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

Generalized circular autoregressive models for isotropic and anisotropic Gaussian textures.

K B Eom1

  • 1Department of Electrical and Computer Engineering, George Washington University, Washington, DC 20052, USA. eom@seas.gwu.edu

Journal of the Optical Society of America. A, Optics, Image Science, and Vision
|August 8, 2001
PubMed
Summary

A new generalized circular autoregressive (GCAR) model offers improved texture modeling. This approach effectively captures complex isotropic and anisotropic natural textures using fewer parameters.

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

2-D moving average models for texture synthesis and analysis.

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

Long-correlation image models for textures with circular and elliptical correlation structures.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
See all related articles

Area of Science:

  • Computer Vision
  • Image Processing
  • Statistical Modeling

Background:

  • Traditional texture modeling often struggles with complex natural textures.
  • Existing models may require numerous parameters or fail to capture anisotropy.

Purpose of the Study:

  • Introduce a novel class of random field models: generalized circular autoregressive (GCAR) models.
  • Enhance the modeling of isotropic and anisotropic natural textures.
  • Develop efficient parameter estimation techniques for GCAR models.

Main Methods:

  • Defined GCAR models with specific neighbor relationships and correlation structures.
  • Developed a multistep algorithm for parameter estimation.
  • Investigated the statistical properties of GCAR model estimators.

Related Experiment Videos

Main Results:

  • GCAR models demonstrate superior performance in modeling both isotropic and anisotropic textures compared to previous methods.
  • The model can represent intricate textures efficiently with a limited number of parameters.
  • Synthesized images closely resembled real textures from the Brodatz album, validating the model's efficacy.

Conclusions:

  • GCAR models provide a powerful and flexible framework for natural texture modeling.
  • The proposed estimation algorithm is effective for GCAR models.
  • Further research can explore the limitations and extensions of GCAR models for advanced image analysis.