Skip Navigation

Mid-America Transportation Center

Freeway Travel Time Estimation using Existing Fixed Traffic Sensors-A Computer-Vision-Based Vehicle Matching Approach

Final Report
click to download report


Researchers

  • Principal Investigator: Zhaozheng Yin ( yinz@mst.edu 573-341-6266)
  • Project Status
    Complete
    About this Project
    Brief Project Description & Background
    We explore a fundamental advancement in the theoretical and practical research related to travel time estimation in a freeway network by matching vehicles in a network of traffic surveillance cameras. The accurately computed travel time will sustain the transportation system in a manner that is more effective, more efficient, and more economically competitive.
    Research Objective
    Using a network of 317 traffic surveillance cameras in the greater St. Louis area, we propose to automatically detect and match vehicles in images captured by cameras at upstream and downstream detection points, then compute corresponding travel time from the matched vehicles.
    Potential Benefits
    The accurately computed travel time is of interest to both road users and road network operations, therefore sustaining the transportation system in a manner that is more effective, more efficient, and more economically competitive. The travel time estimation by our proposed computer-vision-based method can be integrated with other travel time estimation methods.
    Abstract
    Travel time information is of interest to both road users and road network operators. The direct travel time estimation method, such as probe vehicles, has high accuracy, but it requires a high probe rate to collect the complete travel time information of a road network, which is too costly for the daily operation of a road network. The indirect method by point sensors assumes stable speed within a roadway segment and it has low accuracy when the traffic becomes congested. We explore a fundamental advancement in the theoretical and practical research related to travel time estimation in a freeway network by matching vehicles in a network of traffic surveillance cameras. The accurately computed travel time will sustain the transportation system in a manner that is more effective, more efficient, and more economically competitive.
    Project Amount
    $ $38,681