计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (22): 197-200.DOI: 10.3778/j.issn.1002-8331.2008.22.059

• 图形、图像、模式识别 • 上一篇    下一篇

基于Gabor变换和变精度粗糙集的人脸识别

刘海霞1,周 军1,梅红岩1,刘立军2   

  1. 1.辽宁工业大学 计算机科学与工程学院,辽宁 锦州 121001
    2.辽宁移动通信有限公司 锦州分公司,辽宁 锦州 121000
  • 收稿日期:2007-10-11 修回日期:2008-01-21 出版日期:2008-07-11 发布日期:2008-07-11
  • 通讯作者: 刘海霞

Face recognition based on Gabor transform and variable precision rough set

LIU Hai-xia1,ZHOU Jun1,MEI Hong-yan1,LIU Li-jun2   

  1. 1.School of Computer Science and Engineering,Liaoning University of Technology,Jinzhou,Liaoning 121001,China
    2.Jinzhou Branch,Liaoning Mobile Communication Ltd.,Jinzhou,Liaoning 121000,China
  • Received:2007-10-11 Revised:2008-01-21 Online:2008-07-11 Published:2008-07-11
  • Contact: LIU Hai-xia

摘要: 针对主成分分析(PCA)方法在特征提取和降维方面的不足,提出一种基于变精度粗糙集的人脸Gabor特征筛选方法。首先提取人脸图像Gabor特征向量,经2DPCA方法处理后用粗糙集对其进行最佳特征选择。然后训练最近邻分类器进行分类识别。在ORL人脸库上验证了该方法的有效性。实验结果显示,该方法的分类能力优于PCA和Gabor_PCA等方法。

关键词: Gabor变换, 主成分分析, 变精度粗糙集, 人脸识别

Abstract: In order to overcome the shortcoming of PCA,a method for extracting Gabor features of face images based on variable precision rough set is presented.First,Gabor features are extracted from face images.After reduced by 2DPCA algorithm,the features are reduced further by rough set.Then the nearest classifier is trained for classification.The experiments being performed on ORL human face image database show the method presented in this paper is superior to that of PCA algorithm and Gabor_PCA algorithm and so on.

Key words: Gabor transform, Principal Component Analysis(PCA), variable precision rough set, face recognition