Automated vehicle density estimation from raw surveillance videos

Mehboob, Fozia, Abbas, Muhammad, Jiang, Richard, Tahir, Muhammad Atif, Al-Maadeed, Somaya and Bouridane, Ahmed (2016) Automated vehicle density estimation from raw surveillance videos. In: 2016 SAI Computing Conference (SAI). IEEE, pp. 1024-1030. ISBN 978-1-4673-8461-2

Full text not available from this repository.
Official URL: http://dx.doi.org/10.1109/SAI.2016.7556104

Abstract

To enable an effective traffic management and signal control, it is important to know the road traffic density. In recent years, video surveillance based systems and monitoring tools have been widely used for obtaining road traffic density for traffic management. To address the needs of autonomous traffic data extraction and video analysis, a vast body of research exists. However, these schemes are either prone to noise or the analysis methods are based on the manually provided data. Here, a state-of-the-art algorithm is developed for measuring the traffic density from the processing of surveillance videos obtained from different sources and conditions. The developed algorithm, keeping the user input to the minimum, automatically detects the traffic data. To get rid of the noise and false alarms, salient motion based method is used for the detection of the objects of interest. To show the efficacy of the proposed scheme, several raw surveillance videos are acquired and our algorithm is tested on them without any apriori information about the videos or their pertaining field conditions. For benchmark purposes, the outcomes of the developed algorithm are compared with that of a classical baseline method. The experimental results indicate that the traffic density is adequately determined and gives better accuracy than the classical approach. This is despite the fact that no threshold tuning for the individual videos is done in this algorithm.

Item Type: Book Section
Uncontrolled Keywords: Traffic Management, Density Estimation, Object Detection
Subjects: G900 Others in Mathematical and Computing Sciences
Department: Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User: Becky Skoyles
Date Deposited: 06 Jun 2019 12:06
Last Modified: 10 Oct 2019 18:31
URI: http://nrl.northumbria.ac.uk/id/eprint/39533

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics