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

Blind deconvolution of images using optimal sparse representations.

Michael M Bronstein1, Alexander M Bronstein, Michael Zibulevsky

  • 1Department of Computer Science, Technion-Israel Institute of Technology, Haifa 32000, Israel. bronstein@ieee.org

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|June 24, 2005
PubMed
Summary
This summary is machine-generated.

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

Description of a collaborative sperm whale birth and shifts in coda vocal styles during key events.

Scientific reports·2026
Same author

Exploring "dark-matter" protein folds using deep learning.

Cell systems·2024
Same author

Message-Passing Monte Carlo: Generating low-discrepancy point sets via graph neural networks.

Proceedings of the National Academy of Sciences of the United States of America·2024
Same author

A dataset of alternately located segments in protein crystal structures.

Scientific data·2024
Same author

Graph Kernel Neural Networks.

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

RECOVER identifies synergistic drug combinations in vitro through sequential model optimization.

Cell reports methods·2023
Same journal

Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

GoP-based Quality Enhancement on Video Compression.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Align then Tensorize: Multi-Level Consistent Anchor Graph Learning for Scalable Multi-View Clustering.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Beyond Fidelity: Diverse Image Synthesis via Retrieval-Augmented Diffusion.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
See all related articles

This study generalizes the relative Newton algorithm for image blind deconvolution using smooth approximations for sparse sources. It also introduces a sparsification method for arbitrary sources, optimized via supervised learning.

Area of Science:

  • Image processing
  • Signal processing
  • Machine learning

Background:

  • Blind deconvolution aims to recover original signals from mixed or degraded observations without prior knowledge.
  • Existing methods, like the relative Newton algorithm, are effective for 1D signals but limited for complex 2D image data.
  • Sparse signal representation is crucial for effective blind deconvolution.

Purpose of the Study:

  • To extend the relative Newton algorithm for blind deconvolution of images.
  • To develop a robust method for deconvolution of arbitrary sources, not just sparse ones.
  • To optimize the deconvolution process using supervised learning techniques.

Main Methods:

  • Generalization of the relative Newton algorithm for 2D image blind deconvolution.

Related Experiment Videos

  • Utilizing smooth approximation of the absolute value function as a nonlinear term for sparse sources.
  • Introducing a novel sparsification method to handle arbitrary source signals.
  • Employing supervised learning to determine optimal sparsifying transformations.

    • Main Results:

    • Successful generalization of the relative Newton algorithm to image deconvolution.
    • Demonstration of effective blind deconvolution for sparse sources using smooth approximations.
    • Validation of the proposed sparsification method for arbitrary sources.
    • Achieved improved deconvolution performance through learned sparsifying transformations.

    Conclusions:

    • The generalized relative Newton algorithm offers a powerful tool for image blind deconvolution.
    • The proposed sparsification technique enhances the applicability of blind deconvolution to a wider range of sources.
    • Supervised learning provides an effective approach for optimizing blind deconvolution parameters.