东北大学学报（自然科学版）

东北大学学报:自然科学版 ›› 2015, Vol. 36 ›› Issue (11): 1526-1529.DOI: 10.12068/j.issn.1005-3026.2015.11.002

• 信息与控制 • 上一篇    下一篇

基于MBC和POEM特征的人脸识别方法

张祥德， 朱和贵， 李倩颖， 唐青松   

  1. (东北大学 理学院， 辽宁 沈阳110819)
  • 收稿日期:2014-10-21 修回日期:2014-10-21 出版日期:2015-11-15 发布日期:2015-11-10
  • 通讯作者: 张祥德
  • 作者简介:张祥德(1963-)，男，山东昌乐人，东北大学教授.
  • 基金资助:
    国家自然科学基金资助项目(61202085)； 辽宁省自然科学基金资助项目(201202074)； 中央高校基本科研业务费专项资金资助项目(N140503004).

Face Recognition Algorithm Based on MBC and POEM Feature

ZHANG Xiang-de，ZHU He-gui， LI Qian-ying， TANG Qing-song   

  1. School of Sciences， Northeastern University， Shenyang 110819， China.
  • Received:2014-10-21 Revised:2014-10-21 Online:2015-11-15 Published:2015-11-10
  • Contact: ZHU He-gui
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

662

摘要： 针对人脸识别中单一特征难以取得理想效果的问题，提出了基于MBC和POEM特征融合的人脸识别方法.首先，在归一化的人脸图像上提取MBC编码图和POEM编码图，在每个编码图块上生成特征向量，应用线性判别分析对特征向量进行低维映射，并对其进行赋权相加得到最终相似度. 所提算法在FERET的Dup1，Dup2，Fb和Fc 4个测试库上取得了较高的识别率，分别为93.77%，90.60%，99.58%和99.49%；在误识率为0.1%的条件下，在4个测试库上的认证率分别为95.70%，92.31%，99.75%和100%，进一步验证了该方法的有效性.

关键词: 人脸识别, MBC特征, POEM特征, 特征融合, 赋权分段线性判别分析

Abstract: Due to the representation difficulty of a face image by a single type feature used in face recognition， a MBC feature and POEM feature-based face recognition scheme was proposed. Firstly，MBC and POEM coding patterns were extracted from normalized face images. Secondly， feature vector of every block was generated by dividing the MBC and POEM coding patterns into blocks. Finally， the classification capacity of features was enhanced by using weighted piecewise LDA algorithm. Recognition and verification test were carried out using the proposed algorithm on Dup1，Dup2， Fb and Fc， respectively， which were the four subsets of FERET. The recognition rates were 93.77%， 90.60%， 99.58%， and 99.49%， respectively， and the verification rates(false accepted rate is 0.1%) were 95.70%， 92.31%， 99.75%， and 100% ， respectively. All these results indicated the effectiveness of the proposed algorithm.

Key words: face recognition, MBC feature, POEM feature, feature fusion, weighted piecewise LDA

中图分类号: