Search research articles

Search research articles

Related Experiment Videos

Hopfield network with constraint parameter adaptation for overlapped shape recognition.

P N Suganthan¹, E K Teoh, D P Mital

¹Department of Computer Science and Electrical Engineering, University of Queensland, St. Lucia QLD 4072, Australia.

IEEE Transactions on Neural Networks

|February 7, 2008

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

Shallow and ensemble deep randomized neural network for anomaly detection.

Neural networks : the official journal of the International Neural Network Society·2025

Same author

Online dynamic ensemble deep random vector functional link neural network for forecasting.

Neural networks : the official journal of the International Neural Network Society·2023

Same author

Multimodal Neuroimaging Based Alzheimer's Disease Diagnosis Using Evolutionary RVFL Classifier.

IEEE journal of biomedical and health informatics·2023

Same author

Conv-eRVFL: Convolutional Neural Network Based Ensemble RVFL Classifier for Alzheimer's Disease Diagnosis.

IEEE journal of biomedical and health informatics·2022

Same author

Oblique and rotation double random forest.

Neural networks : the official journal of the International Neural Network Society·2022

Same author

Sample-Based Data Augmentation Based on Electroencephalogram Intrinsic Characteristics.

IEEE journal of biomedical and health informatics·2022

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

This study introduces an adaptive Hopfield network for homomorphic graph matching, improving mapping quality without manual parameter tuning. The method enhances image matching accuracy for applications like key silhouette recognition.

Area of Science:

Computer Vision
Computational Neuroscience
Artificial Intelligence

Background:

Homomorphic graph matching is crucial for pattern recognition.
Traditional Hopfield networks require empirical tuning of constraint parameters.
Adaptive learning methods can improve network performance.

Purpose of the Study:

To develop an energy formulation for homomorphic graph matching using Hopfield networks.
To introduce a Lyapunov indirect method for adaptive constraint parameter learning.
To evaluate the proposed method's performance against fixed-parameter networks.

Main Methods:

Formulating an energy function for Hopfield network-based graph matching.
Implementing a Lyapunov indirect method for adaptive parameter learning.

Related Experiment Videos

Applying the adaptive Hopfield network to silhouette image matching of keys.

Main Results:

The adaptive scheme eliminates the need for empirical constraint parameter specification.
The proposed method generates valid and higher-quality mappings compared to fixed-parameter networks.
Successful application to matching silhouette images of keys demonstrated effectiveness.

Conclusions:

Adaptive constraint parameter learning significantly enhances homomorphic graph matching.
The Lyapunov-based approach offers a robust alternative to manual tuning.
This method shows promise for improved image analysis and pattern recognition tasks.