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

Bayesian separation of images modeled with MRFs using MCMC.

Koray Kayabol1, Ercan E Kuruoğlu, Bülent Sankur

  • 1Electrical and Electronics Engineering Department, Istanbul University, Istanbul, Turkey. koray.kayabol@isti.cnr.it

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|April 2, 2009
PubMed
Summary

This study introduces a novel Bayesian approach for random field source separation, significantly outperforming existing methods. The technique effectively separates mixed signals using advanced Markov random fields and modified-Gibbs sampling.

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

Exploring Neurofunctional Phase Transition Patterns in Autism Spectrum Disorder via Thermodynamics Parameters.

Entropy (Basel, Switzerland)·2026
Same author

Modeling brain connectivity dynamics in functional magnetic resonance imaging via particle filtering.

Brain informatics·2021
Same author

Approximate Sparse Multinomial Logistic Regression for Classification.

IEEE transactions on pattern analysis and machine intelligence·2019
Same author

Generalized Bayesian Model Selection for Speckle on Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2018
Same author

The information capacity of the genetic code: Is the natural code optimal?

Journal of theoretical biology·2017
Same author

Time-Dependent Gene Network Modelling by Sequential Monte Carlo.

IEEE/ACM transactions on computational biology and bioinformatics·2015

Area of Science:

  • Signal Processing
  • Computational Imaging
  • Statistical Modeling

Background:

  • Source separation is crucial in various fields.
  • Bayesian methods offer a robust framework for incorporating prior information.
  • Existing methods like Iterated Conditional Modes (ICM) and Independent Component Analysis (ICA) have limitations.

Purpose of the Study:

  • To develop a fully Bayesian method for random field source separation.
  • To incorporate prior image models using Markov random fields.
  • To outperform approximate Bayesian and non-Bayesian source separation techniques.

Main Methods:

  • A Bayesian framework for source separation of random fields.
  • Joint maximization of the a posteriori distribution using numerical methods.

Related Experiment Videos

  • Markov random fields for prior pixel density estimation based on gradient image statistics.
  • Fully Bayesian inference with modified-Gibbs sampling.

    • Main Results:

    • The proposed Bayesian method significantly outperforms approximate Bayesian (ICM) and non-Bayesian (ICA) competitors.
    • Effective source separation demonstrated on synthetic texture and astrophysical images.
    • Robust performance across various noise scenarios.

    Conclusions:

    • The developed fully Bayesian approach provides superior performance for random field source separation.
    • Incorporating prior image models via Markov random fields enhances estimation accuracy.
    • This method offers a powerful alternative to existing source separation techniques.