A Multi-purpose Visual Classification System (original) (raw)

Abstract

A computer vision system which can be trained to classification tasks from sample views is presented. It consists of several artificial neural networks which realize local PCA with subsequent expert nets as classifiers. The major benefit of the approach is that entirely different tasks can be solved with one and the same system without modifications or extensive parameter tuning. Therefore, the architecture is an example for the potential which lies in view based recognition: Making complicated tasks solvable with less and less expert knowledge.

Preview

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. S.C. Ahalt, A.K. Krisnamurthy, P. Chen, and D.E. Melton. Competitive learning algorithms for vector quantizaion. Neural Networks, 3:277–290, 1990
    Article Google Scholar
  2. E. Braun, G. Heidemann, H. Ritter, and G. Sagerer. A multi-directional multiple pathe recognition scheme for complex objects. Pattern Recognition Letters, special issue on Pattern Recognition in Practice VI, Vlieland, June 99, 1999
    Google Scholar
  3. Aristides S. Galanopoulos and Stanley C. Ahalt. Codeward distribution for frequency sensitive competitive learning with one-dimensional input data. IEEE Trans. On Neural Networks, 7(3):752–756, 1996.
    Article Google Scholar
  4. Aristides S. Galanopoulos, Randolph L. Moses, and Stanley C. Ahalt. Diffusion approximation of frequency sensitive competitive learning. IEEE Trans. On Neural Networks, 8(5):1026–1030, 1997.
    Article Google Scholar
  5. S. Grossberg. Competitive learning: From interactive activation to adaptive resonance. Cognitive Sci., 11:23–63, 1987.
    Article Google Scholar
  6. G. Heidemann. Ein flexible einsetzbares Objecterkennungssystem auf der Basis neuronaler Netze. PhD thesis, Univ. Bielefeld, Techn. Fak., 1998. Infix, DISKI 190.
    Google Scholar
  7. G. Heidemann, D. Lücke, and H. Ritter. A system for various visual classification tasks based on neural networks. In A. Sanfeliu et al., editor, Proc. 15th Int. Conf. on Pattern Recognition ICPR 2000, Barcelona, volume I, pages 9–12, 2000.
    Google Scholar
  8. G. Heidemann, and H. Ritter. Combining multiple neural nets for visual feature selection and classification. In ICANN 99, Ninth International Conference on Artificial Neural Networks, pages 365–370, 1999.
    Google Scholar
  9. G. Heidemann and H. Ritter. Visual checking of grasping positions of a three-fingered robor hand. In ICANN 2001, 2001. Accepted for publication.
    Google Scholar
  10. G. Heidemann and H.J. Ritter. Efficient Vector Quantization using the WTA-rule with Activity Equalization. Neural Processing Letters, 13(1): 17–30, 2001.
    Article MATH Google Scholar
  11. T. Kohonen. Self-organized formation of topologically correct feature maps. Biol. Cybernetics, 43:59–69, 1982.
    Article MATH MathSciNet Google Scholar
  12. T. Kohonen. Self-organizationa and associative memory. In Springer Series in Information Sciences 8. Springer-Verlag Heidelberg, 1995.
    Google Scholar
  13. T. Kohonen. Self-Organizing Maps. Springer Verglag, 1995.
    Google Scholar
  14. T. Martinez, S.G. Berkovich, and K. Schulten. “Neural-gas” network for vector quantization and its application to time-series prediction. IEEE Transactions on Neural Networks, 4(4):558–569, 1993.
    Article Google Scholar
  15. J. Moody and C. Darken. Learning with localized receptive fields. In Proc. of the 1988 Connectionist Models Summer School, pages 133–143. Morgan Kaufman Publishers, San Mateo, CA, 1988.
    Google Scholar
  16. H. Murase and S.K. Nayar. Visual learning and recognition of 3-d objects from appearance. Int. Journal of Computer Vision, 14:5–24, 1995.
    Article Google Scholar
  17. S.A. Nen, S.K. Nayar, and H. Murase. Columbia object image library: Coil-100. Technical Report CUCS-006-96, Department of Computer Science, Columbia University, 1996.
    Google Scholar
  18. T. Poggio and S. Edelman. A network that learns to recognize three dimensional objects. Nature, pages 263–266, 1990.
    Google Scholar
  19. H.J. Ritter, T.M. Martinez, and K.J. Schulten. Neuronale Netze. Addision-Wesley, München, 1992.
    MATH Google Scholar
  20. T.D. Sanger. Optimal unsupervised learning in a single-layer linear feedforward neural network. Neural Networks, 2:459–473, 1989.
    Article Google Scholar
  21. Dirk Strotmann. Lokale Klassifikation als Basis für die Erkennung teilverdeckter Objecte. Master’s thesis, Universität Bielefeld, Technische Fakultät, 1999.
    Google Scholar
  22. Michael E. Tipping and Christopher M. Bishop. Mixtures of probabilistic principal component analyzers. Neural Computation, 11(2):443–482, February 15 1999.
    Article Google Scholar

Download references

Author information

Authors and Affiliations

  1. Universität Bielefeld, AG Neuroinformatik, Germany
    Gunther Heidelmann

Authors

  1. Gunther Heidelmann

Editor information

Editors and Affiliations

  1. Computer Science I, University of Dortmund, 44221, Dortmund, Germany
    Bernd Reusch

Rights and permissions

© 2001 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Heidelmann, G. (2001). A Multi-purpose Visual Classification System. In: Reusch, B. (eds) Computational Intelligence. Theory and Applications. Fuzzy Days 2001. Lecture Notes in Computer Science, vol 2206. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45493-4\_34

Download citation

Publish with us