Jove
Visualize
Contact Us

Related Experiment Videos

Morphological associative memories.

G X Ritter1, P Sussner, J L Diza-de-Leon

  • 1University of Florida, Center for Computer Vision and Visualization, Gainesville, FL 32611, USA.

IEEE Transactions on Neural Networks
|February 7, 2008
PubMed
Summary
This summary is machine-generated.

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

Rank-based decompositions of morphological templates.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
Same author

Lattice algebra approach to single-neuron computation.

IEEE transactions on neural networks·2008
Same author

Morphological bidirectional associative memories.

Neural networks : the official journal of the International Neural Network Society·2003
Same author

A new parallel binary image shrinking algorithm.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·1995
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles
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

This study introduces morphological neural networks, replacing traditional multiplication and addition with addition and maximum/minimum operations. These novel networks exhibit distinct properties, particularly in associative memory applications.

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Pattern Recognition

Background:

  • Traditional artificial neural networks rely on linear operations (multiplication and addition) followed by nonlinear activation functions.
  • These linear operations are fundamental to computing neuron states and network layer computations.
  • Existing models like the Hopfield net represent semilinear approaches.

Purpose of the Study:

  • Introduce a novel class of artificial neural networks: morphological neural networks.
  • Explore the computational and storage capabilities of morphological associative memories.
  • Differentiate morphological neural networks from traditional semilinear models.

Main Methods:

  • Replaced multiplication and addition operations with addition and maximum (or minimum) operations, respectively.

Related Experiment Videos

  • Developed morphological neural networks where computation is inherently nonlinear.
  • Focused analysis on morphological associative memories.

    • Main Results:

    • Morphological neural networks exhibit nonlinear computation prior to activation functions.
    • The properties of morphological neural networks are significantly different from traditional models.
    • Examined the unique computing and storage characteristics of morphological associative memories.

    Conclusions:

    • Morphological neural networks offer a fundamentally different approach to neural network design.
    • These networks present distinct advantages and properties, especially for associative memory tasks.
    • Further research into morphological models is warranted due to their unique characteristics.