Monitoring of Urban Roadway Safety Hazards from Existing Bus-based Video Imagery

Contributors to traffic safety hazards include lane changes in the presence of queues at signalized intersections and bus stops, mixes of vehicle classes, speed variability, and vehicle overtaking. Assessing locations with increased recurring traffic safety hazards requires extensive and ongoing collection of data on these contributors. Traditional methods entail using sensors at permanent or temporary fixed locations, which are costly, labor intensive, provide limited coverage over time and space, and only provide information on some contributors. Moreover, the location of these sensors may be influenced by factors other than optimal sampling, such as requests from well-organized constituencies. Therefore, relying on such data to assess traffic safety hazards to support improved policies and designs could lead to missing high-risk conditions resulting in inequitable outcomes. Clearly, low cost and extensive, equitable data collection is desirable. Transit buses operate regularly over wide networks. Most agencies have equipped their bus fleet with cameras that record the environment inside and outside buses for liability, security, and safety purposes. Consequently, the imagery is available for other uses at near-zero marginal cost and the coverage provides comprehensive views that could potentially be used to determine times and locations of regularly occurring safety hazards. Moreover, this imagery has been shown by the principal investigators (PIs) to be effective in monitoring traffic volumes across time and space, information that provides exposure-based context for identifying safety hazards. In this project the PIs would investigate if the available, repeated, and extensive imagery recorded by cameras mounted on transit buses in regular operation could be used to identify “hazardous hotspots”. The PIs have been obtaining transit bus-based video imagery to estimate traffic flows across the OSU campus and provide summary results to campus planners and operators on a regular basis. The OSU campus will again be used as a living lab testbed. The size and diversity of land uses make the campus representative of urban areas. Moreover, the campus is undergoing major construction activities, which would allow investigation of different infrastructure conditions that could influence traffic safety. The use of the campus as an experimental testbed would also allow for in-situ ground-truth observations to assess the accuracy of the video-based results. The hazards to be considered in the proposed one-year project (phase 1) include lane specific queue lengths at intersections and bus stops, frequency of lane-changing in the presence of queues, and vehicle type mix with an emphasis on vulnerable vehicles (e.g., bicycles, scooters, and motorcycles). Speeds and their variation where autos conflict with vulnerable vehicles are also important safety factors. In phase 1, exploring the ability to measure speeds from the imagery will begin. In addition, imagery of traffic conditions at construction zones will be collected for future investigations of speed changes and other safety related hazards. The identification of hazards will be demonstrated using semi-automatic techniques based on an extension of a Graphical User Interface used in a previous project focused on traffic volumes. Moreover, steps toward automation for large scale analysis will be taken during phase 1.

• Record URL:
  https://ppms.cit.cmu.edu/projects/detail/444

Language

• English

Project

• Status: Active
• Funding: $190782
• Contract Numbers:

  69A3552344811

• Sponsor Organizations:

  Carnegie Mellon University

  Safety21 National UTC for Promoting Safety
  Pittsburgh, PA  United States  15213

  Office of the Assistant Secretary for Research and Technology

  University Transportation Center Program
  ,    
• Managing Organizations:

  Carnegie Mellon University

  Safety21 National UTC for Promoting Safety
  Pittsburgh, PA  United States  15213
• Project Managers:

  Stearns, Amy

• Performing Organizations:

  The Ohio State University

  ,    
• Principal Investigators:

  Mishalani, Rabi

• Start Date: 20230701
• Expected Completion Date: 20240630
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers

Subject/Index Terms

• TRT Terms: Buses; Data collection; Image analysis; Lane changing; Traffic queuing; Traffic safety; Vehicle mix; Video
• Subject Areas: Data and Information Technology; Highways; Safety and Human Factors;

Filing Info

• Accession Number: 01900375
• Record Type: Research project
• Source Agency: Safety21
• Contract Numbers: 69A3552344811
• Files: UTC, RIP
• Created Date: Nov 21 2023 8:40PM