Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Gray-scale morphological associative memories.

Peter Sussner1, Marcos Eduardo Valle

  • 1nstitute of Mathematics, Statistics, and Scientific Computation, State University of Campinas, Campinas, CEP13081-970, São Paulo, Brazil. sussner@ime.unicamp.br

IEEE Transactions on Neural Networks
|May 26, 2006
PubMed
Summary

This study analyzes gray-scale morphological associative memories (MAMs) using minimax algebra. Researchers characterized fixed points and basins of attraction, demonstrating MAMs

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Dynamics of structured complex-valued Hopfield neural networks.

Cognitive neurodynamics·2025
Same author

Universal approximation theorem for vector- and hypercomplex-valued neural networks.

Neural networks : the official journal of the International Neural Network Society·2024
Same author

Extreme learning machine for a new hybrid morphological/linear perceptron.

Neural networks : the official journal of the International Neural Network Society·2020
Same author

A broad class of discrete-time hypercomplex-valued Hopfield neural networks.

Neural networks : the official journal of the International Neural Network Society·2019
Same author

Max-plus and min-plus projection autoassociative morphological memories and their compositions for pattern classification.

Neural networks : the official journal of the International Neural Network Society·2018
Same author

On the Dynamics of Hopfield Neural Networks on Unit Quaternions.

IEEE transactions on neural networks and learning systems·2017

Area of Science:

  • Computational Neuroscience
  • Image Processing
  • Algebraic Methods

Background:

  • Traditional neural associative memory models focus on binary or bipolar patterns.
  • Existing research on morphological associative memory systems primarily uses binary models.
  • Key features like optimal storage capacity and one-step convergence in autoassociative morphological memories (AMMs) are known to extend to gray-scale settings.

Purpose of the Study:

  • To analyze gray-scale autoassociative morphological memories (AMMs) extensively using minimax algebra.
  • To provide a comprehensive characterization of fixed points and basins of attraction for gray-scale AMMs.
  • To illustrate the storage capacity and noise tolerance of gray-scale MAMs and introduce a modified model for classification.

Main Methods:

Related Experiment Videos

  • Extensive application of minimax algebra for analyzing gray-scale AMMs.
  • Complete characterization of fixed points and basins of attraction.
  • Computer simulations using gray-scale images to validate mathematical results.

    • Main Results:

    • A detailed description of storage and recall mechanisms in gray-scale AMMs.
    • Mathematical insights into storage capacity and noise tolerance of gray-scale MAMs.
    • Introduction of a modified AMM yielding a fixed point closest to the input via Chebyshev distance.

    Conclusions:

    • Minimax algebra provides a robust framework for understanding gray-scale AMMs.
    • Gray-scale AMMs exhibit significant storage capacity and noise tolerance.
    • Modified gray-scale AMMs can be effectively utilized for pattern classification.