A Video Structural Event Description Model for Traffic Surveillance System

DOI

10.2991/amitp-16.2016.69How to use a DOI?

Keywords

video structural analysis, event description, event detection, camshift algorithm

Abstract

The traditional video event detection only identifies a single event for one model and needs to extract features manually to train a mathematical model. In order to automatically extract features of objects and detect various potential accidents in the video surveillance, a structural analysis based approach to detect traffic events is proposed. In the surveillance video, a structural model is used in this way to collect information of the moving targets and the background things, including objects, attributes, temporal relationships and spatial relationships. To further detect the hidden accidents of traffic video, those above five elements are used to make arithmetical logic expression. It can be adapted for different scenarios easily. The experimental results demonstrate that the approach proposed can detect multiple traffic events in surveillance effectively.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Lei Xu
AU  - Jianxin Song
PY  - 2016/09
DA  - 2016/09
TI  - A Video Structural Event Description Model for Traffic Surveillance System
BT  - Proceedings of the 2016 4th International Conference on Advanced Materials and Information Technology Processing (AMITP 2016)
PB  - Atlantis Press
SP  - 356
EP  - 361
SN  - 2352-538X
UR  - https://doi.org/10.2991/amitp-16.2016.69
DO  - 10.2991/amitp-16.2016.69
ID  - Xu2016/09
ER  -