Automatic Extraction of Vehicle, Motorcycle, Bicycle, and Pedestrian Traffic from Video Data

The objective of this research is to develop image processing algorithms to automatically extract vehicle counts and classifications, as well as counts of motorcycles, bicycles, and pedestrians from real-time and offline videos. An easy-to-use graphical user interface will enable SCDOT staff to obtain multimodal traffic data accurately, safely, and cost-effectively to use for HPMS reporting and prioritize infrastructure design improvements and investments.

• Record URL:
  https://scdot.scltap.org/wp-content/uploads/2022/11/SPR-742-Final-Report-2-8-22.pdf

Language

• English

Project

• Status: Completed
• Funding: $148,890.00
• Contract Numbers:

  SPR 742

• Sponsor Organizations:

  South Carolina Department of Transportation

  955 Park Street
  P.O. Box 191
  Columbia, SC  United States  29202-0191

  Federal Highway Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Managing Organizations:

  South Carolina Department of Transportation

  955 Park Street
  P.O. Box 191
  Columbia, SC  United States  29202-0191
• Project Managers:

  Swygert, Terry

  Watford, Jade

• Performing Organizations:

  University of South Carolina, Columbia

  502 Byrnes Building
  Columbia, SC  United States  29208
• Principal Investigators:

  Huynh, Nathan

• Start Date: 20190115
• Expected Completion Date: 20200715
• Actual Completion Date: 20211231

Subject/Index Terms

• TRT Terms: Image analysis; Image processing; Motorcycles; Pedestrian detectors; Vehicle characteristics; Vehicle classification; Video imaging detectors
• Identifier Terms: Highway Pavement Management Application
• Subject Areas: Pedestrians and Bicyclists; Planning and Forecasting;

Filing Info

• Accession Number: 01733138
• Record Type: Research project
• Source Agency: South Carolina Department of Transportation
• Contract Numbers: SPR 742
• Files: RIP, STATEDOT
• Created Date: Mar 9 2020 7:35AM