Using Security Cameras to Count Transit Passengers

The Institute for Transportation Research and Education at North Carolina State University performed research to determine if transit vehicle security camera footage could be paired with Autoscope software to automatically count transit passenger boardings. The research was limited to community transportation systems vehicles. The Autoscope software allows analog footage to be analyzed using the strategic placement of detectors and functions of detectors to determine where the background image changes. After attempting multiple detector placements in multiple vehicle types with multiple camera configurations, the results show promise that algorithms can be set up in a cost effective manner for counting passenger trips repeatedly on the same vehicle, but less promise that a standard algorithm can be developed that will accurately count passenger trips even on the same vehicle type with similar camera configurations. Because this research stems from a specific need for North Carolina Department of Transportation (NCDOT) to be able to validate passenger trip reporting from transit systems, two sampling plans are suggested as well as expected costs. One involves manual counts using spot-checks of data and the other involves more robust analysis of a single vehicle from each transit system operating deviated fixed route service.

• Record URL:
  https://connect.ncdot.gov/projects/research/Pages/ProjDetails.aspx?ProjectID=2017-16
• Record URL:
  https://rosap.ntl.bts.gov/view/dot/64403
• Corporate Authors:

  North Carolina State University

  Institute for Transportation Research and Education
  Campus Box 8601
  Raleigh, NC  United States  27695-8601

  North Carolina Department of Transportation

  Research and Analysis Group, 1 South Wilmington Street
  Raleigh, NC  United States  27601

  Federal Highway Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Authors:
  • Monast, Kai
  • Vaughan, Christopher
  • Jones, Porter
  • Carnes, Christopher
• Publication Date: 2017-5-31

Language

• English

Media Info

• Media Type: Digital/other
• Edition: Final Report
• Features: Figures; Photos; References; Tables;
• Pagination: 20p

Subject/Index Terms

• TRT Terms: Algorithms; Cameras; Data analysis; Data collection; Detectors; Passenger counting; Software; Transit riders
• Identifier Terms: AUTOSCOPE (Video vehicle detection system)
• Subject Areas: Data and Information Technology; Passenger Transportation; Public Transportation;

Filing Info

• Accession Number: 01687906
• Record Type: Publication
• Report/Paper Numbers: FHWA/NC/2017-16
• Contract Numbers: NCDOT Project 2017-16
• Files: NTL, TRIS, ATRI, USDOT, STATEDOT
• Created Date: Dec 11 2018 9:21AM