ASGCS: A new self-organizing network for automatic selection of feature variables

  • J. Ruiz-del-Solar
  • D. Kottow
Engeneering Applications
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1607)


The automatic selection of invariant feature variables is very important in pattern recognition systems. Recently, neural models have begun to be employed for this task. Among other models the ASSOM stands out because of its simplicity and biological plausibility. However, the main drawback of the application of the ASSOM in image processing systems is that a priori information is necessary to choose a suitable network size and topology in advance. The main purpose of this article is to present the Adaptive-Subspace Growing Cell Structures (ASGCS) network, which corresponds to a further improvement of the ASSOM that overcomes its main drawbacks. The ASGCS network introduces some GCS (Growing Cell Structures) concepts into the ASSOM model. The ASGCS network is described and some examples of automatic Gabor-like feature filter generation are given.


Adaptive-Subspace Self-Organizing Map (ASSOM) Growing Cell Structures (GCS) Adaptive-Subspace Growing Cell Structures (ASGCS) Gabor-Filters Automatic features extraction 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Daugman, J.G. (1980). Two-dimensional spectral analysis of cortical receptive field profiles. Vision Research, 20, 847–856.CrossRefGoogle Scholar
  2. Fritzke, B. (1994). Growing Cell Structures—A self-organizing network for unsupervised and supervised learning. Neural Networks, Vol. 7, No. 9, 1441–1460.CrossRefGoogle Scholar
  3. Kohonen, T. (1995a). Self-Organizing Maps. Springer-Verlag, Heidelberg.Google Scholar
  4. Kohonen, T. (1995b). The Adaptive-Subspace SOM (ASSOM) and its use for the implementation of invariant feature detection. Proc. of the Int. Conf. on Artificial Neural Networks—ICANN 95, October 9–13, Paris, France.Google Scholar
  5. Kohonen, T. (1996). Emergence of invariant-feature detectors in the adaptive-subspace self-organizing map. Biol. Cybern., Vol. 75, No. 4, 281–291.MATHCrossRefGoogle Scholar
  6. Ruiz-del-Solar, J., and Köppen, M. (1996). Automatic generation of Oriented Filters for Texture Segmentation. Proc. of the Int. Workshop on Neural Networks for Identification, Control, Robotics & Signal/Image Processing—NICROSP 96, August 21–23, Venice, Italy.Google Scholar
  7. Ruiz-del-Solar, J., and Köppen, M. (1997). A Texture Segmentation Architecture based on automatically generated Oriented Filters. Journal of Microelectronic Systems Integration, Vol. 5, No. 1, 43–52.Google Scholar
  8. Ruiz-del-Solar, J. (1998). TEXSOM: Texture segmentation using self-organizing maps. Neurocomputing 21, 7–18.CrossRefGoogle Scholar
  9. Sanger, T.D. (1989). Optimal unsupervised learning in a single-layer linear feed-forward neural network. Neural Networks, Vol. 2, No. 6, 459–473.CrossRefGoogle Scholar
  10. Sirosh, J. (1995). A Self-Organizing neural network model of the primary visual cortex, Ph.D. Thesis, The University of Texas at Austin, USA.Google Scholar
  11. Van Sluyters, R.C., Atkinson, J., Banks, M.S., Held, R.M., Hoffmann, K.-P., and Shatz, C.J. (1990). The Development of Vision and Visual Perception. In L. Spillman and J. Werner (Eds.), Visual Perception: The Neurophysiological Foundations, Academic Press.Google Scholar
  12. Wilson, H.R., Levi, D., Maffei, L., Rovamo, J., and DeValois, R. (1990). The Perception of form: Retina to Striate Cortex. In L. Spillman and J. Werner (Eds.), Visual Perception: The Neurophysiological Foundations, Academic Press.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • J. Ruiz-del-Solar
    • 1
  • D. Kottow
    • 1
  1. 1.Department of Electrical EngineeringUniversidad de ChileSantiagoChile

Personalised recommendations