Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (8): 175-177.DOI: 10.3778/j.issn.1002-8331.2009.08.053

• 图形、图像、模式识别 • Previous Articles     Next Articles

Application of localized subspace projections on ear recognition

LENG Jia-fu1,MU Zhi-chun1,WAN Jie2   

  1. 1.School of Information Engineering,University of Science and Technology Beijing,Beijing 100083,China
    2.BERIS Engineering and Research Corporation,Baotou,Inner Mongolia 014010,China
  • Received:2008-08-29 Revised:2008-12-03 Online:2009-03-11 Published:2009-03-11
  • Contact: LENG Jia-fu

局部子空间映射在人耳识别中的应用

冷加福1,穆志纯1,万 捷2   

  1. 1.北京科技大学 信息工程学院,北京 100083
    2.中冶东方工程技术有限公司,内蒙古 包头 014010
  • 通讯作者: 冷加福

Abstract: Three different localized representation methods to face recognition are compared in terms of recognition accuracy.A systematic comparative analysis is conducted in terms of distance metric used,number of selected features on USTB2 ear database.Results indicate that the performances vary significantly upon the distance metric used.

Key words: ear recognition, Local Nonnegative Matrix Factorization(LNMF), Independent Component Analysis(ICA), Nonnegative Matrix Factorization with sparse constraints(NMFsc)

摘要: 比较了三种用于人耳识别的局部表征方法的识别准确率。通过在USTB2人耳库上的实验,对所采用的空间几何距离及选择的特征数进行了一个系统的比较分析。结果表明,方法的识别性能随所用的空间几何距离公式发生较大变化。

关键词: 人耳识别, 局部非负矩阵分解, 独立分量分析, 非负矩阵稀疏分解