COUNTING CARS BY TRACKING MOVING OBJECTS IN THE OUTDOOR PARKING LOT

Image processing systems have been utilized in congested areas to alert drivers to parking availability. For outdoor parking facilities, though, image processing procedures can be affected negatively by noise, shadows, and changes in brightness. The authors propose a method to count cars by using the locus of moving objects with sequential images. A robust tracking process, a judging process, and a counting process of parked cars are combined in this investigation. First, while tracking some regions obtained in time differential images, the extraction of a locus of an object by clustering them along time sequences is executed. Second, the locus is judged as to whether it is made by a car. Finally, the number of parked cars is counted depending on the integration to increase or to decrease the number by the start and the end points of the locus. Results of the experiment are positive. The advantages and disadvantages of the system were analyzed for various conditions, in terms of weather and time of day. Accuracy is improved by using a method to recognize the conditions and adjusting the parameter according to the conditions.

  • 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:
    • Hasegawa, T
    • Nohsoh, K
    • Ozawa, S
  • Conference:
  • Publication Date: 1995-11

Language

  • English

Media Info

  • Pagination: p. 230

Subject/Index Terms

Filing Info

  • Accession Number: 00721079
  • Record Type: Publication
  • Report/Paper Numbers: Volume 1
  • Files: TRIS
  • Created Date: May 20 1996 12:00AM