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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Reciprocal repulsions enforce heterotypic dendrite segregation in an olfactory circuit.

bioRxiv : the preprint server for biology·2026
Same author

Differential analysis of artificial intelligence teaching tools acceptance among medical teachers based on theory of planned behavior: a cross-sectional study with latent profile and multi-class logistic regression analysis.

BMC medical education·2026
Same author

Transcriptomics and network pharmacology reveal that baicalin alleviates acute lung injury induced by H1N1 via targeting STAT1.

Archives of microbiology·2026
Same author

Electrochemical-Colorimetric Dual-Mode Sensor Based on a Co-Porphyrin Covalent Organic Framework Supported on MWCNTs.

Langmuir : the ACS journal of surfaces and colloids·2026
Same author

Ocular Surface Microbiota-Derived Lipoteichoic Acid Promotes Dry Eye Disease by Inducing Corneal Inflammation.

Investigative ophthalmology & visual science·2026
Same author

Endocytome profiling uncovers cell-surface protein dynamics underlying neuronal connectivity.

Neuron·2026
Same journal

CAFF-CIL: Causality-Aware Freedom Forgetting Approach for Class-Incremental Learning.

IEEE transactions on neural networks and learning systems·2026
Same journal

Harmonic Autoencoding Framework for Multiple Tasks in Magnetic Particle Imaging Reconstruction.

IEEE transactions on neural networks and learning systems·2026
Same journal

A Survey on Human-Centric Voice-Face Multimodal Learning.

IEEE transactions on neural networks and learning systems·2026
Same journal

Vision-Assisted Foundation Model for Solving Multitask Vehicle Routing Problems.

IEEE transactions on neural networks and learning systems·2026
Same journal

FP3O: Enabling Proximal Policy Optimization in Multiagent Cooperation With Parameter-Sharing Versatility.

IEEE transactions on neural networks and learning systems·2026
Same journal

Hierarchical Semantic Concept Modeling for Generalizable Myocardial Pathology Segmentation on Multisequence CMR Images.

IEEE transactions on neural networks and learning systems·2026
See all related articles

Related Experiment Video

Updated: May 10, 2025

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

8.9K

Robust Unsupervised Deep Learning for Nonblind Image Deconvolution With Inaccurate Kernels.

Xinran Qin, Yuhui Quan, Zhuojie Chen

    IEEE Transactions on Neural Networks and Learning Systems
    |April 25, 2025
    PubMed
    Summary
    This summary is machine-generated.

    This study introduces a novel unsupervised deep learning method for nonblind image deconvolution (NBID), eliminating the need for ground truth images. The approach effectively handles noise and kernel errors, outperforming existing unsupervised techniques.

    More Related Videos

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

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

    Published on: November 30, 2022

    2.6K
    Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition
    09:37

    Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

    Published on: August 18, 2022

    2.2K

    Related Experiment Videos

    Last Updated: May 10, 2025

    Deep Neural Networks for Image-Based Dietary Assessment
    13:19

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    8.9K
    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

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

    Published on: November 30, 2022

    2.6K
    Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition
    09:37

    Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

    Published on: August 18, 2022

    2.2K

    Area of Science:

    • Computer Vision
    • Image Processing
    • Machine Learning

    Background:

    • Nonblind image deconvolution (NBID) aims to restore sharp images from blurred and noisy versions using a blur kernel.
    • Current deep learning (DL) models for NBID often require ground truth (GT) images for supervision, limiting their real-world applicability, especially in scientific imaging.
    • Kernel inaccuracies are common in both training and testing data, posing a significant challenge for deconvolution algorithms.

    Purpose of the Study:

    • To develop a fully unsupervised deep learning approach for nonblind image deconvolution (NBID) that does not rely on ground truth (GT) images.
    • To effectively address both measurement noise and kernel errors within an end-to-end training framework.
    • To improve the applicability of NBID methods in real-world scenarios, including scientific imaging.

    Main Methods:

    • A GT-free, end-to-end deep learning training process is proposed for NBID.
    • A self-reconstruction loss is introduced to manage measurement noise without GT supervision.
    • A self-ensemble loss function and ensemble inference scheme, incorporating phase-keeping kernel perturbation, are utilized to handle kernel errors.
    • A shifting mechanism is integrated into the loss functions to resolve shift ambiguity caused by kernel errors.

    Main Results:

    • The proposed unsupervised NBID method demonstrates superior performance compared to existing unsupervised approaches.
    • The method achieves competitive results when compared to recent supervised NBID techniques.
    • The approach effectively handles measurement noise and kernel inaccuracies without requiring ground truth data.

    Conclusions:

    • The developed fully unsupervised DL approach offers a viable solution for NBID, overcoming the limitations of GT-dependent methods.
    • The proposed techniques for handling noise and kernel errors are effective, enabling robust image deconvolution in practical settings.
    • This work advances the field of unsupervised image deconvolution, broadening its potential applications in scientific and other imaging domains.