EUVIMEDEUVIMED entdecken
Crossref

A Dual-Stage Residual Diffusion Model With Perceptual Decoding for Remote Sensing Image Dehazing

Hao Zhou, Yalun Wang, Qian Zhang, Tao Tao, Wenqi Ren

IEEE Transactions on Geoscience and Remote Sensing · 2025 · 63 · 1-12

DOI: 10.1109/tgrs.2025.3600540 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2025. Verfasst von Hao Zhou und 4 weiteren Autor:innen. Die Arbeit wurde bisher 27-mal zitiert (Crossref) und nennt 47 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Collaborative Innovation Project of Anhui Universities, Key Project of Natural Science Foundation of Anhui Provincial Department of Education.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Atmospheric pollutants, such as haze, severely affect the quality of remote sensing images, leading to blurred details and impairing their effectiveness in applications like environmental monitoring and agricultural resource management. In recent years, diffusion models have attracted widespread attention due to their powerful generative capabilities. However, striking a balance between their expensive training costs and actual recovery effectiveness has become a major challenge. To address this key challenge, we propose a dual-stage residual diffusion model with perceptual decoding (DS-RDMPD) for remote sensing image dehazing. The core innovation of our work lies in a DS coarse-to-fine architecture that integrates the traditional U-Net with a diffusion model, enabling efficient adaptation of diffusion-based restoration to the dehazing task. This design not only achieves strong performance but also demonstrates remarkable generalization across various image restoration scenarios. In the first stage, we use the multichannel efficient selective synthesis U-Net (MCESS-UNet) to preprocess the remote sensing haze images. This architecture performs initial dehazing and feature extraction through a multiscale channel attention (MC) block and then performs enhanced spatial feature aggregation through an efficient selective synthesis (ESS) block. The preprocessed image is then used as the conditional input of the RDMPD, where the perceptual decoder improves the generation quality by further decoupling the condition to refine the residual estimate. Extensive experiments on multiple datasets show that DS-RDMPD can achieve satisfactory results with only 300000 iterations and about five sampling steps. It has achieved satisfactory results in both qualitative and quantitative experiments, and also exhibits strong performance in rain removal and deblurring tasks, demonstrating the excellent generalization ability of the model. The code is available at https://github.com/Aaronwangz/DS-RDMPD

Förderung

  • National Natural Science Foundation of China
  • Collaborative Innovation Project of Anhui Universities
  • Key Project of Natural Science Foundation of Anhui Provincial Department of Education

Zitieren

APA

Zhou, H., Wang, Y., Zhang, Q., Tao, T., & Ren, W. (2025). A Dual-Stage Residual Diffusion Model With Perceptual Decoding for Remote Sensing Image Dehazing. IEEE Transactions on Geoscience and Remote Sensing, 63, 1-12. https://doi.org/10.1109/tgrs.2025.3600540

Vancouver

Zhou H, Wang Y, Zhang Q, Tao T, Ren W. A Dual-Stage Residual Diffusion Model With Perceptual Decoding for Remote Sensing Image Dehazing. IEEE Transactions on Geoscience and Remote Sensing. 2025;63:1-12. doi:10.1109/tgrs.2025.3600540

BibTeX
@article{Zhou2025,
  author = {Zhou, Hao and Wang, Yalun and Zhang, Qian and Tao, Tao and Ren, Wenqi},
  title = {A Dual-Stage Residual Diffusion Model With Perceptual Decoding for Remote Sensing Image Dehazing},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2025},
  volume = {63},
  pages = {1--12},
  doi = {10.1109/tgrs.2025.3600540}
}
Kostenlos starten →