A face detection and recognition system for intelligent vehicles

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Face recognition is one of the key components for future intelligent vehicle applications such as determining whether a person is authorized to operate the vehicle. This study describes the development and implementation of an automatic face recognition system in the car environment. The challenge is to build a fast and accurate system that is able to detect, recognize and verify a driver's identity with the constraint introduced in the car environment in daylight lighting conditions. A further constraint is to use a low-cost web camera to capture the frontal images. The system consists of two parts. The first is face detection, which is based on combining fast and classical Neural Networks (NN) methods. The second is face recognition and verification, which is based on combining Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) techniques. Lighting correction techniques are applied to improve the overall performance. The proposed system has been tested in the car environment and has a recognition rate of 91.4% with 0.75% false acceptance rate. Face detection is achieved in 3.7 sec while face recognition for a car with two persons authorized to operate the vehicle is 1.4 sec.

Original languageEnglish
Pages (from-to)507-515
Number of pages9
JournalInformation Technology Journal
Volume5
Issue number3
DOIs
Publication statusPublished - May 2006

Fingerprint

Intelligent vehicle highway systems
Face recognition
Railroad cars
Lighting
Discriminant analysis
Principal component analysis
Cameras
Neural networks
Costs

Keywords

  • Biometrics
  • Face detection
  • Face recognition
  • Intelligent vehicles
  • Linear Discriminant Analysis (LDA)
  • Neural Networks (NN)
  • Principal Component Analysis (PCA)

ASJC Scopus subject areas

  • Computer Science (miscellaneous)

Cite this

@article{bb6c60f71e264315919d9f495255e95a,
title = "A face detection and recognition system for intelligent vehicles",
abstract = "Face recognition is one of the key components for future intelligent vehicle applications such as determining whether a person is authorized to operate the vehicle. This study describes the development and implementation of an automatic face recognition system in the car environment. The challenge is to build a fast and accurate system that is able to detect, recognize and verify a driver's identity with the constraint introduced in the car environment in daylight lighting conditions. A further constraint is to use a low-cost web camera to capture the frontal images. The system consists of two parts. The first is face detection, which is based on combining fast and classical Neural Networks (NN) methods. The second is face recognition and verification, which is based on combining Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) techniques. Lighting correction techniques are applied to improve the overall performance. The proposed system has been tested in the car environment and has a recognition rate of 91.4{\%} with 0.75{\%} false acceptance rate. Face detection is achieved in 3.7 sec while face recognition for a car with two persons authorized to operate the vehicle is 1.4 sec.",
keywords = "Biometrics, Face detection, Face recognition, Intelligent vehicles, Linear Discriminant Analysis (LDA), Neural Networks (NN), Principal Component Analysis (PCA)",
author = "Ishak, {Khairul Anuar} and {Abdul Samad}, Salina and Aini Hussain",
year = "2006",
month = "5",
doi = "10.3923/itj.2006.507.515",
language = "English",
volume = "5",
pages = "507--515",
journal = "Information Technology Journal",
issn = "1812-5638",
publisher = "Asian Network for Scientific Information",
number = "3",

}

TY - JOUR

T1 - A face detection and recognition system for intelligent vehicles

AU - Ishak, Khairul Anuar

AU - Abdul Samad, Salina

AU - Hussain, Aini

PY - 2006/5

Y1 - 2006/5

N2 - Face recognition is one of the key components for future intelligent vehicle applications such as determining whether a person is authorized to operate the vehicle. This study describes the development and implementation of an automatic face recognition system in the car environment. The challenge is to build a fast and accurate system that is able to detect, recognize and verify a driver's identity with the constraint introduced in the car environment in daylight lighting conditions. A further constraint is to use a low-cost web camera to capture the frontal images. The system consists of two parts. The first is face detection, which is based on combining fast and classical Neural Networks (NN) methods. The second is face recognition and verification, which is based on combining Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) techniques. Lighting correction techniques are applied to improve the overall performance. The proposed system has been tested in the car environment and has a recognition rate of 91.4% with 0.75% false acceptance rate. Face detection is achieved in 3.7 sec while face recognition for a car with two persons authorized to operate the vehicle is 1.4 sec.

AB - Face recognition is one of the key components for future intelligent vehicle applications such as determining whether a person is authorized to operate the vehicle. This study describes the development and implementation of an automatic face recognition system in the car environment. The challenge is to build a fast and accurate system that is able to detect, recognize and verify a driver's identity with the constraint introduced in the car environment in daylight lighting conditions. A further constraint is to use a low-cost web camera to capture the frontal images. The system consists of two parts. The first is face detection, which is based on combining fast and classical Neural Networks (NN) methods. The second is face recognition and verification, which is based on combining Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) techniques. Lighting correction techniques are applied to improve the overall performance. The proposed system has been tested in the car environment and has a recognition rate of 91.4% with 0.75% false acceptance rate. Face detection is achieved in 3.7 sec while face recognition for a car with two persons authorized to operate the vehicle is 1.4 sec.

KW - Biometrics

KW - Face detection

KW - Face recognition

KW - Intelligent vehicles

KW - Linear Discriminant Analysis (LDA)

KW - Neural Networks (NN)

KW - Principal Component Analysis (PCA)

UR - http://www.scopus.com/inward/record.url?scp=33749593100&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749593100&partnerID=8YFLogxK

U2 - 10.3923/itj.2006.507.515

DO - 10.3923/itj.2006.507.515

M3 - Article

VL - 5

SP - 507

EP - 515

JO - Information Technology Journal

JF - Information Technology Journal

SN - 1812-5638

IS - 3

ER -