L1-Norm-Based 2DLPP

Hao-Xin ZHAO, Hong-Jie XING*, Xi-Zhao WANG, Jun-Fen CHEN

*Corresponding author for this work

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

10 Citations (Scopus)

Abstract

In this paper, we propose a new L1-Norm-Based two-dimensional locality preserving projections (2DLPP-L1). Traditional 2D-LPP can preserve local structure and extract feature directly form matrices, which shows great advantages. However, it is based on L2 norm. It is well known that L2-norm-based criterion is sensitive to outliers. We generalize 2D-LPP to its corresponding L1-norm-based version, i.e. 2DLPP-L1, which is more robust against outliers. To evaluate the performance of 2DLPP-L1, several experiments are performed on the ORL face databases. Experimental results demonstrate that 2DLPP-L1 has better performance than its related methods.

Original languageEnglish
Title of host publicationProceedings of the 2011 Chinese Control and Decision Conference, CCDC 2011
PublisherIEEE
Pages1259-1264
Number of pages6
ISBN (Print)9781424487363
DOIs
Publication statusPublished - 2011
Externally publishedYes
Event2011 Chinese Control and Decision Conference, CCDC 2011 - Mianyang, China
Duration: 23 May 201125 May 2011

Publication series

NameChinese Control and Decision Conference, CCDC
PublisherIEEE
ISSN (Print)1948-9439
ISSN (Electronic)1948-9447

Conference

Conference2011 Chinese Control and Decision Conference, CCDC 2011
Country/TerritoryChina
CityMianyang
Period23/05/1125/05/11

Bibliographical note

This work is partly supported by the National Natural Science Foundation of China (No. 60903089; 61073121), the China Postdoctoral Science Foundation (No. 20080440820), the Natural Science Foundation of Hebei Province (No. F2009000231), the Scientific Research Project of Department of Education of Hebei Province (No. 2009410), the Postdoctoral Science Foundation of Hebei University, the Foundation of Hebei University (No. 2008123), and 2010 Baoding science Research and Development Project(10ZG008).

Keywords

  • 2DLPP
  • L1 norm
  • outliers
  • two dimensional projections

Fingerprint

Dive into the research topics of 'L1-Norm-Based 2DLPP'. Together they form a unique fingerprint.

Cite this