Jump to the content zone at the center

Intersection Traffic Cameras and Conflict Detection Empirical Project

  • Category:Smart Transportation

Status:Complete


In a conventional field of traffic accidents, most of the improvement plans on accident-prone intersections refer to the three types of traffic accident data specified by the Ministry of Transportation and Communications, namely A1, A2, and A3. Moreover, the statistics are analyzed via the MOTC Road Safety Information Platform to screen out the information on accident-prone sites, including road segments, time, type of traffic accidents, car model of accidents, cause determination, and accident site maps. Then, the characteristics of each site are reviewed in written format, or the relevant agencies will co-survey the site to discuss the solutions for improvement. 

 

This PoC project is a collaboration between the Traffic Engineering Office of Taipei City Government and Asia Pacific Telecom Co., which plans to install cameras and AI servers at the intersection of testing fields, in order to analyze the identification and records of car models, traffic flow trajectory, and relevant coordinates. This will establish the indicators of traffic conflict incidents, including conflicts and level of accidents, in order to calculate the technological results from the development of automatic road traffic safety diagnosis. The competent authorities will use such data as a reference for making decisions on the improvement plans, upgrading the effectiveness of accident prevention, and reducing the number of road conflicts occurring at the accident-prone road segments. Meanwhile, the instant analysis of image transmission also reduces the number of surveys conducted by personnel at the sites.