A Three-Stage Anomaly Detection Framework for Traffic Videos

As reported by the United Nations in 2021, road accidents cause 1.3 million deaths and 50 million injuries worldwide each year. Detecting traffic anomalies timely and taking immediate emergency response and rescue measures are essential to reduce casualties, economic losses, and traffic congestion....

Full description

Saved in:
Bibliographic Details
Main Authors: Junzhou Chen, Jiancheng Wang, Jiajun Pu, Ronghui Zhang
Format: Article
Language:English
Published: Wiley 2022-01-01
Series:Journal of Advanced Transportation
Online Access:http://dx.doi.org/10.1155/2022/9463559
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832558373634048000
author Junzhou Chen
Jiancheng Wang
Jiajun Pu
Ronghui Zhang
author_facet Junzhou Chen
Jiancheng Wang
Jiajun Pu
Ronghui Zhang
author_sort Junzhou Chen
collection DOAJ
description As reported by the United Nations in 2021, road accidents cause 1.3 million deaths and 50 million injuries worldwide each year. Detecting traffic anomalies timely and taking immediate emergency response and rescue measures are essential to reduce casualties, economic losses, and traffic congestion. This paper proposed a three-stage method for video-based traffic anomaly detection. In the first stage, the ViVit network is employed as a feature extractor to capture the spatiotemporal features from the input video. In the second stage, the class and patch tokens are fed separately to the segment-level and video-level traffic anomaly detectors. In the third stage, we finished the construction of the entire composite traffic anomaly detection framework by fusing outputs of two traffic anomaly detectors above with different granularity. Experimental evaluation demonstrates that the proposed method outperforms the SOTA method with 2.07% AUC on the TAD testing overall set and 1.43% AUC on the TAD testing anomaly subset. This work provides a new reference for traffic anomaly detection research.
format Article
id doaj-art-28784bc89727444b9885a687ce052f4f
institution Kabale University
issn 2042-3195
language English
publishDate 2022-01-01
publisher Wiley
record_format Article
series Journal of Advanced Transportation
spelling doaj-art-28784bc89727444b9885a687ce052f4f2025-02-03T01:32:31ZengWileyJournal of Advanced Transportation2042-31952022-01-01202210.1155/2022/9463559A Three-Stage Anomaly Detection Framework for Traffic VideosJunzhou Chen0Jiancheng Wang1Jiajun Pu2Ronghui Zhang3School of Intelligent Systems EngineeringSchool of Intelligent Systems EngineeringSchool of Intelligent Systems EngineeringSchool of Intelligent Systems EngineeringAs reported by the United Nations in 2021, road accidents cause 1.3 million deaths and 50 million injuries worldwide each year. Detecting traffic anomalies timely and taking immediate emergency response and rescue measures are essential to reduce casualties, economic losses, and traffic congestion. This paper proposed a three-stage method for video-based traffic anomaly detection. In the first stage, the ViVit network is employed as a feature extractor to capture the spatiotemporal features from the input video. In the second stage, the class and patch tokens are fed separately to the segment-level and video-level traffic anomaly detectors. In the third stage, we finished the construction of the entire composite traffic anomaly detection framework by fusing outputs of two traffic anomaly detectors above with different granularity. Experimental evaluation demonstrates that the proposed method outperforms the SOTA method with 2.07% AUC on the TAD testing overall set and 1.43% AUC on the TAD testing anomaly subset. This work provides a new reference for traffic anomaly detection research.http://dx.doi.org/10.1155/2022/9463559
spellingShingle Junzhou Chen
Jiancheng Wang
Jiajun Pu
Ronghui Zhang
A Three-Stage Anomaly Detection Framework for Traffic Videos
Journal of Advanced Transportation
title A Three-Stage Anomaly Detection Framework for Traffic Videos
title_full A Three-Stage Anomaly Detection Framework for Traffic Videos
title_fullStr A Three-Stage Anomaly Detection Framework for Traffic Videos
title_full_unstemmed A Three-Stage Anomaly Detection Framework for Traffic Videos
title_short A Three-Stage Anomaly Detection Framework for Traffic Videos
title_sort three stage anomaly detection framework for traffic videos
url http://dx.doi.org/10.1155/2022/9463559
work_keys_str_mv AT junzhouchen athreestageanomalydetectionframeworkfortrafficvideos
AT jianchengwang athreestageanomalydetectionframeworkfortrafficvideos
AT jiajunpu athreestageanomalydetectionframeworkfortrafficvideos
AT ronghuizhang athreestageanomalydetectionframeworkfortrafficvideos
AT junzhouchen threestageanomalydetectionframeworkfortrafficvideos
AT jianchengwang threestageanomalydetectionframeworkfortrafficvideos
AT jiajunpu threestageanomalydetectionframeworkfortrafficvideos
AT ronghuizhang threestageanomalydetectionframeworkfortrafficvideos