ROAD TRAFFIC MONITORING USING THE TRIP 2 SYSTEM. SECOND INTERNATIONAL CONFERENCE ON ROAD TRAFFIC MONITORING

This paper contains a brief review of present image processing systems used for traffic monitoring, including a discussion of the disadvantages of such systems. The trip 2 system is described in more detail. This differs from many earlier systems as it is based upon an inexpensive IBM at microcomputer and can process complex pattern information by simulating the operation of a neural network. The vehicle detection algorithm relies upon the ability of the trip 2 system to learn from example and discriminate between complex patterns within video images of traffic scenes. During site trials of the system it was possible to detect 99% of the vehicles and individual vehicle speeds to an accuracy of between plus or minus 8% and 17% respectively. The algorithms were capable of: (a) removing unwanted shadows from images; and (b) operating over periods during which light levels varied significantly. Research is currently being undertaken to investigate the application of such neural network pattern recognition techniques for traffic monitoring, (traffic counting, traffic flow monitoring and vehicle classification). (TRRL)

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/isbn/0852963734
• Corporate Authors:

  Institution of Electrical Engineers

  Savoy Place
  London WC2R 0BL, NY  England  10016-5997
• Authors:
  • Dickinson, K W
  • Wan, C L
• Publication Date: 1989

Media Info

• Features: References;
• Pagination: p. 56-60
• Serial:
  • Publication of: INSTITUTION OF ELECTRICAL ENGINEERS
  • Publisher: Institution of Electrical Engineers
  • ISSN: 0537-9989

Subject/Index Terms

• TRT Terms: Automatic control; Automation; Classification; Computer programs; Conferences; Errors; Incident detection; Recording; Speed; Television; Traffic congestion; Traffic counts; Traffic flow; Traffic surveys; Vehicles
• ITRD Terms: 3882: Automatic; 8513: Classification; 8525: Conference; 632: Congestion (traffic); 6440: Error; 1632: Incident detection; 6969: Recording; 8645: Software; 5408: Speed; 6949: Television; 689: Traffic count; 671: Traffic flow; 676: Traffic survey; 1255: Vehicle
• Subject Areas: Highways; Operations and Traffic Management; Vehicles and Equipment;

Filing Info

• Accession Number: 00493051
• Record Type: Publication
• Source Agency: Transport Research Laboratory
• ISBN: 0-85296373-4
• Files: ITRD, TRIS
• Created Date: May 31 1990 12:00AM