Automated Video-Based Traffic Count Analysis

The goal of this effort has been to develop techniques that could be applied to the detection and tracking of vehicles in overhead footage of intersections. To that end the project team has developed and published techniques for vehicle tracking based on detecting and tracking point features. The project team showed how this approach could be combined with image based background subtraction to find and group vehicles. In subsequent work the team showed how they could extract the 3D structure of a vehicle from a collection of feature tracks by exploiting the non-linearities associated with the pinhole projection model.


    • English


    • Status: Completed
    • Sponsor Organizations:

      University of Pennsylvania


      Office of the Assistant Secretary for Research and Technology

      University Transportation Centers Program
      Department of Transportation
      Washington, DC  United States  20590
    • Project Managers:

      Ehrlichman, Courtney

    • Start Date: 20160101
    • Expected Completion Date: 0
    • Actual Completion Date: 0

    Subject/Index Terms

    Filing Info

    • Accession Number: 01595873
    • Record Type: Research project
    • Source Agency: Technologies for Safe and Efficient Transportation University Transportation Center
    • Files: UTC, RiP
    • Created Date: Apr 8 2016 2:39PM