Synthetic aperture radar (SAR) images are affected by a spatially correlated and signal-dependent noise called speckle, which is very severe and may hinder image exploitation. Despeckling is an important task that aims to remove such noise so as to improve the accuracy of all downstream image processing tasks. The first despeckling methods date back to the 1970s, and several model-based algorithms have been developed in the years since. The field has received growing attention, sparked by the availability of powerful deep learning models that have yielded excellent performance for inverse problems in image processing. This article surveys the literature on deep learning methods applied to SAR despeckling, covering both supervised and the more recent self-supervised approaches. We provide a critical analysis of existing methods, with the objective of recognizing the most promising research lines; identify the factors that have limited the success of deep models; and propose ways forward in an attempt to fully exploit the potential of deep learning for SAR despeckling.
Deep Learning Methods for Synthetic Aperture Radar Image Despeckling: An Overview of Trends and Perspectives
Poggi G.;Scarpa G.;
2021-01-01
Abstract
Synthetic aperture radar (SAR) images are affected by a spatially correlated and signal-dependent noise called speckle, which is very severe and may hinder image exploitation. Despeckling is an important task that aims to remove such noise so as to improve the accuracy of all downstream image processing tasks. The first despeckling methods date back to the 1970s, and several model-based algorithms have been developed in the years since. The field has received growing attention, sparked by the availability of powerful deep learning models that have yielded excellent performance for inverse problems in image processing. This article surveys the literature on deep learning methods applied to SAR despeckling, covering both supervised and the more recent self-supervised approaches. We provide a critical analysis of existing methods, with the objective of recognizing the most promising research lines; identify the factors that have limited the success of deep models; and propose ways forward in an attempt to fully exploit the potential of deep learning for SAR despeckling.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
