SAR image despeckling through convolutional neural networks

TL;DR

This paper introduces a CNN-based method for SAR image despeckling that learns to estimate and subtract speckle noise, demonstrating superior performance on synthetic and real data.

Contribution

It proposes a novel CNN approach using residual learning to effectively remove speckle noise from SAR images, outperforming existing methods.

Findings

Better despeckling performance than state-of-the-art techniques

Effective on both synthetic and real SAR data

Utilizes multitemporal images for training

Abstract

In this paper we investigate the use of discriminative model learning through Convolutional Neural Networks (CNNs) for SAR image despeckling. The network uses a residual learning strategy, hence it does not recover the filtered image, but the speckle component, which is then subtracted from the noisy one. Training is carried out by considering a large multitemporal SAR image and its multilook version, in order to approximate a clean image. Experimental results, both on synthetic and real SAR data, show the method to achieve better performance with respect to state-of-the-art techniques.