Optimization of Signal Timing Based on Pedestrian Intervals Using Computer Vision and Deep Learning Technology

This project is to conduct extensive use of computer vision based deep learning methodologies for detection, tracking and prediction of pedestrian movements. This will in turn aid in achieving the following objectives: (1) Increase pedestrian compliance with crossing phases by (a) Reduced wait times with skipped phases (b) Reduced false calls (2) Increase signal efficiency by (a) Reduced cycle length and split times to accommodate pedestrians (b) Reduced false calls (3) Identify a methodology and display to provide feedback to pedestrians that a call has been placed and the approximate wait time.

Language

• English

Project

• Status: Completed
• Contract Numbers:

  BED26 977-03

• Sponsor Organizations:

  Florida Department of Transportation

  Research Center
  605 Suwannee Street MS-30
  Tallahassee, FL  United States  32399-0450
• Managing Organizations:

  Florida Department of Transportation

  Research Center
  605 Suwannee Street MS-30
  Tallahassee, FL  United States  32399-0450
• Project Managers:

  Dilmore, Jeremy

• Performing Organizations:

  University of Central Florida, Orlando

  12443 Research Parkway, Suite 207
  Orlando, FL  United States  32826-
• Principal Investigators:

  Abdel-Aty, Mohamed

• Start Date: 20230103
• Expected Completion Date: 20240630
• Actual Completion Date: 20240718

Subject/Index Terms

• TRT Terms: Compliance; Computer vision; Identification systems; Optimization; Pedestrian movement; Traffic signal timing
• Subject Areas: Highways; Operations and Traffic Management; Pedestrians and Bicyclists; Safety and Human Factors;

Filing Info

• Accession Number: 01869590
• Record Type: Research project
• Source Agency: Florida Department of Transportation
• Contract Numbers: BED26 977-03
• Files: RIP, STATEDOT
• Created Date: Jan 4 2023 7:28AM