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

Enhanced nonlinear optical limiting in Mn 1.25 Fe 0.7 P 0.44 Si 0.56 Sn 0.2 /rGO nanocomposites via two-photon absorption.

Nanotechnology·2026

Same author

Efficacy of higher-dose versus lower-dose corticosteroids in community-acquired pneumonia: a systematic review and network meta-analysis.

Critical care (London, England)·2026

Same author

Resting-state EEG microstates as biomarkers for major depressive disorder.

Biomedizinische Technik. Biomedical engineering·2026

Same author

Perinatal Outcomes With Continuous Glucose Monitoring in Gestational Diabetes: A Systematic Review and Meta-Analysis.

Journal of diabetes science and technology·2026

Same author

Local fluorescence enhancement of perovskite quantum dots glass via picosecond laser micromachining.

Optics letters·2026

Same author

Adaptively balanced Poisson-constrained physics-informed neural networks for robust displacement integration in background-oriented Schlieren.

Journal of the Optical Society of America. A, Optics, image science, and vision·2026

Search research articles

Related Experiment Video

Updated: Jun 14, 2025

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

Gradient domain model-driven algorithm unfolding network for blind image deblurring.

Zheng Guo¹, Zirui Zhang¹, Wei Yan²

¹School of Physics, Nanjing University of Science and Technology, Nanjing, 210094, China.

Neural Networks : the Official Journal of the International Neural Network Society

|June 12, 2025

Summary

This study introduces a novel gradient-driven algorithm unfolding network (GDUNet) for blind image deblurring. GDUNet effectively restores clear images by leveraging image gradients within a deep unfolding framework.

Keywords:

Algorithm unfolding network Blind image deblurring Image gradient prior

More Related Videos

Magnetic Resonance Derived Myocardial Strain Assessment Using Feature Tracking

Magnetic Resonance Derived Myocardial Strain Assessment Using Feature Tracking

Published on: February 12, 2011

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Related Experiment Videos

Last Updated: Jun 14, 2025

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

Magnetic Resonance Derived Myocardial Strain Assessment Using Feature Tracking

Magnetic Resonance Derived Myocardial Strain Assessment Using Feature Tracking

Published on: February 12, 2011

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Area of Science:

Computer Vision
Artificial Intelligence
Image Processing

Background:

Blind image deblurring is an ill-posed problem requiring simultaneous estimation of clear images and blur kernels.
Algorithm unfolding techniques have advanced image deblurring, but classical image gradient priors are underutilized in deep unfolding frameworks.

Purpose of the Study:

To introduce a gradient-driven algorithm unfolding network (GDUNet) for enhanced blind image deblurring.
To exploit the role of image gradients more effectively within deep unfolding frameworks.

Main Methods:

Generalized a classical sparse gradient deblurring model into a deep unfolding network (GDUNet).
Designed specific proximal mapping modules for diverse prior terms to learn accurate prior distributions.
Incorporated a blur pattern attention module (BPAM) to refine image features and aid blur kernel restoration.

Main Results:

GDUNet demonstrated superior performance on multiple color-blurred image datasets compared to existing state-of-the-art methods.
The proposed network effectively integrates image gradient priors into the deep unfolding paradigm.

Conclusions:

The gradient-driven algorithm unfolding network (GDUNet) offers a promising approach for blind image deblurring.
Integrating image gradients and attention mechanisms enhances the restoration of both clear images and blur kernels.