A neural network face recognition system

Matthew Aitkenhead, Allan James Stuart McDonald

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

A neural network based facial recognition program (FADER-FAce DEtection and Recognition) was developed and tested. The hardware and software components were all standard commercial design, allowing the system to be built for minimal cost. Using a set of 1000 face and 1000 'no-face' images, we achieved 94.7% detection rate, and a 0.6% false positive rate. Three different neural network models were applied to face recognition, using single images of each Subject to train the system. A novel adaptation of the Hebbian connection strength adjustment model gave the best results, with 74.1% accuracy achieved. Each of the system's components, including an intermediate substructure detection network, was subject to evolutionary computation in order to optimise the system performance. (C) 2003 Elsevier Ltd. All rights reserved.

Original languageEnglish
Pages (from-to)167-176
Number of pages9
JournalArtificial Intelligence
Volume16
DOIs
Publication statusPublished - 2003

Keywords

  • neural networks
  • face recognition
  • substructure detection
  • evolutionary computation

Cite this

A neural network face recognition system. / Aitkenhead, Matthew; McDonald, Allan James Stuart.

In: Artificial Intelligence, Vol. 16, 2003, p. 167-176.

Research output: Contribution to journalArticle

Aitkenhead, Matthew ; McDonald, Allan James Stuart. / A neural network face recognition system. In: Artificial Intelligence. 2003 ; Vol. 16. pp. 167-176.
@article{e7f2440a15d249d19edb4e21ca9505f7,
title = "A neural network face recognition system",
abstract = "A neural network based facial recognition program (FADER-FAce DEtection and Recognition) was developed and tested. The hardware and software components were all standard commercial design, allowing the system to be built for minimal cost. Using a set of 1000 face and 1000 'no-face' images, we achieved 94.7{\%} detection rate, and a 0.6{\%} false positive rate. Three different neural network models were applied to face recognition, using single images of each Subject to train the system. A novel adaptation of the Hebbian connection strength adjustment model gave the best results, with 74.1{\%} accuracy achieved. Each of the system's components, including an intermediate substructure detection network, was subject to evolutionary computation in order to optimise the system performance. (C) 2003 Elsevier Ltd. All rights reserved.",
keywords = "neural networks, face recognition, substructure detection, evolutionary computation",
author = "Matthew Aitkenhead and McDonald, {Allan James Stuart}",
year = "2003",
doi = "10.1016/S0952-1976(03)00042-3",
language = "English",
volume = "16",
pages = "167--176",
journal = "Artificial Intelligence",
issn = "0004-3702",
publisher = "Elsevier",

}

TY - JOUR

T1 - A neural network face recognition system

AU - Aitkenhead, Matthew

AU - McDonald, Allan James Stuart

PY - 2003

Y1 - 2003

N2 - A neural network based facial recognition program (FADER-FAce DEtection and Recognition) was developed and tested. The hardware and software components were all standard commercial design, allowing the system to be built for minimal cost. Using a set of 1000 face and 1000 'no-face' images, we achieved 94.7% detection rate, and a 0.6% false positive rate. Three different neural network models were applied to face recognition, using single images of each Subject to train the system. A novel adaptation of the Hebbian connection strength adjustment model gave the best results, with 74.1% accuracy achieved. Each of the system's components, including an intermediate substructure detection network, was subject to evolutionary computation in order to optimise the system performance. (C) 2003 Elsevier Ltd. All rights reserved.

AB - A neural network based facial recognition program (FADER-FAce DEtection and Recognition) was developed and tested. The hardware and software components were all standard commercial design, allowing the system to be built for minimal cost. Using a set of 1000 face and 1000 'no-face' images, we achieved 94.7% detection rate, and a 0.6% false positive rate. Three different neural network models were applied to face recognition, using single images of each Subject to train the system. A novel adaptation of the Hebbian connection strength adjustment model gave the best results, with 74.1% accuracy achieved. Each of the system's components, including an intermediate substructure detection network, was subject to evolutionary computation in order to optimise the system performance. (C) 2003 Elsevier Ltd. All rights reserved.

KW - neural networks

KW - face recognition

KW - substructure detection

KW - evolutionary computation

U2 - 10.1016/S0952-1976(03)00042-3

DO - 10.1016/S0952-1976(03)00042-3

M3 - Article

VL - 16

SP - 167

EP - 176

JO - Artificial Intelligence

JF - Artificial Intelligence

SN - 0004-3702

ER -