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

A spatially constrained mixture model for image segmentation.

K Blekas, A Likas, N P Galatsanos

    IEEE Transactions on Neural Networks
    |March 25, 2005
    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

    Regularized total least squares approach for nonconvolutional linear inverse problems.

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

    Hierarchical Bayesian image restoration from partially known blurs.

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

    Artificial neural networks for solving ordinary and partial differential equations.

    IEEE transactions on neural networks·2008
    Same author

    Neural-network methods for boundary value problems with irregular boundaries.

    IEEE transactions on neural networks·2008
    Same author

    A novel method for automated EMG decomposition and MUAP classification.

    Artificial intelligence in medicine·2005
    Same author

    Mixture model analysis of DNA microarray images.

    IEEE transactions on medical imaging·2005

    This study introduces a new constrained optimization method for the M-step in Gaussian mixture models (GMMs) used for image segmentation. The novel approach improves segmentation accuracy and objective function values compared to existing methods.

    Area of Science:

    • Artificial Intelligence
    • Computer Vision
    • Machine Learning

    Background:

    • Gaussian mixture models (GMMs) are probabilistic neural networks widely used in image segmentation.
    • Spatially constrained GMMs are typically trained using the expectation-maximization (EM) algorithm.
    • Existing EM implementations for GMMs in image segmentation have limitations.

    Discussion:

    • This work proposes a novel constrained optimization formulation for the M-step of the EM algorithm.
    • The new methodology enhances the training process for spatially constrained GMMs.
    • This addresses limitations in previous implementations of the EM algorithm for GMM-based image segmentation.

    Key Insights:

    • The proposed M-step optimization significantly improves segmentation accuracy.

    Related Experiment Videos

  • The method achieves a higher maximum value of the objective function.
  • Numerical experiments on simulated images validate the superior performance.

    • Outlook:

    • Potential for broader applications of constrained optimization in probabilistic models.
    • Further research into advanced optimization techniques for image analysis.
    • Development of more robust and accurate image segmentation algorithms.