Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Deconvolution

Deconvolution

Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A hyperspectral imaging dataset and Grassmann manifold method for intraoperative pixel-wise classification of metastatic colon cancer in the liver.

Computers in biology and medicine·2025

Same author

Low tensor train and low multilinear rank approximations of 3D tensors for compression and de-speckling of optical coherence tomography images.

Physics in medicine and biology·2023

Same author

Non-negative Least Squares Approach to Quantification of <sup>1</sup>H Nuclear Magnetic Resonance Spectra of Human Urine.

Analytical chemistry·2020

Same author

Library-assisted nonlinear blind separation and annotation of pure components from a single <sup>1</sup>H nuclear magnetic resonance mixture spectra.

Analytica chimica acta·2019

Same author

l<sub>0</sub> -Motivated Low-Rank Sparse Subspace Clustering.

IEEE transactions on cybernetics·2018

Same author

Single-Channel Sparse Non-Negative Blind Source Separation Method for Automatic 3-D Delineation of Lung Tumor in PET Images.

IEEE journal of biomedical and health informatics·2016

Same journal

Gaussian-modulated continuous-variable quantum key distribution over 60 km fiber using an integrated silicon photonic receiver.

Optics letters·2026

Same journal

E2E-OCT: end-to-end joint learning model using optical coherence tomography images for vocal cord leukoplakia diagnosis.

Optics letters·2026

Same journal

Holographic generation of panoramic 3D scenes by concave ellipsoidal mirror reflection.

Optics letters·2026

Same journal

Dual-pilot phase recovery with pair-wise maximum-ratio combining for coherent PONs.

Optics letters·2026

Same journal

Mapping the whispering gallery modes of a CaF<sub>2</sub> disk resonator with half-tapered fibers to estimate the fundamental mode volume.

Optics letters·2026

Same journal

Quantitative estimation of deep-subwavelength scale via dark-field scattering axial energy concentration decay profiles.

Optics letters·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Jun 20, 2026

Whole-cell Super-Resolution Imaging via DNA-PAINT on a Spinning Disk Confocal with Optical Photon Reassignment

Whole-cell Super-Resolution Imaging via DNA-PAINT on a Spinning Disk Confocal with Optical Photon Reassignment

Published on: January 6, 2026

3D tensor factorization approach to single-frame model-free blind-image deconvolution.

Ivica Kopriva¹

¹Division of Laser and Atomic Research and Development, Ruder Bosković Institute, Zagreb, Croatia. ikopriva@gmail.com

|September 17, 2009

Summary

This summary is machine-generated.

This study introduces a novel 3D tensor factorization method for single-frame blind-image deconvolution. This approach effectively recovers original images and their derivatives without needing to know the blur kernel size or origin.

More Related Videos

Determining 3D Flow Fields via Multi-camera Light Field Imaging

Determining 3D Flow Fields via Multi-camera Light Field Imaging

Published on: March 6, 2013

Related Experiment Videos

Last Updated: Jun 20, 2026

Whole-cell Super-Resolution Imaging via DNA-PAINT on a Spinning Disk Confocal with Optical Photon Reassignment

Whole-cell Super-Resolution Imaging via DNA-PAINT on a Spinning Disk Confocal with Optical Photon Reassignment

Published on: January 6, 2026

Determining 3D Flow Fields via Multi-camera Light Field Imaging

Determining 3D Flow Fields via Multi-camera Light Field Imaging

Published on: March 6, 2013

Area of Science:

Image processing
Computer vision
Signal processing

Background:

Single-frame blind-image deconvolution (SF BID) is challenging due to unknown blur kernels.
Previous methods often rely on 2D representations and matrix factorization (e.g., NMF, ICA).
These matrix methods require restrictive constraints (sparseness, independence) that are often not met.

Purpose of the Study:

To formulate SF BID as a 3D tensor factorization (TF) problem.
To develop a method that does not require prior knowledge or estimation of the blur kernel's origin or size.
To leverage the advantages of 3D TF over traditional matrix factorization for image deconvolution.

Main Methods:

Applying a bank of 2D Gabor filters to blurred images.
Utilizing the Tucker3 model for 3D tensor factorization of the multichannel blurred image.
Identifying the mixing matrix, original image, and spatial derivatives from the TF factors.

Main Results:

3D TF preserves local image structure, unlike matrix factorization methods.
The PARAFAC model-based 3D TF offers uniqueness under mild conditions.
Demonstrated successful deconvolution on an experimental defocused RGB image.

Conclusions:

3D tensor factorization provides a robust framework for SF BID.
This method overcomes limitations of previous approaches by avoiding blur kernel estimation.
The technique effectively recovers image information and its spatial derivatives.