HSFMamba: Hierarchical Selective Fusion Mamba Network for Optics-Guided Joint Super-Resolution and Denoising of Noise-Corrupted SAR Images

Synthetic Aperture Radar (SAR) image interpretation has attracted widespread attention in remote sensing applications. However, the performance of existing methods is severely hindered by inherent limitations of SAR imaging mechanisms, such as speckle noise and low resolution. With the continuous advancement of remote sensing, it has become increasingly feasible to simultaneously acquire optical and SAR images. Given rich details in optical images, it is crucial to exploit this valuable information to guide quality enhancement of SAR images, thereby significantly improving their performance for practical applications. In this work, we propose a novel Hierarchical Selective Fusion Mamba Network (HSFMamba) for optics-guided joint super-resolution and denoising of SAR images, which simultaneously addresses resolution limitations and noise corruption in a unified framework. HSFMamba leverages the long-range modeling capability of the state space model with linear complexity and incorporates optical images through two progressive cross-selection scan mechanisms to perform high-quality reconstruction of SAR images corrupted by speckle noise. Specifically, we design a cross-modal feature selection module that dynamically identifies significant representations in optical images, thereby progressively extracting key information. To further leverage optical details while mitigating SAR speckle noise, we develop a frequency-spatial adaptive aggregation module aimed at better restoring image details, effectively enhancing critical high-frequency information. Additionally, we construct a well-aligned and high-resolution dataset for optics-guided joint SAR image super-resolution and denoising, comprising 3,200 optical-SAR image pairs, totaling 25,600 pairs across eight degradation modes. Extensive experiments demonstrate that HSFMamba effectively utilizes optical information to improve SAR image quality, outperforming several state-of-the-art methods.