AN IMAGE-PROCESSING SYSTEM FOR PEDESTRIAN DATA COLLECTION

A project is described which investigated the use of software techniques for the collection of information on pedestrian movements automatically using image processing technology. Two algorithms were developed: a) to measure pedestrian density within a crowd scene and b) to count and determine the walking direction of individual pedestrians. The mechanism of each technique is outlined. Real-time performance was achieved for the density measurements using a small network of 3 transputers operating in parallel. The automatic counting of individual pedestrians was demonstrated successfully but has not been optimised to work in real time. For the covering abstract see IRRD 865401.

Language

  • English

Media Info

  • Features: References;
  • Pagination: p. 123-6

Subject/Index Terms

Filing Info

  • Accession Number: 00668590
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • Files: ITRD, ATRI
  • Created Date: Nov 16 1994 12:00AM