Abstract:Hyperspectral image (HSI) denoising is an essential preprocessing step for downstream applications. Fully characterizing the spatial-spectral priors of HSI is crucial for HSI denoising tasks. In recent years, denoising methods based on low-rank subspaces have garnered attention. Within the low-rank decomposition framework (LRDF), the restoration of HSIs can be formulated as a problem of restoring two subspace factors. Since the rank of the HSI data has been predetermined by LRDF, subspace-based methods have already characterized the spectral low-rankness information. Next, subspace-based methods only need to encode spatial priors for HSIs. Existing subspace-based methods either rely on a manual-designed regularization or a pre-trained deep neural network. The former fails to fully capture the intrinsic priors of the HSI, while the latter may encounter generalization issues. Inspired by the unsupervised deep image prior (DIP) technique, this article proposes a double subspace deep prior (DSDP) model to track the mentioned issues. In this model, the two subspace factors are parallelly represented by two deep neural networks. By incorporating popular attention modules into classical convolutional neural networks, the well-designed subspace factor neural network can effectively capture the deep prior of the two subspace factors separately from each HSI in an unsupervised manner. Additionally, the total variation (TV) regularizer is added to constrain the generation of the subspace factor neural network, and further to ensure the effectiveness and robustness of the parameter learning process. Extensive experiments demonstrate that our method outperforms a series of competing methods.

Hyperspectral Image Denoising Via Generalized Kronecker Decomposition-Based Subspace Representation

Hyperspectral Image Denoising Via Double Subspace Deep Prior

Group Sparse Nonnegative Matrix Factorization for Hyperspectral Image Denoising

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

Hyperspectral Image Denoising via Nonlocal Spectral Sparse Subspace Representation

A hyperspectral image denoising method based on land cover spectral autocorrelation

Hyperspectral Image Denoising Via Spectral and Spatial Low-Rank Approximation

Hyperspectral Imagery Denoising Via Reweighed Sparse Low-Rank Nonnegative Tensor Factorization

Spatial-spectral convolutional sparse neural network for hyperspectral image denoising

Deep Parameterized Neural Networks for Hyperspectral Image Denoising

Multitask Sparse Nonnegative Matrix Factorization for Joint Spectral–Spatial Hyperspectral Imagery Denoising

Fast Large-Scale Hyperspectral Image Denoising Via Non-Iterative Low-Rank Subspace Representation

Nonlocal Spatial–Spectral Neural Network for Hyperspectral Image Denoising

Hyperspectral Image Denoising Via Sparse Representation and Low-Rank Constraint

Hyperspectral Image Denoising Based on Subspace Low Rank Representation

Hyperspectral Images Denoising via Nonconvex Regularized Low-Rank and Sparse Matrix Decomposition

Hyperspectral Image Denoising Using Nonconvex Fraction Function.

Hyperspectral Image Denoising via Group Sparsity Regularized Hybrid Spatio-Spectral Total Variation

A guidable nonlocal low-rank approximation model for hyperspectral image denoising

Multitask Sparse Representation Model-Inspired Network for Hyperspectral Image Denoising

Hyperspectral Image Denoising Using Global Weighted Tensor Norm Minimum And Nonlocal Low-Rank Approximation