Deep Learning-based Prediction of Traffic Accident Risk in Vehicular Networks

Haitao Zhao, Jun Zhang, Xiaoqing Li, Qin Wang, Hongbo Zhu

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review


With the growing quantity of vehicles, traffic security is in a grim state. In order to improve the safety of road traffic, this paper proposes a forecasting algorithm of traffic accident risk based on deep learning for edge-cloud internet of vehicles. Specifically, the gathered real-time traffic data is input into a convolutional neural network (CNN) for feature extraction. Then, the output of CNN is input in a random forest for feature classification, and the risk of traffic accidents can be predicted. The edge servers pick the warnings with the high risk of traffic accidents and transmit them to the corresponding vehicle units. The drivers can reduce the risk of traffic accidents via adjusting their behaviors according to the warnings. Simulations show that the proposed forecasting algorithm has a larger area under the curve of Receiver Operating Characteristic, higher accuracy, and lower loss than the CNN based method.

Original languageEnglish
Title of host publication2020 IEEE Globecom Workshops, GC Wkshps 2020 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728173078
Publication statusPublished - Dec 2020
Event2020 IEEE Globecom Workshops, GC Wkshps 2020 - Virtual, Taipei, Taiwan
Duration: 7 Dec 202011 Dec 2020

Publication series

Name2020 IEEE Globecom Workshops, GC Wkshps 2020 - Proceedings


Conference2020 IEEE Globecom Workshops, GC Wkshps 2020
CityVirtual, Taipei


  • accident prediction
  • Road traffic
  • Vehicular Networks

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Software

Cite this