English

SAR Despeckling via Regional Denoising Diffusion Probabilistic Model

Computer Vision and Pattern Recognition 2024-01-09 v1 Image and Video Processing

Abstract

Speckle noise poses a significant challenge in maintaining the quality of synthetic aperture radar (SAR) images, so SAR despeckling techniques have drawn increasing attention. Despite the tremendous advancements of deep learning in fixed-scale SAR image despeckling, these methods still struggle to deal with large-scale SAR images. To address this problem, this paper introduces a novel despeckling approach termed Region Denoising Diffusion Probabilistic Model (R-DDPM) based on generative models. R-DDPM enables versatile despeckling of SAR images across various scales, accomplished within a single training session. Moreover, The artifacts in the fused SAR images can be avoided effectively with the utilization of region-guided inverse sampling. Experiments of our proposed R-DDPM on Sentinel-1 data demonstrates superior performance to existing methods.

Keywords

Cite

@article{arxiv.2401.03122,
  title  = {SAR Despeckling via Regional Denoising Diffusion Probabilistic Model},
  author = {Xuran Hu and Ziqiang Xu and Zhihan Chen and Zhengpeng Feng and Mingzhe Zhu and LJubisa Stankovic},
  journal= {arXiv preprint arXiv:2401.03122},
  year   = {2024}
}

Comments

5 pages, 5 figures

R2 v1 2026-06-28T14:09:59.757Z