Abstract:Image denoising is often empowered by accurate prior information. In recent years, data-driven neural network priors have shown promising performance for RGB natural image denoising. Compared to classic handcrafted priors (e.g., sparsity and total variation), the "deep priors" are learned using a large number of training samples -- which can accurately model the complex image generating process. However, data-driven priors are hard to acquire for hyperspectral images (HSIs) due to the lack of training data. A remedy is to use the so-called unsupervised deep image prior (DIP). Under the unsupervised DIP framework, it is hypothesized and empirically demonstrated that proper neural network structures are reasonable priors of certain types of images, and the network weights can be learned without training data. Nonetheless, the most effective unsupervised DIP structures were proposed for natural images instead of HSIs. The performance of unsupervised DIP-based HSI denoising is limited by a couple of serious challenges, namely, network structure design and network complexity. This work puts forth an unsupervised DIP framework that is based on the classic spatio-spectral decomposition of HSIs. Utilizing the so-called linear mixture model of HSIs, two types of unsupervised DIPs, i.e., U-Net-like network and fully-connected networks, are employed to model the abundance maps and endmembers contained in the HSIs, respectively. This way, empirically validated unsupervised DIP structures for natural images can be easily incorporated for HSI denoising. Besides, the decomposition also substantially reduces network complexity. An efficient alternating optimization algorithm is proposed to handle the formulated denoising problem. Semi-real and real data experiments are employed to showcase the effectiveness of the proposed approach.

Hyperspectral Unmixing Powered by Deep Image Priors and Denoising Regularization.

A Plug-and-Play Priors Framework for Hyperspectral Unmixing

Unrolling Plug-and-Play Network for Hyperspectral Unmixing

Hyperspectral Denoising Using Unsupervised Disentangled Spatio-Spectral Deep Priors

AE-RED: A Hyperspectral Unmixing Framework Powered by Deep Autoencoder and Regularization by Denoising

Hyperspectral Image Denoising Via Double Subspace Deep Prior

Hyperspectral Denoising Using Unsupervised Disentangled Spatiospectral Deep Priors

Tuning-free Plug-and-Play Hyperspectral Image Deconvolution with Deep Priors

Deep Spectral Unmixing Framework Via 3D Denoising Convolutional Autoencoder

Robust Hyperspectral Unmixing with Practical Learning-Based Hyperspectral Image Denoising

Hyperspectral Blind Unmixing Using a Double Deep Image Prior

Hyperspectral Unmixing via Nonnegative Matrix Factorization with Handcrafted and Learnt Priors

Hyperspectral Mixed Noise Removal Via Spatial-Spectral Constrained Unsupervised Deep Image Prior

Deep Parameterized Neural Networks for Hyperspectral Image Denoising

Learning spectral-spatial prior via 3ddncnn for hyperspectral image deconvolution

Joint denoising and unmixing for hyperspectral image

DAEN: Deep Autoencoder Networks for Hyperspectral Unmixing

Hyperspectral Unmixing Network Accounting for Spectral Variability Based on a Modified Scaled and a Perturbed Linear Mixing Model

Learning a Model-Based Deep Hyperspectral Denoiser from a Single Noisy Hyperspectral Image.

Hyperspectral Unmixing Using Robust Deep Nonnegative Matrix Factorization