Development of Dual-Station Automated Expressway Incident Detection Algorithms

This paper presents a new dual-station algorithm for applications in incident. While traditional algorithms for this purpose implement loop-based detection, the combined detector evaluation (CODE) algorithm and the CODE flow-based algorithm presented here incorporate video-based detection acquired by wayside cameras. Two types of technologically-enhanced incident detection are: 1) the non-automated technique of pan-tilt-zoom (PTZ) surveillance, and 2) automated incident detection algorithms (AIDA) that incorporate video image processing or vehicle-to-roadside communication. CODE, as part of AIDA, enhances earlier systems through efficiency and effectiveness of prediction. Testing of CODE found that it detected incidents approximately 200 seconds faster than existing means on the roadway in question. One constraint of the system as presented is that it is not calibrated to detect incidents in low-traffic situations.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/41297384
• Authors:
  • Mak, Chin Long
  • Fan, Henry S L
• Publication Date: 2007-9

Language

• English

Media Info

• Media Type: Print
• Features: Figures; Photos; References; Tables;
• Pagination: pp 480-490
• Serial:
  • IEEE Transactions on Intelligent Transportation Systems
  • Volume: 8
  • Issue Number: 3
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 1524-9050
  • Serial URL: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6979

Subject/Index Terms

• TRT Terms: Automation; Computer algorithms; Highway traffic control; Incident detection; Intelligent transportation systems
• Subject Areas: Highways; Operations and Traffic Management; I73: Traffic Control;

Filing Info

• Accession Number: 01080258
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: BTRIS, TRIS
• Created Date: Oct 31 2007 6:38AM