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

Noise histogram regularization for iterative image reconstruction algorithms.

Samuel T Thurman1, James R Fienup

  • 1The Institute of Optics, University of Rochester, New York 14627, USA. thurman@optics.rochester.edu

Journal of the Optical Society of America. A, Optics, Image Science, and Vision
|February 16, 2007
PubMed
Summary

A new regularization term for image reconstruction uses residual difference histograms to prevent noise overfitting. This method ensures reconstructions statistically match expected noise, improving accuracy for Gaussian and quantization noise.

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

Phase diversity improves retinal image quality in adaptive optics scanning light ophthalmoscopy.

Biomedical optics express·2026
Same author

Phase diversity improves retinal image quality in adaptive optics scanning light ophthalmoscopy.

bioRxiv : the preprint server for biology·2025
Same author

Efficient split-step modeling for imaging through atmospheric turbulence.

Journal of the Optical Society of America. A, Optics, image science, and vision·2025
Same author

Fringe visibility for pulsed digital holography with large optical path differences.

Journal of the Optical Society of America. A, Optics, image science, and vision·2025
Same author

Fourier-optics imaging analysis with ABCD matrices: tutorial.

Journal of the Optical Society of America. A, Optics, image science, and vision·2025
Same author

Using a broadband long-wavelength channel to increase the capture range of segment piston phase retrieval for segmented-aperture systems.

Applied optics·2024

Area of Science:

  • Medical imaging
  • Computational imaging
  • Signal processing

Background:

  • Iterative image reconstruction algorithms are crucial for medical imaging.
  • These algorithms can overfit noise in the image data, degrading reconstruction quality.
  • A robust method to mitigate noise overfitting is needed.

Purpose of the Study:

  • To derive a novel regularization term for iterative image reconstruction.
  • To prevent overfitting of noise in the reconstructed images.
  • To improve the statistical accuracy of reconstructions.

Main Methods:

  • A regularization term was derived based on the histogram of residual differences.
  • The residual difference is between a forward-model image and noisy image data.

Related Experiment Videos

  • The term constrains the residual histogram to match the expected noise histogram.

    • Main Results:

    • Reconstruction results were evaluated using simulated imagery.
    • Performance was tested for both Gaussian and quantization noise.
    • The proposed method demonstrated effectiveness in preventing noise overfitting.

    Conclusions:

    • The derived regularization term effectively constrains image reconstructions.
    • This approach improves reconstruction quality by preventing noise overfitting.
    • The method shows promise for various iterative image reconstruction applications.