Probabilistic Decisions in Production Net An Example from Vehicle Recognition

  • Eckart Michaelsen
  • Uwe Stilla
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2396)


A structural knowledge-based vehicle recognition method is modified yielding a new probabilistic foundation for the decisions. The method uses a pre-calculated set of hidden line projected views of articulated polyhedral models of the vehicles. Model view structures are set into correspondence with structures composed from edge lines in the image. The correspondence space is searched utilizing a 4D Hough-type accumulator. Probabilistic models of the background and the error in the measurements of the image structures lead to likelihood estimations that are used for the decision. The likelihood is propagated along the structure of the articulated model. The system is tested on a cluttered outdoor scene. To ensure any-time performance the recognition process is implemented in a data-driven production system.


Search Area Probabilistic Decision Line Object Generalize Hough Transformation Correspondence Space 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Ballard D. H., Brown C. M.: Computer Vision. Prentice Hall, Englewood Cliffs, New Jersey, (1982).Google Scholar
  2. 2.
    Binfort T. O., Levitt T. S.: Model-based Recognition of Objects in Complex Scenes. In: ARPA (ed.). Image Understanding Workshop 1994. Morgan Kaufman, San Francisco (1994), 149–155.Google Scholar
  3. 3.
    Dickinson S. J., Pentland A. P., Rosenfeld A.: From Volumes to Views: An Approach to 3-D Object Recognition. CVGIP:IU, Vol. 55, No. 2 (1992), 130–154.zbMATHCrossRefGoogle Scholar
  4. 4.
    Eggert D. W., Bowyer K. W., Dyer C.R.: Aspect Graphs: State-of-the-Art and Applications in Digital Photogrammetry. ISPRS-XXiX, Vol. 5, Com V, (1992) 633–645.Google Scholar
  5. 5.
    Grimson W. E. L.: Object Recognition by Computer: The Role of Geometric Constraints. MIT Press, Cambridge, Mass., (1990).Google Scholar
  6. 6.
    Hermitson K. J., Booth D. M., Foulkes S. B., Reno A. L.: Pose Estimation and Recognition of Ground Vehicles in Aerial Reconnaissance Imagery. ICPR 1998, Vol. 1, IEEE, Los Alamitos, California, (1998), 578–582.Google Scholar
  7. 7.
    Hoogs A., Mundy J.:An Integrated Boundary and Region Approach to Perceptual Grouping. ICPR 2000, Vol. 1, IEEE, Los Alamitos, California, (2000), 284–290.CrossRefGoogle Scholar
  8. 8.
    Michaelsen E., Stilla U.: Ansichtenbasierte Erkennung von Fahrzeugen. In: Sommer G., Krüger N., Perwas C. (eds.): Mustererkennung 2000, Springer, Berlin, (2000) 245–252.Google Scholar
  9. 9.
    Michaelsen E., Stilla U.: Assessing the Computational Effort for Structural 3D Vehicle Recognition. In: Ferri F.J., Inesta J.M., Amin A., Pudil P. (eds): Advances in Pattern Recognition (SSPR-SPR 2000), Springer, Berlin, (2000) 357–366.CrossRefGoogle Scholar
  10. 10.
    Tan T., Sullivan G., Baker K.: Model-Based Localisation and Recognition of Road Vehicles.. Int. Journ. of Comp. Vision, Vol. 27, (1998) 5–25.CrossRefGoogle Scholar
  11. 11.
    Wang P. S. P.: Parallel Matching of 3D Articulated Object Reccognition. Int. Journ. of Pattern Recognition and Artificial Intelligence, Vol. 13, (1999) 431–444.CrossRefGoogle Scholar
  12. 12.
    Viola P., Wells W. M. III: Alignment by Maximalization of Mutual Information. Int. Journ. of Comp. Vision, Vol. 24, (1997) 137–154.CrossRefGoogle Scholar
  13. 13.
    Wells W. M. III: Statistical Approaches to Feature-Based Object Recognition. IJCV, 21, (1997) 63–98.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Eckart Michaelsen
    • 1
  • Uwe Stilla
    • 1
  1. 1.FGAN-FOM Research Institute for Optronics and Pattern RecognitionEttlingenGermany

Personalised recommendations