Advertisement

Median Associative Memories: New Results

  • Humberto Sossa
  • Ricardo Barrón
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3773)

Abstract

Median associative memories (MEDMEMs) first described in [1] have proven to be efficient tools for the reconstruction of patterns corrupted with mixed noise. First formal conditions under which these tools are able to reconstruct patterns either from the fundamental set of patterns and from distorted versions of them were given in [1]. In this paper, new more accurate conditions are provided that assure perfect reconstruction. Numerical and real examples are also given.

Keywords

Associative Memory Noisy Image Pepper Noise Distorted Version Pattern Vector 
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.

References

  1. 1.
    Sossa, H., Barrón, R., Vázquez, R.A.: New Associative Memories to Recall Real-Valued Patterns. In: Sanfeliu, A., Martínez Trinidad, J.F., Carrasco Ochoa, J.A. (eds.) CIARP 2004, vol. 3287, pp. 195–202. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  2. 2.
    Sossa, H., Barrón, R.: Transforming Fundamental Set of Patterns to a Canonical Form to Improve Pattern Recall. In: Lemaître, C., Reyes, C.A., González, J.A. (eds.) IBERAMIA 2004. LNCS, vol. 3315, pp. 687–696. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  3. 3.
    Steinbuch, K.: Die Lernmatrix. Kybernetik 1(1), 26–45 (1961)CrossRefGoogle Scholar
  4. 4.
    Anderson, J.A.: A simple neural network generating an interactive memory. Mathematical Biosciences 14, 197–220 (1972)zbMATHCrossRefGoogle Scholar
  5. 5.
    Hopfield, J.J.: Neural networks and physical systems with emergent collective computational abilities. Proceedings of the National Academy of Sciences 79, 2554–2558 (1982)CrossRefMathSciNetGoogle Scholar
  6. 6.
    Ritter, G.X., et al.: Morphological associative memories. IEEE Transactions on Neural Networks 9, 281–293 (1998)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Humberto Sossa
    • 1
  • Ricardo Barrón
    • 1
  1. 1.Centro de Investigación en Computación-IPNMexico CityMexico

Personalised recommendations