基于小波分析和复矩阵稀疏表示的人脸识别方法  

Face recognition method based on wavelet analysis and sparse representation of complex matrix

作　　者：曹玉涛 吴爱弟[1] CAO Yutao;WU Aidi(School of Sciences,Tianjin University of Technology and Education,Tianjin 300222,China)

机构地区：[1]天津职业技术师范大学理学院,天津300222

出　　处：《天津职业技术师范大学学报》2021年第1期62-66,共5页Journal of Tianjin University of Technology and Education

摘　　要：为提高人脸识别效果,提出一种高频子带和低频子带融合为复矩阵的人脸识别方法。该方法通过对人脸图像进行2层小波分解,提取出第2层小波分解得到的低频部分和高频部分,将3个方向的高频子带分别做分块处理,计算每个子块能量值,选取每个方向信息量多的若干子块,并将其整合为1个新的高频子带,再将新的高频子带和原低频子带构成1个复矩阵。最后,用稀疏表示进行分类,分别在ORL和AR人脸图像数据库中进行测试。测试结果表明,本文设计的方法优于其他人脸识别方法。In order to improve the face recognition effect,a face recognition method is proposed with complex matrix of high frequency and low frequency subbands.Firstly,the low frequency and high frequency parts obtained by the second layer wavelet decomposition were extracted,and the high frequency subbands in three directions were divided into blocks to calculate the energy value of each subblock.Several subblocks with more information in each direction were selected and integrated into a new high frequency subband.The new high frequency subband and the original low frequency subband formed a complex matrix.Finally,a classification was made by sparse representation to test in ORL and AR face image databases,respectively.The test results show that the method used in this paper is superior to other face recognition methods.

关 键 词：字典学习 稀疏表示 人脸识别 小波分析 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]