Improved Signalized Intersection Performance Using Computer Vision and Artificial Intelligence: Research Project Capsule [24-4SS]

Details

• Creators:
  Moomen, Milhan ; Rupnow, Tyson ; Codjoe, Julius
• Corporate Creators:
  Louisiana Transportation Research Center
• Corporate Contributors:
  Louisiana. Department of Transportation and Development ; Louisiana State University
• Subject/TRT Terms:
  • Signalized intersections
  • Drones
  • Artificial intelligence
  • Computer vision
  • Performance measurement
  • Traffic surveillance
  • Video
• Publication/ Report Number:
  24-4SS
• Resource Type:
  Brief
• Geographical Coverage:
  United States
• Edition:
  Research Project Capsule
• Corporate Publisher:
  Louisiana Transportation Research Center
• Abstract:
  The primary objectives of this research are to: • Assess the feasibility and accuracy of using computer vision technology for performance evaluation at signalized intersections. • Provide intersection video footage data captured by drones. • Use computer vision and artificial intelligence to automatically convert data from video recordings at selected intersections into trajectories of road users. • Use computer vision and artificial intelligence to count road users, and detect queuing and demand for each approach at selected intersections using drone footage. • Develop tools to facilitate DOTD traffic engineers in understanding road users' behaviors, evaluating intersection performance measures, and assisting in determining effective measures for improving safety and efficiency at intersections.
• Format:
  PDF
• Funding:
  TT-Fed/TT-Reg-5
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/73189/dot_73189_DS1.pdf
• File Type:
  [PDF - 643.98 KB]
• Collection(s):
  Artificial Intelligence Deployment Center of Excellence (AI COE) ; US Transportation Collection
• Main Document Checksum:
  urn:sha-512:4450c3285ccde0a2f20c7eba1cfdc3ec9bd994e6a970c624989f2b4ce3b171480b18dd5d8b764f0aec53a70756075ce6fdc92cffc689c7a0ae9c0c1ee3ca4eab