A qualitative and quantitative comparison of real-time background subtraction algorithms for video surveillance applications

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Background subtraction is a widely used technique for segmenting a foreground object from its background The aim of this paper is to review and compare the performance of the most common statistical background subtraction methods, including median-based, Gaussian-based and Kernel density-based approaches. To obtain a fair evaluation, four challenging scenarios were selected based on Wallflower datasets. All review methods are based on processing speed, memory usage and segmentation accuracy. The overall evaluation shows that the Gaussian-based method gives the best performance in accuracy, speed and memory consumption. In addition, this paper provides a better understanding of algorithm behaviours applied to different situations for real-time video surveillance applications. 1553-9105/

Original languageEnglish
Pages (from-to)493-505
Number of pages13
JournalJournal of Computational Information Systems
Volume8
Issue number2
Publication statusPublished - Feb 2012

Fingerprint

Data storage equipment
Processing

Keywords

  • Background subtraction
  • Gaussian mixture modal
  • KDE
  • Median
  • Real-time video surveillance

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems

Cite this

@article{60d2138e7b2e48a0978b5829d273bd19,
title = "A qualitative and quantitative comparison of real-time background subtraction algorithms for video surveillance applications",
abstract = "Background subtraction is a widely used technique for segmenting a foreground object from its background The aim of this paper is to review and compare the performance of the most common statistical background subtraction methods, including median-based, Gaussian-based and Kernel density-based approaches. To obtain a fair evaluation, four challenging scenarios were selected based on Wallflower datasets. All review methods are based on processing speed, memory usage and segmentation accuracy. The overall evaluation shows that the Gaussian-based method gives the best performance in accuracy, speed and memory consumption. In addition, this paper provides a better understanding of algorithm behaviours applied to different situations for real-time video surveillance applications. 1553-9105/",
keywords = "Background subtraction, Gaussian mixture modal, KDE, Median, Real-time video surveillance",
author = "M. Hedayati and {Wan Zaki}, {Wan Mimi Diyana} and Aini Hussain",
year = "2012",
month = "2",
language = "English",
volume = "8",
pages = "493--505",
journal = "Journal of Computational Information Systems",
issn = "1553-9105",
publisher = "Binary Information Press",
number = "2",

}

TY - JOUR

T1 - A qualitative and quantitative comparison of real-time background subtraction algorithms for video surveillance applications

AU - Hedayati, M.

AU - Wan Zaki, Wan Mimi Diyana

AU - Hussain, Aini

PY - 2012/2

Y1 - 2012/2

N2 - Background subtraction is a widely used technique for segmenting a foreground object from its background The aim of this paper is to review and compare the performance of the most common statistical background subtraction methods, including median-based, Gaussian-based and Kernel density-based approaches. To obtain a fair evaluation, four challenging scenarios were selected based on Wallflower datasets. All review methods are based on processing speed, memory usage and segmentation accuracy. The overall evaluation shows that the Gaussian-based method gives the best performance in accuracy, speed and memory consumption. In addition, this paper provides a better understanding of algorithm behaviours applied to different situations for real-time video surveillance applications. 1553-9105/

AB - Background subtraction is a widely used technique for segmenting a foreground object from its background The aim of this paper is to review and compare the performance of the most common statistical background subtraction methods, including median-based, Gaussian-based and Kernel density-based approaches. To obtain a fair evaluation, four challenging scenarios were selected based on Wallflower datasets. All review methods are based on processing speed, memory usage and segmentation accuracy. The overall evaluation shows that the Gaussian-based method gives the best performance in accuracy, speed and memory consumption. In addition, this paper provides a better understanding of algorithm behaviours applied to different situations for real-time video surveillance applications. 1553-9105/

KW - Background subtraction

KW - Gaussian mixture modal

KW - KDE

KW - Median

KW - Real-time video surveillance

UR - http://www.scopus.com/inward/record.url?scp=84859061885&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859061885&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84859061885

VL - 8

SP - 493

EP - 505

JO - Journal of Computational Information Systems

JF - Journal of Computational Information Systems

SN - 1553-9105

IS - 2

ER -