Full Length Research Paper

Vehicle speed detection in video image sequences using CVS method

Arash Gholami Rad^1,3*, Abbas Dehghani² and Mohamed Rehan Karim¹

¹Department of civil Engineering, University of Malaya. 50603 Kuala Lumpur, Malaysia.

²Department of Electrical Engineering, Sadjad Institute of Higher Education, Mashhad, Iran.

³Department of Computer Science, UCLA, 420 Westwood Plaza, Los Angeles, CA 90095, USA.

*Corresponding author. E-mail: Hsara56@perdana.um.edu.my.

Accepted 5 November, 2010

Abstract

Video and image processing has been used for traffic surveillance, analysis and monitoring of traffic conditions in many cities and urban areas. This paper aims to present another approach to estimate the vehicles velocity. In this study, the captured traffic movies are collected with a stationary camera which is mounted on a freeway. The camera was calibrated based on geometrical equations that were supported directly by using references. Camera calibration for exact measurements may be possible while accurate speed estimation can still be quite difficult to achieve. The designed system has the ability to be extended to another related traffic application. The average error of the detected vehicle speed was ± 7 km/h and the experiment was operated at different resolutions and different video sequences.

Key words: Vehicle speed detection, video sequence, computer vision, background modeling, traffic monitoring.