MEASURING TRAVEL TIME USING PATTERN MATCHING TECHNIQUE

In this paper, the authors propose a new method for measuring travel time. Conventional methods require vehicle identification at several passing points by reading the license plate numbers. Consequently, high resolution cameras are needed at each point and it becomes prohibitively expensive. The proposed method, though, needs less expensive sensors, which can recognize large and small vehicles. The proposed method identifies groups of vehicles by features. One of the features is local rate of large vehicles against all vehicles, called feature sequence. The feature sequences have unique patterns. Vehicles are automatically identified at several passing points using a pattern matching technique. An experiment was performed on a trunk road to test the method. Four observation points were established along the road and travel times were measured. The results indicate that the travel times obtained by the proposed method had less than 10 percent errors. Group identification failure due to change of vehicle order by computer simulation was checked. Finally, the authors illustrate by computer simulation that feature sequences usually have unique patterns for the vehicle group identification. The method is applicable to actual traffic by the first experiment and it is robust against the change of vehicle order by the two simulations.

  • Supplemental Notes:
    • Five volumes of papers and one volume of abstracts comprise the published set of conference materials.
  • Corporate Authors:

    VERTIS

    TORANOMOM 34 MORI BUILDING 1-25-5
    TORANOMON, MINATOKU, TOKYO 105  Japan 
  • Authors:
    • Takahashi, K
    • Inoue, T
    • Yokota, T
    • Kobayashi, Y
    • Yamane, K
  • Conference:
  • Publication Date: 1995-11

Language

  • English

Media Info

  • Pagination: p. 78

Subject/Index Terms

Filing Info

  • Accession Number: 00716544
  • Record Type: Publication
  • Report/Paper Numbers: Volume 1
  • Files: TRIS
  • Created Date: Feb 28 1996 12:00AM