SAR Despeckling Using Overcomplete Convolutional Networks

TL;DR

This paper introduces an overcomplete convolutional neural network architecture for SAR image despeckling, emphasizing local feature learning to improve despeckling performance over existing methods.

Contribution

It proposes a novel overcomplete CNN architecture that focuses on local features, contrasting with traditional CNNs that emphasize global features for SAR despeckling.

Findings

Improved despeckling results on synthetic SAR images.

Enhanced detail preservation compared to recent methods.

Effective handling of local speckle features.

Abstract

Synthetic Aperture Radar (SAR) despeckling is an important problem in remote sensing as speckle degrades SAR images, affecting downstream tasks like detection and segmentation. Recent studies show that convolutional neural networks(CNNs) outperform classical despeckling methods. Traditional CNNs try to increase the receptive field size as the network goes deeper, thus extracting global features. However,speckle is relatively small, and increasing receptive field does not help in extracting speckle features. This study employs an overcomplete CNN architecture to focus on learning low-level features by restricting the receptive field. The proposed network consists of an overcomplete branch to focus on the local structures and an undercomplete branch that focuses on the global structures. We show that the proposed network improves despeckling performance compared to recent despeckling methods on synthetic and real SAR images.