器学习与计算智能重点实验室(河北大学数学与计算机学院),河北保定071002 [2]河北省信息工程学校数学教研室,河北保定071000

出　　处：《山东大学学报（工学版）》2012年第2期1-6,29,共7页Journal of Shandong University（Engineering Science）

基　　金：国家自然科学基金资助项目(61170040);河北省自然科学基金资助项目(F2010000323);河北省高等学校科学技术研究重点项目(ZD2010139);河北大学大学生科技创新项目(2011043)

摘　　要：基于小波变换的人脸识别方法通常选用低频子图进行人脸识别,这样会丢失其他子段图像中的识别信息。针对这一问题,提出了两种小波子空间集成人脸识别方法并与其他相关方法进行了实验比较。第1种方法集成每1层小波低频子空间图像进行人脸识别;第2种方法首先对人脸图像做L层小波分解,然后对每1层的3个高频子空间图像求平均,连同每层的1个低频子空间图像得到L个小波子空间图像,最后集成这L个小波子空间图像进行人脸识别。本文提出的方法充分利用了不同频率小波子段图像的识别信息,能够提高人脸识别的精度。在ORL、YALE和JAFFE 3个人脸数据库上的实验结果显示,本文提出的方法特别是方法 2在识别精度方面都优于其他方法。The low frequency subimage is usually used for face recognition based on wavelet transform （WT） methods. However, some important information hidden in other high frequency subimages will be unavoidably lost. To solve this problem, two methods were presented for face recognition by ensemble of wavelet subspaces, and the comparisons with other related methods were put forth by experiments. In the first method, the wavelet low frequency subimages at each layer were integrated for face recognition. In the second method, face images were first decomposed into different sub- images with L layer wavelet transform, and then L wavelet subspace images were obtained by averaging three high fre- quency subimages of each layer and integrating the low frequency subimage of each layer. Finally the L wavelet sub- space images were integrated for face recognition. The proposed methods could make full use of the information provid- ed by the different frequency subimages and the accuracy of face recognition was improved. The experimental results of three face databases （ORL, YALE, and JAFFE） showed that the proposed methods, especially the second method, could obtain a higher accuracy than other related methods.

关 键 词：人脸识别 小波变换 子空间集成 二维主成分分析 二维线性判别分析 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]