ICANN ’93 pp 978-981 | Cite as

Invariant Pattern Recognition with Recovery of Transformation Parameters

  • K. Jeschke
  • J. Reinhardt
  • J. A. Maruhn
Conference paper


In this paper a method is described for using a multilayered perceptron for recognizing a rotation and translation invariant pattern and for recovering the transformation parameters. This is achieved by employing an invariant preprocessing stage. By using the knowledge of the pattern class and a second preprocessed input the original rotation angle is determined. Simulation results are presented and discussed.


Multilayered Perceptron Zernike Moment Pattern Class Invariant Input Invariant Pattern 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    D.E. Rumelhard, G.E. Hinton, R.J. Williams, ”Learning Internal Representations by Error Propagation” PDP vol 1, Cambridge, MA: MIT Press, pp 318–362, 1986Google Scholar
  2. [2]
    G. Elliman R.N. Banks, ”Shift Invariant Neural Net for Machine Vision”, Proc. IEEE, vol 137, pt 1, pp 183–187, 1990Google Scholar
  3. [3]
    A.P. Reeves, R.J. Prokop, S.E. Andrews, F.P. Kuhl, ”3D Shape Analysis Using Moment and Fourier Descriptors”, IEEE Trans. Pattern Anal. Mach. Intell., vol 12, pp 487–497, 1990Google Scholar
  4. [4]
    A. Khotanzad, J.H. Lu, ”Distortion Invariant Character Recognition by a MLP and Back-Propagation Learning”, Proc 2nd Intl. Conf. on NN vol I, pp 625–632, 1988.Google Scholar
  5. [5]
    M.K. Hu, ”Visual Pattern Recognition by Moment Invariants” IRE Trans. Inform. Theory, vol IT-8, pp 1028–1038, 1990Google Scholar
  6. [6]
    A. Khotanzad, Y. H. Hong, ”Invariant Image Recognition by Zernike Moments”, IEEE Trans. Pattern Anal. Mach. Intell., vol 12, pp 489–497, 1990CrossRefGoogle Scholar
  7. [7]
    M.B. Reid, L. Spirkovska, E. Ochoa, ”Rapid Training of Higher Order Neural Networks for Invariant Pattern Recognition”, Conf. Proc. IJCNN 1989, vol I, pp 689–692Google Scholar
  8. [8]
    P. Zimmerer and A. Zell, ”Translationsinvariante Erkennung von Werkstücken mit Neuronalen Netzwerken”, Informatik Fachberichte 290, pp 51–58, 1991CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London Limited 1993

Authors and Affiliations

  • K. Jeschke
    • 1
  • J. Reinhardt
    • 1
  • J. A. Maruhn
    • 1
  1. 1.Institut für Theoretische PhysikUniversität FrankfurtGermany

Personalised recommendations