English

应用科学学报 ›› 2013, Vol. 31 ›› Issue (5): 488-494.doi: 10.3969/j.issn.0255-8297.2013.05.008

• 信号与信息处理 • 上一篇    下一篇

基于样本扩展和线性子空间特征提取的单样本人脸识别

孟一飞1， 袁雪1， 魏学业1， 秦飞舟2， 覃庆努1   

  1. 1. 北京交通大学电子信息工程学院，北京100044
    2. 宁夏大学物理电气信息学院，银川750021
  • 收稿日期:2012-05-23 修回日期:2012-10-15 出版日期:2013-09-26 发布日期:2012-10-15
  • 作者简介:孟一飞，博士生，副教授，研究方向：数字图像处理、模式识别，E-mail: 10111045@bjtu.edu.cn；魏学业，教授，博导，研究方向：自动控制理论，E-mail: xywei@bjtu.edu.cn
  • 基金资助:

    国家自然科学基金(No. 61301186)；高等学校博士学科点专项科研基金（No. 20110009120003）；北京交通大学校基金（No.W11JB00460）资助

Single Sample Face Recognition with Virtual Samples and Linear Subspace Feature Extraction

MENG Yi-fei1, YUAN Xue1, WEI Xue-ye1, QIN Fei-zhou2, QIN Qing-nu1   

  1. 1. School of Electronic and Information Engineering , Beijing Jiaotong University,
    Beijing 100044, China
    2. School of Physics Electrical Information Engineering, Ningxia University,
    Yinchuan 750021, China
  • Received:2012-05-23 Revised:2012-10-15 Online:2013-09-26 Published:2012-10-15

PDF

1818

被引次数

8

摘要： 针对单训练样本情况下的人脸识别问题，提出一种虚拟样本扩展方法. 利用光照模板映射将单一样本扩展为一组虚拟样本，从而增强单训练样本的分类信息. 采用主成分分析（principal component analysis, PCA）对扩展的虚拟样本进行降维，并用Fisher 鉴别变换作二次特征抽取，然后用最近邻分类器识别人脸图像. 所提方法在人脸图像库Yale B 和Extended Yale B 上进行试验，用PCA+LDA 方法把扩展图像作为训练集对测试图像进行特征提取和识别. 相对于以单样本图像为训练集的PCA 特征提取，该方法显著提高了识别率.

关键词: 人脸识别, 单人单样本, 主成分分析, 线性鉴别分析

Abstract:  A method of reference model illumination mapping is proposed to deal with the problem of one sample per person. To enhance the classification information of single training sample, we extend virtual images generated from the given single training image. Using discrete wavelet transform (DWT), the low-frequency band is processed to map the illumination information from the reference model to create a virtual sample. Principle-component-analysis plus linear-discriminant-analysis (PCA+LDA) is performed on the virtual training set to extract features. Experiments are performed on the Yale B and extended Yale B facial image database. The results show that, compared with the PCA feature extraction with single sample, recognition rate is significantly higher using the proposed method.

Key words:  face recognition, single sample per person, principal component analysis (PCA), linear discriminant analysis ( LDA )

中图分类号: