EUVIMEDEUVIMED entdecken
Crossref

WaveDH: wavelet sub-bands guided ConvNet for efficient image dehazing

Seongmin Hwang, Daeyoung Han, Cheolkon Jung, Moongu Jeon

Machine Vision and Applications · 2027-01 · 38 (1)

DOI: 10.1007/s00138-026-01943-5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Machine Vision and Applications, online erschienen am 23.10.2026. Verfasst von Seongmin Hwang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 91 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The growing interest in image dehazing has led to significant advancements in deep learning-based single image dehazing approaches, achieving remarkable performance in recent studies. However, many existing methods fail to meet the efficiency demands of practical applications. In this paper, we introduce WaveDH, a novel and compact ConvNet designed to address this efficiency gap. Our WaveDH leverages wavelet sub-bands for guided up-and-downsampling and frequency-aware feature refinement. The key idea lies in utilizing wavelet decomposition to extract low- and high-frequency components from feature levels, allowing for faster processing while upholding high-quality reconstruction. The downsampling block employs a novel squeeze-and-attention scheme to optimize the feature downsampling process in a structurally compact manner, preserving discriminative features while discarding noise components. In our upsampling block, we introduce a dual-upsample and fusion mechanism to enhance high-frequency component awareness, aiding in the reconstruction of fine details. Departing from conventional dehazing methods that treat frequency components equally, our feature refinement block adopts a frequency-aware, coarse-to-fine approach to process features more effectively. By refining details in a maximum 8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times $$\end{document} downsampled feature space, WaveDH strikes a favorable trade-off between accuracy and computational cost. Extensive experiments demonstrate that WaveDH outperforms many state-of-the-art methods across several benchmarks. Notably, on the SOTS-indoor dataset, WaveDH achieves a peak signal-to-noise ratio (PSNR) of 39.35 dB and a structural similarity index (SSIM) of 0.995, surpassing existing high-complexity models while requiring significantly reduced computational costs. Our code is available at https://github.com/smin-hwang/WaveDH.

Zitieren

APA

Hwang, S., Han, D., Jung, C., & Jeon, M. (2027). WaveDH: wavelet sub-bands guided ConvNet for efficient image dehazing. Machine Vision and Applications, 38(1). https://doi.org/10.1007/s00138-026-01943-5

Vancouver

Hwang S, Han D, Jung C, Jeon M. WaveDH: wavelet sub-bands guided ConvNet for efficient image dehazing. Machine Vision and Applications. 2027;38(1). doi:10.1007/s00138-026-01943-5

BibTeX
@article{Hwang2027,
  author = {Hwang, Seongmin and Han, Daeyoung and Jung, Cheolkon and Jeon, Moongu},
  title = {WaveDH: wavelet sub-bands guided ConvNet for efficient image dehazing},
  journal = {Machine Vision and Applications},
  year = {2027},
  volume = {38},
  number = {1},
  doi = {10.1007/s00138-026-01943-5}
}
Kostenlos starten →