MACHINE VISION AND ROAD RESEARCH: NEW TASKS, OLD PROBLEMS

This paper covers the application of techniques derived from machine vision and from computer-based image analysis to several well known problem areas in road research, viz, vehicle classification and identification, and road surface rating. The paper commences by surveying and discussing recent work relevant to the application of computer-based techniques in these areas, and then presents an overview of some of the kinds of computational requirements and loads typically associated with machine processing in these domains. Several of the advanced chip-level devices which are expected to enable the economical realisation of devices for the broad range of applications in question within the next two to five years are described. The program of investigation and research towards which the paper is directed has as its aim the development of algorithmic procedures. These will make it possible to exploit the rapidly declining costs of image acquisition and processing hardware, and permit developed and tested applications software and field tests to have been completed in a timely manner. In specific terms it is proposed to apply digital image acquisition and processing techniques to automatic monitoring of traffic flow and to vehicle classification and identification, and to assess the potential for application to road condition ratings. The number of the covering abstract for the conference is TRIS No. 393385. (Author/TRRL)

  • Record URL:
  • Availability:
  • Supplemental Notes:
    • This paper was presented during the 12th Australian Road Research Board Conference, Hobart, Tasmania, 27-31 August 1984.
  • Corporate Authors:

    ARRB

    Melbourne, Victoria  Australia 
  • Authors:
    • WIGAN, M R
    • CULLINAN, M
  • Publication Date: 1984

Media Info

  • Features: Photos; References;
  • Pagination: p. 76-86
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00393446
  • Record Type: Publication
  • Source Agency: ARRB
  • Files: ITRD, TRIS, ATRI
  • Created Date: Jun 30 1985 12:00AM