A comparative study of face recognition using improved AAM, PCA and ICA via Feret date base

Khamiss Masaoud S Algabary, Khairuddin Omar

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Face identification is a popular research area where there are different methods studied in the literature. An Active Appearance Model (AAM) based feature extraction in geometrical illumination-invariant matching approach is studied in details, to enhance invariant recognition of human faces and analysis the face verification and identification performance. This paper proposed to develop a general model of pattern appearance that can be used to detect and classify patterns from face images. A digital camera integrated with computer based software designed was built to capture the face and compare it with the faces stored in the Face Recognition Technology (FERET) database. The results are compared with respect to the effects of changes in illumination. The experiments performed on illumination part of the FERET face data set. Simulation results confirmed that Subspace AAM technique performed slightly well comparing with PCA and ICA by 95.99%, 90% and 86% respectively.

Original languageEnglish
Pages (from-to)263-272
Number of pages10
JournalEuropean Journal of Scientific Research
Volume48
Issue number2
Publication statusPublished - 2010

Fingerprint

Active Appearance Models
Passive Cutaneous Anaphylaxis
Independent component analysis
Face recognition
Lighting
Face Recognition
Date
Comparative Study
comparative study
Face
Technology
Illumination
Digital cameras
lighting
Feature extraction
Software
Databases
software
Invariant
Research

Keywords

  • AAM
  • Digital camera
  • Error rate
  • Face recognition
  • False acceptance
  • False rejection
  • FERET database
  • Image ICA
  • PCA

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Earth and Planetary Sciences(all)
  • Engineering(all)
  • Materials Science(all)
  • Mathematics(all)
  • Computer Science(all)

Cite this

A comparative study of face recognition using improved AAM, PCA and ICA via Feret date base. / Algabary, Khamiss Masaoud S; Omar, Khairuddin.

In: European Journal of Scientific Research, Vol. 48, No. 2, 2010, p. 263-272.

Research output: Contribution to journalArticle

@article{fc80174ace1c4e82aee09c00f0c91bfa,
title = "A comparative study of face recognition using improved AAM, PCA and ICA via Feret date base",
abstract = "Face identification is a popular research area where there are different methods studied in the literature. An Active Appearance Model (AAM) based feature extraction in geometrical illumination-invariant matching approach is studied in details, to enhance invariant recognition of human faces and analysis the face verification and identification performance. This paper proposed to develop a general model of pattern appearance that can be used to detect and classify patterns from face images. A digital camera integrated with computer based software designed was built to capture the face and compare it with the faces stored in the Face Recognition Technology (FERET) database. The results are compared with respect to the effects of changes in illumination. The experiments performed on illumination part of the FERET face data set. Simulation results confirmed that Subspace AAM technique performed slightly well comparing with PCA and ICA by 95.99{\%}, 90{\%} and 86{\%} respectively.",
keywords = "AAM, Digital camera, Error rate, Face recognition, False acceptance, False rejection, FERET database, Image ICA, PCA",
author = "Algabary, {Khamiss Masaoud S} and Khairuddin Omar",
year = "2010",
language = "English",
volume = "48",
pages = "263--272",
journal = "European Journal of Scientific Research",
issn = "1450-202X",
publisher = "European Journals Inc.",
number = "2",

}

TY - JOUR

T1 - A comparative study of face recognition using improved AAM, PCA and ICA via Feret date base

AU - Algabary, Khamiss Masaoud S

AU - Omar, Khairuddin

PY - 2010

Y1 - 2010

N2 - Face identification is a popular research area where there are different methods studied in the literature. An Active Appearance Model (AAM) based feature extraction in geometrical illumination-invariant matching approach is studied in details, to enhance invariant recognition of human faces and analysis the face verification and identification performance. This paper proposed to develop a general model of pattern appearance that can be used to detect and classify patterns from face images. A digital camera integrated with computer based software designed was built to capture the face and compare it with the faces stored in the Face Recognition Technology (FERET) database. The results are compared with respect to the effects of changes in illumination. The experiments performed on illumination part of the FERET face data set. Simulation results confirmed that Subspace AAM technique performed slightly well comparing with PCA and ICA by 95.99%, 90% and 86% respectively.

AB - Face identification is a popular research area where there are different methods studied in the literature. An Active Appearance Model (AAM) based feature extraction in geometrical illumination-invariant matching approach is studied in details, to enhance invariant recognition of human faces and analysis the face verification and identification performance. This paper proposed to develop a general model of pattern appearance that can be used to detect and classify patterns from face images. A digital camera integrated with computer based software designed was built to capture the face and compare it with the faces stored in the Face Recognition Technology (FERET) database. The results are compared with respect to the effects of changes in illumination. The experiments performed on illumination part of the FERET face data set. Simulation results confirmed that Subspace AAM technique performed slightly well comparing with PCA and ICA by 95.99%, 90% and 86% respectively.

KW - AAM

KW - Digital camera

KW - Error rate

KW - Face recognition

KW - False acceptance

KW - False rejection

KW - FERET database

KW - Image ICA

KW - PCA

UR - http://www.scopus.com/inward/record.url?scp=79955753260&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955753260&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:79955753260

VL - 48

SP - 263

EP - 272

JO - European Journal of Scientific Research

JF - European Journal of Scientific Research

SN - 1450-202X

IS - 2

ER -