计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (17): 207-209.

• 图形、图像、模式识别 • 上一篇    下一篇

子域NMF在人脸识别中的研究

周 劲,王义为,杨全龙,吴 萌   

  1. 济南大学 信息科学与工程学院,济南 250022
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-06-11 发布日期:2011-06-11

Research of face recognition technology based on subspace NMF

ZHOU Jin,WANG Yiwei,YANG Quanlong,WU Meng   

  1. School of Information Science and Engineering,University of Jinan,Jinan 250022,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-06-11 Published:2011-06-11

摘要: 人脸识别是一种通过提取人脸视觉特征信息进行身份鉴别的计算机技术。在非负矩阵分解NMF 算法的基础上提出了改进的基于子域的NMF求解算法,将其应用于人脸识别领域,分别在Yale和ORL公共人脸数据库进行测试,得到了100%和95%的识别率。与其他求解算法相比具有夹逼性好,识别率高等优点。

关键词: 人脸识别, 非负矩阵分解, 子域, 特征提取

Abstract: Face recognition is one kind of computer technology for authentication by extracting visual features of human faces.An improved NMF algorithm in subspace is proposed and applied to the face recognition by using Yale database and ORL database.The experiments have shown that the improved algorithm can generate better recognition rate(respectively 100% and 95%)than other methods.

Key words: face recognition, non-negative matrix factorization, subspace, feature extraction