INTERACTIVE SYSTEM OF ANALYZING TRAFFIC EVENT STATISTICS BASED ON OCCLUSION ROBUST VEHICLE TRACKING METHOD

In this paper, we introduce our interactive system for the acquisition and display system of various traffic statistics. One of the most important applications of the Intelligent Transport Systems (ITS) is to analyze various traffic activities and construct traffic monitoring systems. By using a new tracking algorithm which is robust against occlusion and clutter problems, an automated system which is able to acquire traffic statistics and events such as velocities, vehicle counts with respect to origin-destination, precise trace, and so on was constructed. In this system, from the sequential traffic images obtained at crossroads, the regions of vehicles are obtained and stored by using an occlusion robust tracking algorithm. Users can search for statistics by submitting queries, and the statistics are calculated from the tracking results. The calculations are executed according to the contents of the queries. However, because the statistics to be calculated vary, the preparation of the entire associated calculation method is not practical. Therefore, we analyze the calculation flow and decompose hierarchically to arrive at an elementary operation. According to the result of parsing the submitted query, adequate elements are combined and the statistics are calculated. In this way, we avoid the excessive enlargement of the system; because of the decomposition, it becomes receptive to various kinds of queries. In addition, since we employ the natural language query and graphical user interface (GUI), it is easy to access the required statistics.

  • Supplemental Notes:
    • Full Conference Proceedings available on CD-ROM.
  • Corporate Authors:

    ITS America

    1100 17th Street, NW, 12th Floor
    Washington, DC  United States  20036
  • Authors:
    • Nishida, T
    • Matsushita, T
    • Kamijo, Shunsuke
    • Sakauchi, M
  • Conference:
  • Publication Date: 2002

Language

  • English

Media Info

  • Pagination: 12p

Subject/Index Terms

Filing Info

  • Accession Number: 00960179
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jul 2 2003 12:00AM