Humphreys, James and Hunter, Andrew (2009) Multiple object tracking using a neural cost function. Image and vision computing, 27 (4). pp. 417-424. ISSN 0262-8856

Preview

PDF Humphreys2009IVCMultipleObjectTrackingUsingANeuralCostFunction.pdf - Whole Document Download (281Kb)

Abstract

This paper presents a new approach to the tracking of multiple objects in CCTV surveillance using a combination of simple neural cost functions based on Self-Organizing Maps, and a greedy assignment algorithm. Using a reference standard data set and an exhaustive search algorithm for benchmarking, we show that the cost function plays the most significant role in realizing high levels of performance. The neural cost function’s context-sensitive treatment of appearance, change of appearance and trajectory yield better tracking than a simple, explicitly designed cost function. The algorithm matches 98.8% of objects to within 15 pixels.

Item Type:	Article
Keywords:	Surveillance, Tracking, Background differencing, Self-organizing maps, Neural networks, ref11, refdoi
Subjects:	G Mathematical and Computer Sciences > G400 Computer Science
Divisions:	College of Sciences > Faculty of Science > Lincoln School of Computer Science
Depositing User:	Rosaline Smith
Date Deposited:	10 Jul 2010 20:14
Last Modified:	15 May 2013 08:24
URI:	http://eprints.lincoln.ac.uk/id/eprint/2755

Actions (login required)

View Item