Towards Frequency-Adaptive Learning for SAR Despeckling

TL;DR

This paper introduces SAR-FAH, a frequency-adaptive deep learning model for SAR despeckling that separates image frequencies and employs specialized networks to improve noise reduction and detail preservation.

Contribution

The paper proposes a novel divide-and-conquer architecture using wavelet decomposition and frequency-specific networks, including neural ODEs and deformable U-Nets, for improved SAR image despeckling.

Findings

Outperforms existing methods in noise removal quality.

Better preserves edges and textures in SAR images.

Validated on synthetic and real SAR datasets.

Abstract

Synthetic Aperture Radar (SAR) images are inherently corrupted by speckle noise, limiting their utility in high-precision applications. While deep learning methods have shown promise in SAR despeckling, most methods employ a single unified network to process the entire image, failing to account for the distinct speckle statistics associated with different spatial physical characteristics. It often leads to artifacts, blurred edges, and texture distortion. To address these issues, we propose SAR-FAH, a frequency-adaptive heterogeneous despeckling model based on a divide-and-conquer architecture. First, wavelet decomposition is used to separate the image into frequency sub-bands carrying different intrinsic characteristics. Inspired by their differing noise characteristics, we design specialized sub-networks for different frequency components. The tailored approach leverages statistical variations across frequencies, improving edge and texture preservation while suppressing noise. Specifically, for the low-frequency part, denoising is formulated as a continuous dynamic system via neural ordinary differential equations, ensuring structural fidelity and sufficient smoothness that prevents artifacts. For high-frequency sub-bands rich in edges and textures, we introduce an enhanced U-Net with deformable convolutions for noise suppression and enhanced features. Extensive experiments on synthetic and real SAR images validate the superior performance of the proposed model in noise removal and structural preservation.