INTENSITY AND EDGE-BASED SYMMETRY DETECTION APPLIED TO CAR-FOLLOWING

Two methods are presented for detecting symmetry in images. One is based directly on the intensity values and another one based on a discrete representation of local orientation. A symmetry finder is presented based on a normalized measure for the degree of intensity symmetry within scan line intervals. A symmetry-enhancing edge detector is presented for detecting symmetry which is formed by a relationship between local orientations. The application for which the symmetry methods have been developed is vision-based car- following. Real-time experiments with a system for automatic headway control on motorways have been successful.

• Supplemental Notes:
  • Publication Date: 1992 Published By: Springer Verlag, Berlin
• Corporate Authors:

  Universitat Karlsruhe. Institut fur Algorithmen und Kognitive Systeme

  ,    

  Universite de Rennes. IRISA

  ,    

  Prometheus (Program)

  ,    

  Universita di Genova. Dept. of Communication, Computer and Systems Science

  ,    

  University of Surrey

  Department of Electronics and Electrical Engineering
  Surrey,   United Kingdom 

  Ruhr-Universitat Bochum. Institut fur Neuroinformatik

  ,    

  Queen Mary and Westfield College

  ,    
• Authors:
  • Zielke, Thomas
  • Brauckmann, Michael
  • Seelen, W von
• Conference:
  • European Conference on Computer Vision (2nd : 1992 : Santa Margherita Ligure, Italy). Computer vision--ECCV '92
• Publication Date: 1992

Language

• English

Media Info

• Pagination: p. 865-873

Subject/Index Terms

• TRT Terms: Car following; Computer vision
• Subject Areas: Operations and Traffic Management;

Filing Info

• Accession Number: 00772760
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: PATH
• Created Date: Nov 17 1999 12:00AM