AI-Enabled Vision System for Intersection Analytics

Phase I of this project revealed limitations of using a single camera per intersection to automatically extract key traffic performance and safety information from video feeds. To overcome these limitations and enhance data accuracy, the Phase II approach will deploy a second camera at selected high-impact intersections. By fusing the views from two different camera angles, the system can establish a true spatial relationship of objects in the intersection, essentially achieving a more complete 3D understanding of vehicle and pedestrian trajectories.

Language

• English

Project

• Status: Active
• Funding: $98,080.00
• Contract Numbers:

  TR202621

• Sponsor Organizations:

  Missouri Department of Transportation

  1617 Missouri Boulevard
  P.O. Box 270
  Jefferson City, MO  United States  65102
• Project Managers:

  Schulte, Brent

• Performing Organizations:

  University of Missouri, Columbia

  Department of Civil and Environmental Engineering
  E2509 Lafferre Hall
  Columbia, MO  United States  65203
• Principal Investigators:

  Adu-Gyamfi, Yaw

• Start Date: 20260101
• Expected Completion Date: 20270801
• Actual Completion Date: 0

Subject/Index Terms

• TRT Terms: Artificial intelligence; Data fusion; Intersections; Pedestrian movement; Pedestrian vehicle interface; Traffic surveillance; Vehicle trajectories; Video cameras
• Subject Areas: Data and Information Technology; Operations and Traffic Management; Pedestrians and Bicyclists; Safety and Human Factors;

Filing Info

• Accession Number: 01981047
• Record Type: Research project
• Source Agency: Missouri Department of Transportation
• Contract Numbers: TR202621
• Files: RIP, STATEDOT
• Created Date: Feb 24 2026 3:00PM