Evaluation of Thermal Image Video Sensors for Stop Bar Detection at Signalized Intersections

Thermal image sensors for stop bar presence detection have recently been introduced to the traffic industry as an alternative to cameras sensitive only to the visual spectrum. This new detection technology, from two manufacturers, was evaluated side by side with a video detection system. Inductive loops were used for a comparison to identify discrepancies that warranted the manual establishment of ground truth of the video images. The video camera and two thermal sensors operated simultaneously over a 24-h period, and discrepancies in the loop versus thermal or video detection calls were validated from recorded video. No missed call events longer than 10 s were observed, and only a modest number of false calls were made by the test systems. The bias in activation and termination times was also evaluated during day and night operation for each system. The study found that the median time difference in activating a detection zone was about 1 s when the day and night operations of the detection system were compared with a video camera. This finding is consistent with past studies that reported nighttime detection challenges related to headlight projections. However, the thermal cameras had virtually no change in median activation times when day and nighttime operations were compared. This encouraging finding suggests that integrating cameras sensitive to the infrared spectrum holds considerable promise for improving the quality of nighttime video detection.

• Record URL:
  https://doi.org/10.3141/2308-20
• Summary URL:
  http://scholar.google.com/scholar_lookup?title=Evaluation+of+Thermal+Image+Video+Sensors+for+Stop+Bar+Detection+at+Signalized+Intersections&author=J.+Grossman&author=A.+Hainen&author=S.+Remias&author=D.+Bullock&publication_year=2012
• Availability:
  • Find a library where document is available. Order URL: http//www.trb.org/Main/Blurbs/168553.aspx
• Authors:
  • Grossman, Jay
  • Hainen, Alexander M
  • Remias, Stephen M
  • Bullock, Darcy M
• Publication Date: 2012

Language

• English

Media Info

• Media Type: Print
• Features: Figures; Photos; References; Tables;
• Pagination: pp 184–198
• Monograph Title: Urban and Traffic Data Systems 2012
• Serial:
  • Transportation Research Record: Journal of the Transportation Research Board
  • Issue Number: 2308
  • Publisher: Transportation Research Board
  • ISSN: 0361-1981

Subject/Index Terms

• TRT Terms: Automatic vehicle detection and identification systems; Infrared detectors; Night; Sensors; Signalized intersections; Thermal imagery; Video imaging detectors
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; I73: Traffic Control;

Filing Info

• Accession Number: 01366383
• Record Type: Publication
• ISBN: 9780309263061
• Report/Paper Numbers: 12-0951
• Files: TRIS, TRB, ATRI
• Created Date: Mar 29 2012 7:14AM