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

Image segmentation and selective smoothing by using Mumford-Shah model.

Song Gao1, Tien D Bui

  • 1Department of Computer Science, Concordia University, Montreal, QC H3G 1M8 Canada. sgao@cse.concordia.ca

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|October 22, 2005
PubMed
Summary

A new hierarchical active contour model offers faster image segmentation and smoothing. This method is robust to initial conditions and effectively denoises images, improving upon existing Chan-Vese models.

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

Gene expression profile analysis in response to α1,2-fucosyl transferase (FUT1) gene transfection in epithelial ovarian carcinoma cells.

Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine·2016
Same author

Biosynthesis of silver nanoparticles by the fungus Arthroderma fulvum and its antifungal activity against genera of Candida, Aspergillus and Fusarium.

International journal of nanomedicine·2016
Same author

The expression profile of toll-like receptor signaling molecules in CD19(+) B cells from patients with primary immune thrombocytopenia.

Immunology letters·2016
Same author

Adrenomedullin 2 Improves Early Obesity-Induced Adipose Insulin Resistance by Inhibiting the Class II MHC in Adipocytes.

Diabetes·2016
Same author

Expression of B7-H4 and hepatitis B virus X in hepatitis B virus-related hepatocellular carcinoma.

World journal of gastroenterology·2016
Same author

OPERA-LG: efficient and exact scaffolding of large, repeat-rich eukaryotic genomes with performance guarantees.

Genome biology·2016

Area of Science:

  • Computer Vision
  • Image Processing
  • Computational Mathematics

Background:

  • Active contour models, like those by Chan and Vese, utilize piecewise constant or smooth image representations for segmentation.
  • Existing methods based on Mumford-Shah variational approach and level-set methods can be sensitive to initialization and computationally intensive.

Purpose of the Study:

  • To introduce a novel hierarchical active contour method for image segmentation and smoothing.
  • To address limitations of existing multiphase active contour models, specifically initialization sensitivity and speed.

Main Methods:

  • Development of a hierarchical approach where curve evolution partial differential equations (PDEs) for different level-set functions are decoupled.
  • Solving individual level-set motion equations hierarchically to enhance computational efficiency.

Related Experiment Videos

  • Integration of the diffusion equation for image denoising capabilities.

    • Main Results:

    • Significant speed-up in the segmentation process due to decoupled curve evolution PDEs.
    • Elimination of sensitivity to initial conditions, a common issue in coupled level-set methods.
    • Effective handling of very noisy images through the incorporated diffusion-based denoising.

    Conclusions:

    • The proposed hierarchical active contour method provides a faster, more flexible, and initialization-independent alternative to current models.
    • The method demonstrates robustness and high-quality results, particularly for noisy image segmentation and smoothing.
    • Decoupling level-set evolution equations is key to improving performance and overcoming initialization challenges.