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 Video

Updated: Jul 7, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

EM algorithm for image segmentation initialized by a tree structure scheme.

J K Fwu1, P M Djuric

  • 1Dept. of Electr. Eng., State Univ. of New York, Stony Brook, NY.

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|January 1, 1997
PubMed
Summary

This study introduces a novel initialization method for segmenting vector images using the expectation-maximization (EM) technique. This approach enhances image segmentation accuracy without needing prior information, as demonstrated on magnetic resonance brain images.

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

Metabolomics of neural progenitor cells: a novel approach to biomarker discovery.

Cold Spring Harbor symposia on quantitative biology·2008
Same author

The frequency distribution of the number of ion pairs in irradiated tissue.

Radiation research·1998
Same author

On the detection of edges in vector images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·1997
Same author

Unsupervised vector image segmentation by a tree structure-ICM algorithm.

IEEE transactions on medical imaging·1996

Area of Science:

  • Medical image analysis
  • Computer vision
  • Statistical modeling

Background:

  • Image segmentation is crucial for analyzing medical data like MRI scans.
  • Existing methods often require prior information for accurate segmentation.
  • Vector images modeled as multivariate finite mixtures present unique segmentation challenges.

Purpose of the Study:

  • To develop an unsupervised initialization procedure for image segmentation.
  • To improve the performance of the expectation-maximization (EM) algorithm in image segmentation.
  • To segment vector images characterized by Markov random fields (MRFs).

Main Methods:

  • Modeling vector images as multivariate finite mixtures.
  • Utilizing Markov random fields (MRFs) to characterize image data.

More Related Videos

Automatic Identification of Dendritic Branches and their Orientation
06:08

Automatic Identification of Dendritic Branches and their Orientation

Published on: September 17, 2021

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Related Experiment Videos

Last Updated: Jul 7, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Automatic Identification of Dendritic Branches and their Orientation
06:08

Automatic Identification of Dendritic Branches and their Orientation

Published on: September 17, 2021

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

  • Applying the expectation-maximization (EM) algorithm for segmentation.
  • Proposing a novel, information-free initialization technique for the EM algorithm.

    • Main Results:

    • The proposed initialization procedure provides excellent initial estimates for the EM method.
    • Successful segmentation of simulated one-dimensional (1D) and multidimensional magnetic resonance (MR) brain images.
    • Demonstrated the effectiveness of the unsupervised initialization in improving segmentation outcomes.

    Conclusions:

    • The novel initialization method enhances the EM-based segmentation of MRF-modeled vector images.
    • This unsupervised approach offers a robust alternative when prior information is unavailable.
    • The technique shows promise for applications in medical image segmentation, particularly for brain MRI.