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 ›› 2010, Vol. 36 ›› Issue (7): 200-202. doi: 10.3969/j.issn.1000-3428.2010.07.069

• 人工智能及识别技术 • 上一篇    下一篇

基于OLPP和信息向量机的人脸识别

刘建伟,徐 翔,罗雄麟   

  1. (中国石油大学自动化研究所,北京 102249)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2010-04-05 发布日期:2010-04-05

Face Recognition Based on Orthogonal Locality Preserving Projection and Informative Vector Machine

LIU Jian-wei, XU Xiang, LUO Xiong-lin   

  1. (Research Institute of Automation, China University of Petroleum, Beijing 102249)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-04-05 Published:2010-04-05

摘要: 结合正交局部保持投影(OLPP)和信息向量机(IVM),提出用于人脸识别的OLPP-IVM算法。应用OLPP对原始人脸图像数据进行特征提取,利用IVM在降维后的数据上实现人脸分类。与主成分分析、线性判别分析等算法的比较实验证明,用该算法进行人脸识别误差更小,性能更优越。

关键词: 信息向量机, 正交局部保持投影, 人脸识别, 特征提取, 多类分类

Abstract: By combining Orthogonal Locality Preserving Projection(OLPP) and Informative Vector Machine(IVM), this paper proposes OLPP-IVM algorithm for face recognition. It uses OLPP to extract the feature of face image and reduces the dimension of data, and uses IVM to classify the face. By compared with PCA, LDA and other algorithms, the algorithm is proved to have lower error rate in face image recognition, and its performance is superior.

Key words: Informative Vector Machine(IVM), Orthogonal Locality Preserving Projection(OLPP), face recognition, feature extraction, multi-class classification

中图分类号: