A new feature selection method for face recognition based on general data field

Long Zhao, Shuliang Wang, Yi Lin

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Feature selection is an important step when building a classifier for face recognition. It is difficult to classify the high dimensional and small sample data sets such as face data sets pose. Because the high dimensions increase the risk of over fitting and the small samples decrease the accuracy. A new feature selection method for face recognition based on general data field is proposed in this paper. This method adopts the Sw (potential value within class) and Sb (potential value between different classes) to calculate the information entropy of each feature. The representative features have been selected to structure classifier. Well known feature selection techniques for face data sets are implemented and compared with our present method to show its effectiveness. The experiments show that our algorithm effectively reduces the dimensionality of face data sets and keeps the classifier performance.

Original languageEnglish
Title of host publicationProceedings of the ASE BigData and SocialInformatics 2015, ASE BD and SI 2015
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450337359
DOIs
Publication statusPublished - 7 Oct 2015
EventASE BigData and SocialInformatics, ASE BD and SI 2015 - Kaohsiung, Taiwan, Province of China
Duration: 7 Oct 20159 Oct 2015

Publication series

NameACM International Conference Proceeding Series
Volume07-09-Ocobert-2015

Conference

ConferenceASE BigData and SocialInformatics, ASE BD and SI 2015
Country/TerritoryTaiwan, Province of China
CityKaohsiung
Period7/10/159/10/15

Keywords

  • Face recognition
  • Feature selection
  • General data field
  • Information entropy
  • Potential value

Fingerprint

Dive into the research topics of 'A new feature selection method for face recognition based on general data field'. Together they form a unique fingerprint.

Cite this