EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HLDA-Net: Hierarchical Local Enhancement and Direction-Aware State-Space Network for Single Image Deraining

Guoqiang Chai, Jin Ge, Xuting Duan, Lesong Pei, Guosheng Zhang

Digital · 06.11.2026 · 6 (4) · 93

DOI: 10.3390/digital6040093 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Digital, online erschienen am 06.11.2026. Verfasst von Guoqiang Chai und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Postgraduate Education Innovation Program of Shanxi Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The problem of image deraining has drawn widespread attention as the blurry images caused by rain streaks can severely impair the practical value of images. Although exiting deep learning-based methods have achieved significant success, single-image deraining still faces great challenges due to the diverse orientations, multi-scale distributions and intense texture entanglement of rain streaks. Specifically, existing state-space-based deraining methods still suffer from relatively uniform local enhancement across different feature hierarchies, representation bias caused by fixed single-direction scanning in two-dimensional (2D) images, and insufficient selection of cross-level information during feature fusion. In this paper, a novel deraining network named HLDA-Net is proposed to address these issues, which unifies hierarchical local enhancement (HLE) and direction-aware dual-branch state-space module (DDS) within a Convolutional Neural Network (CNN) encoder–decoder framework. Level-adaptive detail refinement is embedded in HLE for better background texture restoration, while complementary dual-branch spatial modeling is introduced in DDS to mitigate the directional bias of single-path scanning and strengthen long-distance dependency modeling. In addition, gated skip fusion (GSF) is integrated into the encoder–decoder restoration pipeline to strengthen cross-level feature aggregation and optimize stability. Extensive experiments on synthetic and real-world benchmarks demonstrate the effectiveness of HLDA-Net for single-image deraining, with the peak signal-to-noise ratio (PSNR) and Structural Similarity Index Measure (SSIM) on the Rain100H dataset reaching 31.034 and 0.914, respectively.

Förderung

  • National Natural Science Foundation of China
  • Postgraduate Education Innovation Program of Shanxi Province

Zitieren

APA

Chai, G., Ge, J., Duan, X., Pei, L., & Zhang, G. (2026). HLDA-Net: Hierarchical Local Enhancement and Direction-Aware State-Space Network for Single Image Deraining. Digital, 6(4), 93. https://doi.org/10.3390/digital6040093

Vancouver

Chai G, Ge J, Duan X, Pei L, Zhang G. HLDA-Net: Hierarchical Local Enhancement and Direction-Aware State-Space Network for Single Image Deraining. Digital. 2026;6(4):93. doi:10.3390/digital6040093

BibTeX
@article{Chai2026,
  author = {Chai, Guoqiang and Ge, Jin and Duan, Xuting and Pei, Lesong and Zhang, Guosheng},
  title = {HLDA-Net: Hierarchical Local Enhancement and Direction-Aware State-Space Network for Single Image Deraining},
  journal = {Digital},
  year = {2026},
  volume = {6},
  number = {4},
  pages = {93},
  doi = {10.3390/digital6040093}
}
Kostenlos starten →