Skip to Main content Skip to Navigation
Conference papers

Adaptive video-based algorithm for accident detection on highways

Abstract : For the past few decades, automatic accident detection, especially using video analysis, has become a very important subject. It is important not only for traffic management but also, for Intelligent Transportation Systems (ITS) through its contribution to avoid the escalation of accidents especially on highways. In this paper a novel vision-based road accident detection algorithm on highways and expressways is proposed. This algorithm is based on an adaptive traffic motion flow modeling technique, using Farneback Optical Flow for motions detection and a statistic heuristic method for accident detection. The algorithm was applied on a set of collected videos of traffic and accidents on highways. The results prove the efficiency and practicability of the proposed algorithm using only 240 frames for traffic motion modeling. This method avoids to utilization of a large database while adequate and common accidents videos benchmarks do not exist.
Document type :
Conference papers
Complete list of metadata

https://hal-uphf.archives-ouvertes.fr/hal-03388343
Contributor : Kathleen Torck Connect in order to contact the contributor
Submitted on : Wednesday, October 20, 2021 - 1:22:43 PM
Last modification on : Wednesday, November 3, 2021 - 4:43:29 AM

Identifiers

Collections

Citation

Boutheina Maaloul, Abdelmalik Taleb-Ahmed, Smail Niar, Naim Harb, Carlos Valderrama. Adaptive video-based algorithm for accident detection on highways. 2017 12th IEEE International Symposium on Industrial Embedded Systems (SIES), Jun 2017, Toulouse, France. pp.1-6, ⟨10.1109/SIES.2017.7993382⟩. ⟨hal-03388343⟩

Share

Metrics

Record views

18