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

Efficient Glare Suppression Network for Nighttime Images with Lightweight Parallel Attention and Ghost Convolution.

Ruoyu Yang1, Huaixin Chen1, Sijie Luo1

  • 1School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China.

Sensors (Basel, Switzerland)
|June 26, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Light Acquisition02:16

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
Parallel Processing01:20

Parallel Processing

The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...

You might also read

Related Articles

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

Sort by
Same author

Intra-arterial thrombolysis as adjunct to endovascular treatment in acute basilar artery occlusion.

Journal of neurology·2026
Same author

Procedure time and clinical outcomes after thrombectomy: a pooled cohort analysis of three randomised trials.

European stroke journal·2026
Same author

Trends and insights in non-syndromic craniosynostosis research globally: a bibliometric analysis of the Web of Science database (2005-2024).

Translational pediatrics·2026
Same author

Impact of age on clinical outcomes after endovascular therapy for large ischemic stroke.

Journal of neurology·2026
Same author

Species and strain diversity in Staphylococcus drive divergent host responses in human skin.

bioRxiv : the preprint server for biology·2026
Same author

Structural atlas of Pakpunavirus P7-1 reveals determinants of virion stability and genome ejection.

Communications biology·2026

This study introduces a lightweight glare suppression network (LGSNet) for nighttime driving. The efficient LGSNet effectively reduces glare and enhances details, offering a practical solution for edge devices.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Image Processing

Background:

  • Nighttime road scenes suffer from glare, overexposure, and detail loss from artificial lights.
  • Existing glare suppression models are computationally intensive and difficult to deploy on edge devices.

Purpose of the Study:

  • To propose a lightweight glare suppression network (LGSNet) for nighttime road scenes.
  • To address the limitations of existing models regarding parameter count and computational complexity.
  • To enable efficient glare suppression on resource-constrained edge devices.

Main Methods:

  • Developed a lightweight glare suppression network (LGSNet) using ghost depthwise separable convolution (GhostDSC) and Lightweight Parallel Attention (LPA).
  • Integrated GhostDSC blocks into a U-Net architecture to reduce parameters and computational cost.
Keywords:
ghost convolutionglare suppressionlightweight networknighttime images

Related Experiment Videos

  • Employed an LPA module to enhance attention to glare regions and details.
  • Utilized a joint loss function (background, glare, reconstruction loss) for optimized suppression and preservation.

    • Main Results:

    • The proposed LGSNet achieved competitive performance on public (Flare7K++) and custom (NRGD) datasets.
    • LGSNet has significantly fewer parameters (7.45 M) compared to U-Net and Uformer.
    • The method demonstrated effectiveness in suppressing glare and restoring details across various metrics (PSNR, SSIM, LPIPS, NIQE, BRISQUE, PIQE).

    Conclusions:

    • LGSNet offers a superior trade-off between model complexity and performance for glare suppression.
    • The network provides an efficient solution for resource-aware glare suppression tasks on edge devices.
    • The proposed method effectively mitigates glare interference while preserving crucial scene details.