SMD position measurement by a Kohonen network compared with image processing

  • Robert P. W. Duin
  • Edwin Th. G. Hoek
Part of the Lecture Notes in Computer Science book series (LNCS, volume 970)


Two techniques for determining the exact position of a surface mounted device in a robot's hand are compared: a Kohonen network and image processing. The algorithms are developed on a sequential machine. Computing times for special purpose parallel hardware are estimated, indicating that the two techniques are comparable.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    R.C. Gonzalez and R.E Woods., Digital image processing, Addison-Wesley, Reading, 1992.Google Scholar
  2. [2]
    R. Hecht-Nielsen, Neurocomputing, Addison-Wesley, Reading, 1990.Google Scholar
  3. [3]
    E.T.G. Hoek, “Fast and Accurate SMD Locating using a Kohonen Network”, Master Thesis, Delft University of Technology, 1994.Google Scholar
  4. [4]
    T. Kohonen, “Self-organized formation of topologically correct feature maps”, Biol. Cyber., 43, 59–69, 1982.Google Scholar
  5. [5]
    W.H. Press et al., Numerical Recipes: the art of scientific computing, Cambridge, 1986.Google Scholar
  6. [6]
    P.P. Jonker, Morphological Image Processing: Architecture and VLSI Design (Ph.D. thesis), Kluwer, Technische Boeken BV, Deventer, 1992.Google Scholar
  7. [7]
    R.P.W. Duin, P.P. Jonker, Processor arrays compared to pipelines for cellular image processing: S. Levialdi (eds.), Multicomputer Vision, Academic Press, London, 1988, 151–169.Google Scholar
  8. [8]
    R.P.W. Duin and E.R. Komen, Massively parallel architectures for cellular logic image processing, in: V. Cantoni, L.P. Cordella, S. Levialdi, G. Sanniti di Baja (eds.), Progress in Image Analysis and Processing, World Scientific, Singapore, 1990, 643–657.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Robert P. W. Duin
    • 1
  • Edwin Th. G. Hoek
    • 1
  1. 1.Pattern Recognition Group Faculty of Applied PhysicsDelft University of TechnologyThe Netherlands

Personalised recommendations