首页 | 本学科首页   官方微博 | 高级检索  
     

人脸图像的特征提取
引用本文:赵振勇,王保华,王力,崔磊.人脸图像的特征提取[J].微机发展,2007,17(5):221-224.
作者姓名:赵振勇  王保华  王力  崔磊
作者单位:贵州大学信息工程学院 贵州贵阳550003(赵振勇,王保华,王力),同济大学计算机学院 上海200092(崔磊)
摘    要:人脸识别的研究是模式识别和人工智能领域内的重要课题,有着十分广泛的应用前景。人脸特征的自动提取是人脸自动识别过程中至关重要的一个环节。主要就基于积分投影的人脸图像的特征提取、基于奇异值分解的特征提取及小波特征提取等几种较好方法进行研究。基于积分投影的人脸图像特征点的提取方法对人脸进行定位特别精确。基于小波分解频谱性分析的人脸特征提取极大减少了图像的存储空间和计算复杂度。基于SVD分解的特征提取处理后的正确率很高,计算复杂度也低。

关 键 词:小波分解  积分投影  奇异值分解
文章编号:1673-629X(2007)05-0221-04
修稿时间:2006年8月13日

The Feature Extraction of Face Images
ZHAO Zhen-yong,WANG Bao-hua,WANG Li,CUI Lei.The Feature Extraction of Face Images[J].Microcomputer Development,2007,17(5):221-224.
Authors:ZHAO Zhen-yong  WANG Bao-hua  WANG Li  CUI Lei
Affiliation:ZHAO Zhen-yong1,WANG Bao-hua1,WANG Li1,CUI Lei2
Abstract:Research of human face recognition is an important topic in the area of pattern recognition and artificial intelligence,and it has very broad application prospects.The extraction of face features is an important part in the process of face automatic recognition.It is essential to introduce the extraction of face features based on integral projection,singular value decomposition and wavelet and so on.Bonus point projection based on the characteristics of face images from the point of methodology is precised for people face special position.Storage space and computing complexity are reduced by wavelet decomposition analysis based on the spectrum of identity from the face of great images.A high rate of correct is come out through based on characteristics extraction based on SVD decomposition.
Keywords:wavelet decomposition  integral projection  singular value decomposition
本文献已被 CNKI 等数据库收录!
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