Face recognition using artificial neural networks in parallel architecture

Batyrkhan Omarov, Azizah Suliman, Kaisar Kushibar

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Face detection and recognition is the main aspect for different important areas such as video surveillance, biometrics, interactive game applications, human computer interaction and access control systems. These systems require fast real time detection and recognition with high recognition rate. In this paper we propose implementation of the Artificial Neural Network by using high performance computing architecture based on Graphics Processing Unit to get face recognition with high accuracy and more speedup. There, we consider a parallel training approach for backpropagation algorithm for face recognition. For the high performance of face recognition it was used Compute Unified Device Architecture (CUDA) on a GPU. The experimental results demonstrate a significant decrease on executing times and greater speedup than serial implementation.

Original languageEnglish
Pages (from-to)238-248
Number of pages11
JournalJournal of Theoretical and Applied Information Technology
Volume91
Issue number2
Publication statusPublished - 30 Sep 2016

Fingerprint

Parallel architectures
Parallel Architectures
Face recognition
Face Recognition
Artificial Neural Network
Neural networks
Speedup
High Performance
Face Detection
Video Surveillance
Back-propagation Algorithm
Graphics Processing Unit
Access Control
Biometrics
Backpropagation algorithms
Human computer interaction
High Accuracy
Access control
Control System
Game

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{7436c2800f834b93a4c9775d1c64031e,
title = "Face recognition using artificial neural networks in parallel architecture",
abstract = "Face detection and recognition is the main aspect for different important areas such as video surveillance, biometrics, interactive game applications, human computer interaction and access control systems. These systems require fast real time detection and recognition with high recognition rate. In this paper we propose implementation of the Artificial Neural Network by using high performance computing architecture based on Graphics Processing Unit to get face recognition with high accuracy and more speedup. There, we consider a parallel training approach for backpropagation algorithm for face recognition. For the high performance of face recognition it was used Compute Unified Device Architecture (CUDA) on a GPU. The experimental results demonstrate a significant decrease on executing times and greater speedup than serial implementation.",
author = "Batyrkhan Omarov and Azizah Suliman and Kaisar Kushibar",
year = "2016",
month = "9",
day = "30",
language = "English",
volume = "91",
pages = "238--248",
journal = "Journal of Theoretical and Applied Information Technology",
issn = "1992-8645",
publisher = "Asian Research Publishing Network (ARPN)",
number = "2",

}

Face recognition using artificial neural networks in parallel architecture. / Omarov, Batyrkhan; Suliman, Azizah; Kushibar, Kaisar.

In: Journal of Theoretical and Applied Information Technology, Vol. 91, No. 2, 30.09.2016, p. 238-248.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Face recognition using artificial neural networks in parallel architecture

AU - Omarov, Batyrkhan

AU - Suliman, Azizah

AU - Kushibar, Kaisar

PY - 2016/9/30

Y1 - 2016/9/30

N2 - Face detection and recognition is the main aspect for different important areas such as video surveillance, biometrics, interactive game applications, human computer interaction and access control systems. These systems require fast real time detection and recognition with high recognition rate. In this paper we propose implementation of the Artificial Neural Network by using high performance computing architecture based on Graphics Processing Unit to get face recognition with high accuracy and more speedup. There, we consider a parallel training approach for backpropagation algorithm for face recognition. For the high performance of face recognition it was used Compute Unified Device Architecture (CUDA) on a GPU. The experimental results demonstrate a significant decrease on executing times and greater speedup than serial implementation.

AB - Face detection and recognition is the main aspect for different important areas such as video surveillance, biometrics, interactive game applications, human computer interaction and access control systems. These systems require fast real time detection and recognition with high recognition rate. In this paper we propose implementation of the Artificial Neural Network by using high performance computing architecture based on Graphics Processing Unit to get face recognition with high accuracy and more speedup. There, we consider a parallel training approach for backpropagation algorithm for face recognition. For the high performance of face recognition it was used Compute Unified Device Architecture (CUDA) on a GPU. The experimental results demonstrate a significant decrease on executing times and greater speedup than serial implementation.

UR - http://www.scopus.com/inward/record.url?scp=84989339650&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84989339650&partnerID=8YFLogxK

M3 - Article

VL - 91

SP - 238

EP - 248

JO - Journal of Theoretical and Applied Information Technology

JF - Journal of Theoretical and Applied Information Technology

SN - 1992-8645

IS - 2

ER -