首页 | 本学科首页   官方微博 | 高级检索  
     

基于三次B样条小波和2DFFT-2DLDA的人脸识别
引用本文:陈勇,冯浩,王宪保. 基于三次B样条小波和2DFFT-2DLDA的人脸识别[J]. 计算机工程与应用, 2010, 46(6): 151-153. DOI: 10.3778/j.issn.1002-8331.2010.06.043
作者姓名:陈勇  冯浩  王宪保
作者单位:1.浙江工业大学 信息工程学院,杭州 310032 2.嘉兴学院,浙江 嘉兴 314001
基金项目:国家自然科学基金(No.60871093)~~
摘    要:提出了一种基于三次B样条小波和2DFFT-2DLDA的人脸识别方法,用三次B样条小波对人脸图像进行多层分解,得到一幅低频子图和3幅边缘细节子图,选取其中两幅效果最好的子图进行二维傅里叶变换后将其连接形成一个特征向量,然后进行2DLDA处理产生最终的特征表达,最后使用最近邻法进行分类。在JAFFE和Yale人脸库中的实验表明算法具有比频谱脸算法和Gabor-2DLDA算法更高的识别率,同时具有很低的算法复杂度。

关 键 词:人脸识别  小波变换  二维线性鉴别分析  
收稿时间:2009-02-16
修稿时间:2009-4-2 

Face recognition based on cubic B-spline wavelet and 2DFFT-2DLDA
CHEN Yong,FENG Hao,WANG Xian-bao. Face recognition based on cubic B-spline wavelet and 2DFFT-2DLDA[J]. Computer Engineering and Applications, 2010, 46(6): 151-153. DOI: 10.3778/j.issn.1002-8331.2010.06.043
Authors:CHEN Yong  FENG Hao  WANG Xian-bao
Affiliation:1.College of Information Engineering,Zhejiang University of Technology,Hangzhou 310032,China 2.Jiaxing University,Jiaxing,Zhejiang 314001,China
Abstract:Face recognition based on cubic B-spline wavelet and 2DFFT-2DLDA is proposed.Cubic B-spline wavelet is used to decompose the facial image at suitable levels to produce an approximation subband and three detail subbands at last level composition.Then two optimal subbands selected from four subbands are performed 2D Fourier transform and concatenated into a feature vector,and then the vector is processed by 2DLDA to produce the last feature vector,the last result will be produced by the nearest neighborhood c...
Keywords:face recognition  wavelet transform  Two Demensional Linear Discriminant Analysis(2DLDA)
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《计算机工程与应用》浏览原始摘要信息
点击此处可从《计算机工程与应用》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号