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: Jun 15, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Multivariate image segmentation using semantic region growing with adaptive edge penalty.

A K Qin1, David A Clausi

  • 1Department of Systems Design Engineering, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada. qfred008@gmail.com

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|March 19, 2010
PubMed
Summary

This study introduces a new multivariate image segmentation algorithm (MIRGS) that improves accuracy and reduces computational cost. The method uses semantic region growing and a novel initialization technique for better performance on complex 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

SynerNet: Broad-to-precise CAM synergy for weakly supervised semantic segmentation.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

AutoSGRL: Automated framework construction for self-supervised graph representation learning.

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

SPCNet: Deep Self-Paced Curriculum Network Incorporated With Inductive Bias.

IEEE transactions on neural networks and learning systems·2025
Same author

Advertising or adversarial? AdvSign: Artistic advertising sign camouflage for target physical attacking to object detector.

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

One-Nearest Neighborhood Guides Inlier Estimation for Unsupervised Point Cloud Registration.

IEEE transactions on neural networks and learning systems·2024
Same author

Toward Explainable Multiparty Learning: A Contrastive Knowledge Sharing Framework.

IEEE transactions on cybernetics·2024

Area of Science:

  • Computer Vision
  • Image Processing
  • Machine Learning

Background:

  • Multivariate image segmentation faces challenges due to high intraclass variation, feature space sparseness, and computational complexity.
  • Existing methods struggle with class distinguishability and algorithmic robustness.

Purpose of the Study:

  • To present a novel Markov random field (MRF)-based algorithm, multivariate iterative region growing using semantics (MIRGS), for improved multivariate image segmentation.
  • To address intraclass variation, computational cost, and initialization sensitivity in segmentation algorithms.

Main Methods:

  • MIRGS utilizes an MRF spatial context model with adaptive edge penalties applied to regions.
  • Semantic region growing, initiated by watershed over-segmentation, is performed iteratively to reduce solution space complexity.

Related Experiment Videos

Last Updated: Jun 15, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

  • A region-level k-means (RKM) initialization method is employed to mitigate sensitivity to initial conditions.

    • Main Results:

    • MIRGS effectively reduces the impact of intraclass variation and computational demands.
    • The RKM initialization method demonstrates superiority over common alternatives in accuracy and cost.
    • Experiments show MIRGS consistently outperforms three other published segmentation algorithms on diverse datasets.

    Conclusions:

    • MIRGS offers a robust and effective solution for multivariate image segmentation.
    • The proposed RKM initialization significantly enhances the performance and reliability of iterative segmentation algorithms.
    • MIRGS represents a significant advancement in handling complex image segmentation tasks.