Efficient and effective automated surveillance agents using kernel tricks

Many schemes have been presented over the years to develop automated visual surveillance systems. However, these schemes typically need custom equipment, or involve significant complexity and storage requirements. In this paper we present three software-based agents built using kernel machines to pe...

Full description

Bibliographic Details
Main Authors:	Ahmed, Tarem, Wei, Xianglin, Ahmed, Supriyo, Pathan, Al-Sakib Khan
Format:	Article
Language:	English English
Published:	Society for Modeling & Simulation International 2012
Subjects:	QA75 Electronic computers. Computer science QA76 Computer software
Online Access:	http://irep.iium.edu.my/25385/ http://irep.iium.edu.my/25385/ http://irep.iium.edu.my/25385/1/ms_S-12-0061-Revised-ACCEPTED.pdf http://irep.iium.edu.my/25385/4/SIMULATION.pdf

id	iium-25385
recordtype	eprints
spelling	iium-253852014-07-16T08:04:42Z http://irep.iium.edu.my/25385/ Efficient and effective automated surveillance agents using kernel tricks Ahmed, Tarem Wei, Xianglin Ahmed, Supriyo Pathan, Al-Sakib Khan QA75 Electronic computers. Computer science QA76 Computer software Many schemes have been presented over the years to develop automated visual surveillance systems. However, these schemes typically need custom equipment, or involve significant complexity and storage requirements. In this paper we present three software-based agents built using kernel machines to perform automated, real-time intruder detection in surveillance systems. Kernel machines provide a powerful data mining technique that may be used for pattern matching in the presence of complex data. They work by first mapping the raw input data onto a (often much) higher dimensional feature space, and then clustering in the feature space instead. The reasoning is that mapping onto the (higher-dimensional) feature space enables the comparison of additional, higher order correlations in determining patterns between the raw data points. The agents proposed here have been built using algorithms that are adaptive, portable, do not require any expensive or sophisticated components, and are lightweight and efficient having run times of the order of hundredths of a second. Through application to real image streams from a simple, run-of-the-mill closed-circuit television surveillance system, and direct quantitative performance comparison with some existing schemes, we show that it is possible to easily obtain high detection accuracy with low computational and storage complexities. Society for Modeling & Simulation International 2012 Article PeerReviewed application/pdf en http://irep.iium.edu.my/25385/1/ms_S-12-0061-Revised-ACCEPTED.pdf application/pdf en http://irep.iium.edu.my/25385/4/SIMULATION.pdf Ahmed, Tarem and Wei, Xianglin and Ahmed, Supriyo and Pathan, Al-Sakib Khan (2012) Efficient and effective automated surveillance agents using kernel tricks. SIMULATION: Transactions of the SCS. ISSN Print ISSN: 0037-5497, Online ISSN: 1741-3133 (In Press) http://sim.sagepub.com/
repository_type	Digital Repository
institution_category	Local University
institution	International Islamic University Malaysia
building	IIUM Repository
collection	Online Access
language	English English
topic	QA75 Electronic computers. Computer science QA76 Computer software
spellingShingle	QA75 Electronic computers. Computer science QA76 Computer software Ahmed, Tarem Wei, Xianglin Ahmed, Supriyo Pathan, Al-Sakib Khan Efficient and effective automated surveillance agents using kernel tricks
description	Many schemes have been presented over the years to develop automated visual surveillance systems. However, these schemes typically need custom equipment, or involve significant complexity and storage requirements. In this paper we present three software-based agents built using kernel machines to perform automated, real-time intruder detection in surveillance systems. Kernel machines provide a powerful data mining technique that may be used for pattern matching in the presence of complex data. They work by first mapping the raw input data onto a (often much) higher dimensional feature space, and then clustering in the feature space instead. The reasoning is that mapping onto the (higher-dimensional) feature space enables the comparison of additional, higher order correlations in determining patterns between the raw data points. The agents proposed here have been built using algorithms that are adaptive, portable, do not require any expensive or sophisticated components, and are lightweight and efficient having run times of the order of hundredths of a second. Through application to real image streams from a simple, run-of-the-mill closed-circuit television surveillance system, and direct quantitative performance comparison with some existing schemes, we show that it is possible to easily obtain high detection accuracy with low computational and storage complexities.
format	Article
author	Ahmed, Tarem Wei, Xianglin Ahmed, Supriyo Pathan, Al-Sakib Khan
author_facet	Ahmed, Tarem Wei, Xianglin Ahmed, Supriyo Pathan, Al-Sakib Khan
author_sort	Ahmed, Tarem
title	Efficient and effective automated surveillance agents using kernel tricks
title_short	Efficient and effective automated surveillance agents using kernel tricks
title_full	Efficient and effective automated surveillance agents using kernel tricks
title_fullStr	Efficient and effective automated surveillance agents using kernel tricks
title_full_unstemmed	Efficient and effective automated surveillance agents using kernel tricks
title_sort	efficient and effective automated surveillance agents using kernel tricks
publisher	Society for Modeling & Simulation International
publishDate	2012
url	http://irep.iium.edu.my/25385/ http://irep.iium.edu.my/25385/ http://irep.iium.edu.my/25385/1/ms_S-12-0061-Revised-ACCEPTED.pdf http://irep.iium.edu.my/25385/4/SIMULATION.pdf
first_indexed	2023-09-18T20:37:54Z
last_indexed	2023-09-18T20:37:54Z
_version_	1777409169611030528

Efficient and effective automated surveillance agents using kernel tricks

Similar Items