Enhanced method for recognizing gender in smart environments from gait biometric

Amer G. Binsaadoon, El Sayed M. El-Alfy

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

Abstract

Gender recognition is becoming an attractive research topic of increasing importance in demographic and medical studies applications. This paper presents an enhanced methodology for texture representation by a local Gabor-based phase quantization method, named LGPQ, and applies it for automatic human gender recognition using texture analysis of gait energy images (GEIs). Gait analysis and recognition is one of the recently explored areas in smart environments with superior features than other biometric traits such as the ability to be recognized at a distance even from low-resolution images. The proposed LGPQ has the advantage of capturing more information to encode the spatio-temporal variations in the Gabor transform of GEI. As a result, better gender classification can be achieved. The proposed method is evaluated using a support vector machine (SVM) classifier with a linear kernel. Moreover, it is compared to several texture-based methods on the CASIA B multi-view gait database. The experimental results demonstrate promising performance of LGPQ for gait-based gender recognition.

Original languageEnglish
Title of host publicationIET Conference Publications
PublisherInstitution of Engineering and Technology
EditionCP747
ISBN (Electronic)9781785618161, 9781785618437, 9781785618468, 9781785618871, 9781785619427, 9781785619694, 9781839530036
ISBN (Print)9781785617911
StatePublished - 2018

Publication series

NameIET Conference Publications
NumberCP747
Volume2018

Bibliographical note

Publisher Copyright:
© 2018 Institution of Engineering and Technology. All rights reserved.

Keywords

  • Biometric technology
  • Gender recognition
  • Local phase quantization
  • Smart environments
  • Texture analysis

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Enhanced method for recognizing gender in smart environments from gait biometric'. Together they form a unique fingerprint.

Cite this