作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (04): 212-214. doi: 10.3969/j.issn.1000-3428.2007.04.074

• 人工智能及识别技术 • 上一篇    下一篇

基于小波域奇异值分解的人脸识别方法

闫荣华1,2,彭进业1,李 岩1,谢明华1,温文龙2   

  1. (1. 西北大学信息科学与技术学院,西安 710127;2. 中国科学院西安光学精密机械研究所,西安 710119)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-02-20 发布日期:2007-02-20

Method of Face Recognition Based on Singular Value Decomposition in the Wavelet Domain

YAN Ronghua1,2, PENG Jinye1, LI Yan1, XIE Minghua1, WEN Wenlong2   

  1. (1. School of Information Science & Technology, Northwest University, Xi’an 710127; 2. Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi’an 710119)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-02-20 Published:2007-02-20

摘要: 通过分析小波分解系数对光照、姿态、表情所表现出来的特性,选择了不同系数的组合进行奇异值分解,提高了在光照、姿态、表情变化等情况下人脸识别的鲁棒性。采用Harvard、Umist和Yale 3个人脸库对该文提出的方法进行了人脸识别实验。结果表明,基于小波分解系数优化组合的奇异值分解方法的识别率高于在原图上的奇异值分解方法。

关键词: 人脸识别, 小波变换, 奇异值分解, 反对称双正交小波

Abstract: This paper studies the properties of different wavelet decomposition coefficients when illumination, pose and expression are changed. Robustness of face recognition is improved by decomposing different coefficient. The standard face database from Harvard, UMIST and Yale are selected to evaluate the recognition accuracy of the method. The experiment indicates that the error rate of the method based on the wavelet transform and SVD is lower than that of the method of singular value decomposition in gray domain.

Key words: Face recognition, Wavelet transform, Singular value decomposition(SVD), Anti-symmetrical biorthogonal wavelet