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

WANG Fei-Ming; LI Meng-Lin; QU Yang; PAN Bin

doi:10.7527/S1000-6893.2025.32804

ACTA AERONAUTICAET ASTRONAUTICA SINICA >

0 1 - 0

DOI: https://doi.org/10.7527/S1000-6893.2025.32804

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

WANG Fei-Ming ,
LI Meng-Lin ,
QU Yang ,
PAN Bin

Expand

1. 南开大学
2.

Received date: 2025-09-19

Revised date: 2025-12-09

Online published: 2025-12-15

Supported by

National Key R&D Program of China;National Natural Science Foundation of China;Fundamental Research Funds for the Central Universities

Fold

Abstract

Due to the influence of scattering characteristics and imaging geometry, single-polarization SAR images suffer from severe speckle noise and significant image degradation, which makes them more prone to texture loss and structural distortion in super-resolution reconstruction. To address this problem, a diffusion-based super-resolution method that integrates frequency-domain processing and structure awareness is proposed. The method uses a latent diffusion model as the backbone, and introduces a wavelet-conditioned guidance module and a direction-aware enhancement module to improve performance.The wavelet-conditioned guidance module performs multi-scale modulation on high-frequency sub-bands through 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 convolution kernels and is embedded into the deep residual structures of the encoder to strengthen the sensitivity of the compressed feature maps to structural information. In the experiments, a realistic degradation model combining speckle noise and blur kernels is also constructed to better approximate the real imaging chain.Experimental results show that the proposed method achieves significant improvements over existing frameworks on multiple datasets, with an average improvement of 8.12% compared with the best competing method, demonstrating its effectiveness and advancement in SAR image super-resolution.

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

Cite this article

WANG Fei-Ming , LI Meng-Lin , QU Yang , PAN Bin . Diffusion super-resolution of SAR images integrating wavelet guidance and structural enhancement[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 0 : 1 -0 . DOI: 10.7527/S1000-6893.2025.32804

References

[1]连文慧.模型驱动神经网络SAR成像特征重建[D]. 中国民航大学,2024.DOI:10.27627/d.cnki.gzmhy.2024.000480.
[2] Xie J, Gao F, Zhou X, Dong J. Wavelet-based bi-dimensional aggregation network for SAR image change detection[J].IEEE Geoscience and Remote Sensing Letters. 2024, 21: 4013705.
[3] 张冰玉, 潘志刚, 姚锴, 等. 基于生成对抗网络的SAR解压缩图像重建算法[J]. 中国科学院大学学报(中英文), 2025, 42(5): 666-676.
[4] Singh P, Shree R. Analysis and effects of speckle noise in SAR images[C]// 2016 2nd International Conference on Advances in Computing, Communication, & Automation (ICACCA) (Fall). 2016: 1–5.
[5] Singh P, Diwakar M, Shankar A, Raj S, Kumar M. A review on SAR image and its despeckling[J]. Archives of Computational Methods in Engineering, 2021, 28(7): 4633–4653.
[6] Wang L, Zheng M, Du W, Wei M, Li L. Super-resolution SAR image reconstruction via generative adversarial network[C]// 2018 12th International Symposium on Antennas, Propagation and EM Theory (ISAPE). 2018: 1–4.
[7] Kanakaraj S, Nair M S, Kalady S. Adaptive importance sampling unscented Kalman filter based SAR image super resolution[J]. Computers & Geosciences, 2019, 133: 104310.
[8] 刘涛, 钱锋, 张葆. 遥感图像的MAP超分辨重建[J]. 液晶与显示, 2018, 33(10): 884–892.
[9] 杨磊, 连文慧, 陈思佳, 等. 稀疏正则驱动的超分辨SAR图像重建[J]. 电讯技术, 2024, 64(9): 1361–1369.
[10] 史振威, 雷森. 图像超分辨重建算法综述[J]. 数据采集与处理, 2020, 35(01): 1–20.
[11] Luo Z, Yu J, Liu Z. The super-resolution reconstruction of SAR image based on the improved FSRCNN[J]. The Journal of Engineering, 2019, 2019(19): 5975–5978.
[12] 罗宇轩, 吴高昌, 高明. 自适应卷积和轻量化Transformer的遥感图像超分辨网络[J]. 计算机工程与应用, 2025, 61(9): 263–276.
[13] Dong W, Xu Y, Qu J, Hou S. Learning multi-modal cross-scale deformable transformer network for unregistered hyperspectral image super-resolution[C]// AAAI. 2024, 38(2): 1573–1581
[14] 李永军, 陈锦智敏, 李孟军, 等. 基于生成对抗网络的图像超分辨重建算法研究[J]. 河南大学学报(自然科学版), 2024, 54(4): 436–442, 485.
[15] 刘艳芳, 李春升, 杨威. 基于深度学习的SAR图像质量提升方法研究[J]. 上海航天(中英文), 2022, 39(3): 91–99.
[16] Ho J, Jain A, Abbeel P. Denoising diffusion probabilistic models[EB/OL]. 2020.
[17] 郭龙飞. 基于扩散模型的图像超分辨率综述[J]. 电视技术, 2025, 49(07): 222–225.
[18] Saharia C, Ho J, Chan W, Salimans T, Fleet D J, Norouzi M. Image super-resolution via iterative refinement[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(4): 4713–4726.
[19] Li H, Yang Y, Chang M, Chen S, Feng H, Xu Z, Li Q, Chen Y. SRDiff: Single image super-resolution with diffusion probabilistic models[J]. Neurocomputing, 2022, 479: 47–59.
[20] Rombach R, Blattmann A, Lorenz D, Esser P, Ommer B. High-resolution image synthesis with latent diffusion models[C]// CVPR. 2022: 10684–10695.
[21] LIU Y F, LI C S, YANG W. Research on SAR image quality enhancement based on deep learning[J]. Shanghai Aerospace (Chinese & English), 2022, 39(3): 91-99.
[22] XIAO Y, YUAN Q, JIANG K, HE J, JIN X, ZHANG L. EDiffSR: An Efficient Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 1-14.
[23] 邹承权, 黄江成, 孙正宝, 等. 遥感图像超分辨率重建方法：综述与实验[J]. 遥感技术与应用, 2025, 40(4): 969-989.
[24] CHEN Z, ZHANG S, XIONG B. Super-Resolving SAR Images with Diffusion Models: A Dual Evaluation of Metrics and Applications[C]//2024 IEEE 17th International Conference on Signal Processing (ICSP). 2024: 340-346.
[25] QIAN Y, CAI Q, PAN Y, LI Y, YAO T, SUN Q, MEI T. Boosting diffusion models with moving average sampling in frequency domain[C]//CVPR. 2024: 8911-8920.
[26] PENG L, CAO Y, PEI R, LI W, GUO J, FU X, WANG Y, ZHA Z J. Efficient Real-world Image Super-Resolution Via Adaptive Directional Gradient Convolution[EB/OL].2024.arXiv:2405.07023
[27] WANG J, YUE Z, ZHOU S, CHAN K C K, LOY C C. Exploiting diffusion prior for real-world image super-resolution[J]. International Journal of Computer Vision, 2024, 132(12): 5929-5949.
[28] HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-507.
[29] 王晓曼, 陈艳平, 杨采薇, 等. 多方向梯度特征提取的嵌套命名实体识别方法[J/OL]. 计算机应用, 2025: 1-10.
[30] 袁姮, 霍欣燃, 姜文涛. 全息梯度差分卷积的图像分类网络[J/OL]. 自动化学报, 2025: 1-25. doi:10.16383/j.aas.c250028.
[31] 孙巍, 王乾宙, 陈雪凌, 等. 真实场景下图像超分辨技术现状与趋势[J]. 中国图象图形学报, 2025, 30(6): 1576-1592.
[32] 闫河, 巫茜. 劣质 SAR 图像退化模型研究[J]. 计算机工程与设计, 2010, 31(24): 5351-5354.
[33] SONG J, MENG C, ERMON S. Denoising diffusion implicit models[J]. arXiv preprint arXiv:2010.02502, 2020.
[34] REN B, MA S, HOU B, HONG D, CHANUSSOT J, WANG J, JIAO L. A dual-stream high resolution network: Deep fusion of GF-2 and GF-3 data for land cover classification[J]. International Journal of Applied Earth Observation and Geoinformation, 2022, 112: 102896.
[35] ICEYE. ICEYE Synthetic Aperture Radar (SAR) Sample Data[EB/OL]. 2025 [2025-07-15]. https://www.iceye.com/resources.
[36] LI X, ZHANG G, CUI H, HOU S, WANG S, LI X, CHEN Y, LI Z, ZHANG L. MCANet: A Joint Semantic Segmentation Framework of Optical and SAR Images for Land Use Classification[J]. International Journal of Applied Earth Observation and Geoinformation, 2022, 106: 102638.
[37] LI Y, LI X, LI W, HOU Q, LIU L, CHENG M M, YANG J. SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection[C]//NeurIPS. 2024.
[38] WU F, ZHANG H, WANG C, LI L, LI J, CHEN W, ZHANG B. SARBuD1.0: A SAR Building Dataset Based on GF-3 FSII Imageries for Built-up Area Extraction with Deep Learning Method[J]. National Remote Sensing Bulletin, 2022, 26(4): 620-631.
[39] ARABBOEV M, BEGMATOV S, RIKHSIVOEV M, NOSIROV K, SAYDIAKBAROV S. A comprehensive review of image super-resolution metrics: classical and AI-based approaches[J]. Acta IMEKO, 2024, 13(1): 1-8.
[40] BLAU Y, MICHAELI T. The perception-distortion tradeoff[C]//CVPR. 2018: 6228-6237.
[41] ZHANG R, ISOLA P, EFROS A A, SHECHTMAN E, WANG O. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric[C]//CVPR. 2018: 586-595.
[42] HEUSEL M, RAMSAUER H, UNTERTHINER T, NESSLER B, HOCHREITER S. GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium[C]//NeurIPS. 2017: 6626-6637.
[43] DENG H, WANG Y, LIU S, XIE X. SAR image despeckling via edge-preserving filtering and nonlocal low-rank approximation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(7): 5973-5989.
[44] ZHANG Y, LIU J, YANG W, GUO Z. Image super-resolution based on structure-modulated sparse representation[J]. IEEE Transactions on Image Processing, 2015, 24(9): 2797-2810.
[45] WANG X, XIE L, DONG C, SHAN Y. Real-ESRGAN: Training real-world blind super-resolution with pure synthetic data[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 1905-1914.
[46] LIM B, SON S, KIM H, NAH S, LEE K M. Enhanced Deep Residual Networks for Single Image Super-Resolution[EB/OL]. 2017.arXiv:1707.02921

Options

Outlines

模态框（Modal）标题

Abstract

Cite this article

References