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 by object color: a unifying framework and connection to color constancy.

M H Brill1

  • 1Science Applications International Corporation, McLean, Virginia 22102.

Journal of the Optical Society of America. A, Optics and Image Science
|October 1, 1990
PubMed
Summary

This study introduces a framework for image segmentation that identifies material boundaries by analyzing multispectral image bands. It achieves this by computing spatially invariant quantities, aiding in object color analysis and segmentation.

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

Retinal model with adaptive contrast sensitivity and resolution.

Applied optics·2010
Same author

Mesopic color matching: some theoretical issues.

Journal of the Optical Society of America. A, Optics and image science·1990
Same author

Heuristic analysis of von Kries color constancy.

Journal of the Optical Society of America. A, Optics and image science·1986
Same author

Analysis of human color mechanisms using sinusoidal spectral power distributions.

Journal of the Optical Society of America. A, Optics and image science·1986
Same author

Remarks on signal-processing explanations of the trichromacy of vision.

Journal of the Optical Society of America. A, Optics and image science·1985
Same author

Decomposition of Cohen's matrix R into simpler color invariants.

The American journal of psychology·1985

Area of Science:

  • Computer Vision
  • Image Processing
  • Computational Photography

Background:

  • Image segmentation aims to partition an image into meaningful regions, often based on material properties.
  • Challenges in image segmentation include variations in lighting and viewing angles, which obscure true material reflectance.
  • Existing algorithms struggle to decouple spectral reflectance from illumination effects.

Purpose of the Study:

  • To present a unifying framework for multispectral image segmentation that robustly identifies material boundaries.
  • To develop methods that are invariant to illumination and viewing geometry variations.
  • To explore the relationship between image segmentation and color constancy.

Main Methods:

  • The framework assumes visual stimulus is a sum of spatial and spectral components.

Related Experiment Videos

  • Computation of spatially invariant quantities within object boundaries based on reflected light.
  • Utilizing the unit normal to the color space plane spanned by object pixels as an invariant.

    • Main Results:

    • Identified spatially invariant quantities for multispectral image segmentation under varying illumination.
    • Demonstrated that these invariants correspond to the components of the unit normal in color space.
    • Showcased limited success in estimating spectral-reflectance parameters from these invariants.

    Conclusions:

    • The proposed framework offers a method for segmenting images at material boundaries, robust to illumination changes.
    • A direct link between color constancy and object color-based image segmentation remains a complex challenge.
    • Further research is needed to fully integrate segmentation and color constancy constraints.