Research on Urban Traffic Incident Detection Based on Vehicle Cameras

Zhuofei Xia, Jiayuan Gong, Hailong Yu, Wenbo Ren and Jingnan Wang
Additional contact information
Zhuofei Xia: Institute of Automotive Engineers, Hubei University of Automotive Technology, Shiyan 442002, China
Jiayuan Gong: Institute of Automotive Engineers, Hubei University of Automotive Technology, Shiyan 442002, China
Hailong Yu: Institute of Medical Robotics, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
Wenbo Ren: Institute of Automotive Engineers, Hubei University of Automotive Technology, Shiyan 442002, China
Jingnan Wang: Institute of Automotive Engineers, Hubei University of Automotive Technology, Shiyan 442002, China

Future Internet, 2022, vol. 14, issue 8, 1-17

Abstract: Situational detection in the traffic system is of great significance to traffic management and even urban management. Traditional detection methods are generally based on roadside equipment monitoring roads, and it is difficult to support large-scale and fine-grained traffic incident detection. In this study, we propose a detection method applied to the mobile edge, which detects traffic incidents based on the video captured by vehicle cameras, so as to overcome the limitations of roadside terminal perception. For swarm intelligence detection, we propose an improved YOLOv5s object detection network, adding an atrous pyramid pooling layer to the network and introducing a fusion attention mechanism to improve the model accuracy. Compared with the raw YOLOv5s, the mAP metrics of our improved model are increased by 3.3% to 84.2%, enabling it to detect vehicles, pedestrians, traffic accidents, and fire traffic incidents on the road with high precision in real time. This provides information for city managers to help them grasp the abnormal operation status of roads and cities in a timely and effective manner.

Keywords: traffic incident; attention mechanism; object detection; computer vision (search for similar items in EconPapers)
JEL-codes: O3 (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1999-5903/14/8/227/pdf (application/pdf)
https://www.mdpi.com/1999-5903/14/8/227/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jftint:v:14:y:2022:i:8:p:227-:d:872591

Access Statistics for this article

Future Internet is currently edited by Ms. Grace You

More articles in Future Internet from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().