deSpeckNet: Generalizing Deep Learning Based SAR Image Despeckling

TL;DR

deSpeckNet is a deep learning approach that adaptively estimates noise distribution for SAR image despeckling, demonstrating strong generalization across diverse landcover types and SAR sensors without relying on specific noise models.

Contribution

It introduces a novel DL model that estimates noise distribution and despeckled images simultaneously, enabling broad applicability without predefined noise assumptions.

Findings

Effective speckle reduction across multiple SAR datasets

Generalizes well without specific noise model assumptions

Restores image details while reducing noise

Abstract

Deep learning (DL) has proven to be a suitable approach for despeckling synthetic aperture radar (SAR) images. So far, most DL models are trained to reduce speckle that follows a particular distribution, either using simulated noise or a specific set of real SAR images, limiting the applicability of these methods for real SAR images with unknown noise statistics. In this paper, we present a DL method, deSpeckNet1, that estimates the speckle noise distribution and the despeckled image simultaneously. Since it does not depend on a specific noise model, deSpeckNet generalizes well across SAR acquisitions in a variety of landcover conditions. We evaluated the performance of deSpeckNet on single polarized Sentinel-1 images acquired in Indonesia, The Democratic Republic of Congo and The Netherlands, a single polarized ALOS-2/PALSAR-2 image acquired in Japan and an Iceye X2 image acquired in Germany. In all cases, deSpeckNet was able to effectively reduce speckle and restore