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 Video

Updated: Jan 8, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

991

Dehaze-attention: enhancing image dehazing with a multi-scale, attention-based deep learning framework.

Hao Huang1, G T S Ho2, M W Geda3

  • 1School of Integrated Circuits, Harbin Institute of Technology, Shenzen, 518055, China.

Scientific Reports
|December 19, 2025
PubMed
Summary

Related Concept Videos

Deconvolution01:20

Deconvolution

524
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...
524

You might also read

Related Articles

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

Sort by
Same author

Comparative analysis of traditional and metaverse-based education: impacts on students' all-round development.

BMC psychology·2026
Same author

A lightweight fully convolutional neural network for efficient and accurate tunnel crack detection.

Journal of advanced research·2025
Same author

Comparative analysis of Student's live online learning readiness during the coronavirus (COVID-19) pandemic in the higher education sector.

Computers & education·2021
Same author

The antioxidant effects of riluzole on the APRE-19 celll model injury-induced by t-BHP.

BMC ophthalmology·2017
Same author

Nischarin-siRNA delivered by polyethylenimine-alginate nanoparticles accelerates motor function recovery after spinal cord injury.

Neural regeneration research·2017
Same author

Magnetron Sputtering Deposition Cu@Onion-like N-C as High-Performance Electrocatalysts for Oxygen Reduction Reaction.

ACS applied materials & interfaces·2017

This study introduces Dehaze-Attention, an advanced deep learning model for image dehazing. It effectively restores visibility in complex conditions by using an attention mechanism and multi-scale processing, outperforming existing methods.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Image Processing

Background:

  • Deep learning significantly advanced image dehazing, but many methods struggle with complex atmospheric conditions, leading to poor visibility restoration.
  • Existing dehazing models often rely on assumptions that fail in variable haze densities, limiting their practical application.
  • Effective image dehazing is crucial for applications like aerial imaging and autonomous systems.

Purpose of the Study:

  • To propose an improved image dehazing model, Dehaze-Attention, capable of handling variable haze densities and preserving structural information.
  • To address the limitations of current dehazing methods in complex atmospheric scenarios.
  • To enhance visibility restoration in hazy images for improved downstream applications.

Main Methods:

Keywords:
Atmospheric imagingDeep learningImage dehazingMulti-Scale networkRemote sensing

Related Experiment Videos

Last Updated: Jan 8, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

991
  • Utilized advanced feature extraction via convolutional layers to capture foundational details from hazy images.
  • Integrated an attention mechanism to enable dynamic focus on relevant features and minimize information loss.
  • Incorporated a multi-scale network structure for processing haze across different densities through combined global and local feature analysis.

Main Results:

  • The Dehaze-Attention model achieved state-of-the-art performance on synthesized hazy images under diverse atmospheric conditions.
  • Demonstrated significant improvements in quantitative metrics (Peak Signal-to-Noise Ratio and Structural Similarity Index Measure) compared to baseline models.
  • Subjective evaluations confirmed superior visibility restoration and detail preservation by the proposed model.

Conclusions:

  • The Dehaze-Attention model effectively handles variable haze densities while preserving essential structural information.
  • The model shows significant improvements in both quantitative and subjective evaluations, outperforming existing dehazing approaches.
  • The enhanced visibility restoration capabilities make Dehaze-Attention suitable for applications in aerial imaging, autonomous systems, and remote sensing.