Abstract:Hyperspectral images (HSIs) are inevitably corrupted with various types of noise, which seriously degrades the data quality and usability. Denoising is an essential preprocessing task of HSI processing. Recently, benefiting from the great learning ability of deep learning, convolutional neural network (CNN) denoisers have obtained state-of-the-art performances for Gaussian noise removal. However, one central problem remains largely unsolved: how to deal with the complicated noise in the real-world HSIs, especially when a paired training data set is unavailable. In this article, a self-supervised hyperspectral image denoising network (SHDN) is proposed, which consists of a noise estimator and a CNN denoiser. Rather than defining a complex noise model to generate training pairs on the clean HSIs, a self-supervised training scheme is first proposed by considering the noisy HSI itself as the training data. Through the noise estimator, the realistic noise samples can be extracted and combined with the clean bands to make up the training pairs. In addition, to jointly restore the target noisy band and to maintain the spectral consistency, a flexible multi-to-single band convolutional network is designed, where the noisy band and the neighboring bands are jointly aggregated via multiscale contextualized dilated blocks and the spectral–spatial convolutional unit. Experiments on HSIs from spaceborne, airborne, unmanned aerial vehicle (UAV)-borne, and ground-based data sets demonstrate the applicability and the generalization of SHDN in the real scenarios. Additionally, the usability of the noisy bands and the suitability of the SHDN framework in the subsequent applications are verified in the land-cover mapping experiments.

Signal Denoising Using 1D Convolutional Neural Network for Compressed Sensing SAR Imaging Improvement

EiCSNet: Efficient Iterative Neural Network for Compressed Sensing Reconstruction

Convolutional Neural Network Based Computational Imaging Spectroscopy

A Self-Supervised Denoising Network for Satellite-Airborne-Ground Hyperspectral Imagery

A Dictionary Learning Based Unsupervised Neural Network for Single Image Compressed Sensing

A CNN-Based Self-Supervised Synthetic Aperture Radar Image Denoising Approach

Convolutional Neural Network and Guided Filtering for SAR Image Denoising

HTCNet: Hybrid Transformer-CNN for SAR Image Denoising

Spatial-spectral convolutional sparse neural network for hyperspectral image denoising

Nonlocal Spatial–Spectral Neural Network for Hyperspectral Image Denoising

Deep Parameterized Neural Networks for Hyperspectral Image Denoising

Single Image Compressed Sensing MRI via a Self-Supervised Deep Denoising Approach

SWDGAN: GAN-based Sampling and Whole Image Denoising Network for Compressed Sensing Image Reconstruction

Multi-Scale Convolutional Neural Network for SAR Image Semantic Segmentation

Nonsparse SAR Scene Imaging Network Based on Sparse Representation and Approximate Observations

Tree-Structured Dilated Convolutional Networks for Image Compressed Sensing

A novel deep-learning model for RDTS signal denoising based on down-sampling and convolutional neural network

SAR Nonsparse Scene Reconstruction Network via Image Feature Representation Learning

An Unsupervised CNN-Based Multichannel Interferometric Phase Denoising Method Applied to TomoSAR Imaging

Better and Faster, when ADMM Meets CNN: Compressive-Sensed Image Reconstruction

DenoDet: Attention as Deformable Multi-Subspace Feature Denoising for Target Detection in SAR Images