Deep Learning Methods For Synthetic Aperture Radar Image Despeckling: An Overview Of Trends And Perspectives

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 at removing such noise, so as to improve the accuracy of all downstream image processing tasks. The first despeckling methods date back to the 1970's, and several model-based algorithms have been developed in the subsequent years. The field has received growing attention, sparkled by the availability of powerful deep learning models that have yielded excellent performance for inverse problems in image processing. This paper surveys the literature on deep learning methods applied to SAR despeckling, covering both the supervised and the more recent self-supervised approaches. We provide a critical analysis of existing methods with the objective to recognize the most promising research lines, to identify the factors that have limited the success of deep models, and to propose ways forward in an attempt to fully exploit the potential of deep learning for SAR despeckling.

References in corpus (16)

• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Remote Sensing Image Scene Classification: Benchmark and State of the Art
• FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoising
• Deep Visual Domain Adaptation: A Survey
• SAR Image Despeckling Using a Convolutional Neural Network
• Multiplicative Noise Removal Using Variable Splitting and Constrained Optimization
• Deep Graph-Convolutional Image Denoising
• MuLoG, or How to apply Gaussian denoisers to multi-channel SAR speckle reduction?
• Nonlocality-Reinforced Convolutional Neural Networks for Image Denoising
• Multi-Objective CNN Based Algorithm for SAR Despeckling
• deSpeckNet: Generalizing Deep Learning Based SAR Image Despeckling
• SAR Image Despeckling by Deep Neural Networks: from a pre-trained model to an end-to-end training strategy
• Guided patch-wise nonlocal SAR despeckling
• A New Ratio Image Based CNN Algorithm For SAR Despeckling
• A Practical Solution for SAR Despeckling With Adversarial Learning Generated Speckled-to-Speckled Images
• A Novel Cost Function for Despeckling using Convolutional Neural Networks

Cited by in corpus (3)

• Diffusion Models Meet Remote Sensing: Principles, Methods, and Perspectives
• Multi-temporal speckle reduction with self-supervised deep neural networks
• Reduction of rain-induced errors for wind speed estimation on SAR observations using convolutional neural networks