航空学报 > 2026, Vol. 47 Issue (10): 532804-532804   doi: 10.7527/S1000-6893.2025.32804

航天遥感图像智能处理与分析专刊

融合小波引导与结构增强的SAR图像扩散超分辨

王飞鸣, 李孟霖, 屈阳, 潘斌()   

  1. 南开大学 统计与数据科学学院,天津 300071
  • 收稿日期:2025-09-19 修回日期:2025-11-06 接受日期:2025-11-28 出版日期:2025-12-17 发布日期:2025-12-15
  • 通讯作者: 潘斌 E-mail:panbin@nankai.edu.cn
  • 基金资助:
    国家重点研发计划(2022YFA1003800);国家自然科学基金(62571273);天津市自然科学基金(25JCLMJC01090)

Diffusion super-resolution of SAR images integrating wavelet guidance and structural enhancement

Feiming WANG, Menglin LI, Yang QU, Bin PAN()   

  1. School of Statistics and Data Science,Nankai University,Tianjin 300071,China
  • Received:2025-09-19 Revised:2025-11-06 Accepted:2025-11-28 Online:2025-12-17 Published:2025-12-15
  • Contact: Bin PAN E-mail:panbin@nankai.edu.cn
  • Supported by:
    National Key Research and Development Program of China(2022YFA1003800);National Natural Science Foundation of China(62571273);Natural Science Foundation of Tianjin(25JCLMJC01090)

摘要:

受散射特性与成像几何影响,单极化SAR遥感图像存在相干斑噪声严重、图像退化显著等特性,导致SAR图像超分辨任务中更容易出现纹理丢失与结构失真。针对这一问题,提出了一种融合频域处理和结构感知的扩散模型超分辨方法。该方法以潜在扩散模型为骨架,引入了小波条件引导模块和方向感知增强模块以提高模型性能。小波条件引导模块通过小波分解和空间自适应归一化操作对高频子带进行多尺度调制,从而依据扩散时间步动态增强纹理表达与高频重建能力。方向感知增强模块集成多种自适应卷积核,以通道注意力方式嵌入编码器深层残差结构,增强压缩特征图对结构信息的敏感性。实验中,还建立了融合相干斑噪声和模糊核的真实退化模型,以贴近真实成像链路。实验表明该方法在多数据集上显著优于现有框架,相比最优方法指标平均改善8.12%,验证了该方法在SAR图像超分辨任务中的有效性与先进性。

关键词: SAR图像超分辨, 遥感, 扩散模型, 小波分解, 通道注意力

Abstract:

Influenced by scattering characteristics and imaging geometry, single-polarization SAR remote sensing images suffer from severe speckle noise and significant degradation, leading to texture loss and structural distortion in SAR image super-resolution tasks. To address this issue, this paper proposes a diffusion model-based super-resolution method that integrates frequency-domain processing and structural perception. The method adopts a latent diffusion model as its backbone and introduces a wavelet-guided module and a direction-aware enhancement module to improve performance. The wavelet-guided module performs multi-scale modulation of high-frequency sub-bands through wavelet decomposition and spatially adaptive normalization, thereby dynamically enhancing texture representation and high-frequency reconstruction capability according to the diffusion timestep. The direction-aware enhancement module incorporates multiple adaptive convolutional kernels, embedded with channel attention into the deep residual structures of the encoder to increase the sensitivity of compressed feature maps to structural information. In experiments, a realistic degradation model combining speckle noise and blur kernels is established to closely approximate the actual imaging pipeline. Results demonstrate that the proposed method significantly outperforms existing frameworks across multiple datasets, achieving an average improvement of 8.12% over the best baseline, verifying its effectiveness and advancement in SAR image super-resolution tasks.

Key words: SAR image super-resolution, remote sensing, diffusion models, wavelet decomposition, channel attention

中图分类号: