Abstract:A hyperspectral image (HSI) is often corrupted by various types of noise during image acquisition, e.g., Gaussian noise, impulse noise, stripes, deadlines, and more. Thus, as a preprocessing step, HSI denoising plays a vital role in many subsequent tasks. Recently, a variety of mixed noise removal approaches have been developed for HSI, and the methods based on spatial–spectral double factor and total variation (DFTV) regularization have achieved comparable performance. Additionally, the nonlocal low-rank tensor model (NLR) is often employed to characterize spatial nonlocal self-similarity (NSS). Generally, fully exploring prior knowledge can improve the denoising performance, but it significantly increases the computational cost when the NSS prior is employed. To solve this problem, this article proposes a novel DFTV-based NLR regularization (DFTVNLR) model for HSI mixed noise removal. The proposed model employs low-rank tensor factorization (LRTF) to characterize the spectral global low-rankness (LR), introduces 2-D and 1-D TV constraints on double-factor to characterize the spatial and spectral local smoothness (LS), respectively. Meanwhile, the NLR is applied to the spatial factor to characterize the NSS. Then, we developed an algorithm based on proximal alternating minimization (PAM) to solve the proposed model effectively. Particularly, we effectively controlled the computational cost from two aspects, namely taking small-sized double factor as regularization object and putting the time-consuming NLR model before the main loop with fewer iterations to solve it independently. Finally, considerable experiments on simulated and real noisy HSI substantiate that the proposed method is superior to the related state-of-the-art methods in balancing the denoising effect and speed.

Mixed Noise Removal by Weighted Low Rank Model.

Mixed Noise Removal by Weighted Encoding with Sparse Nonlocal Regularization

Mixed Noise Removal via Laplacian Scale Mixture Modeling and Nonlocal Low-Rank Approximation.

A new nonlocal means based framework for mixed noise removal

Hyperspectral Image Mixed Noise Removal Based on Multidirectional Low-Rank Modeling and Spatial–Spectral Total Variation

Mixed Noise Removal Algorithm Combining Adaptive Directional Weighted Mean Filter and Improved Adaptive Anisotropic Diffusion Model

Sparse nonlocal priors based two-phase approach for mixed noise removal

Real-World Image Denoising Via Weighted Low Rank Approximation

Mixed Impulse and Gaussian Noise Removal Using L0 Sparse Regularization

Optimal Weights Mixed Filter for removing mixture of Gaussian and impulse noises

Adaptive Image Denoising in Mixed Gaussian-impulse Noise Using Weighted Nuclear Norm Minimization

Mixed Noise Removal Based on a Novel Non-Parametric Bayesian Sparse Outlier Model

A New Weighted Mean Filter with a Two-Phase Detector for Removing Impulse Noise.

Hyperspectral Image Mixed Noise Removal via Double Factor Total Variation Nonlocal Low-Rank Tensor Regularization

Mixed Impulse and Gaussian Noise Removal Using Detail-Preserving Regularization

Weighted Schatten &Lt;inline-Formula> &Lt;tex-Math Notation="latex">$p$ &Lt;/tex-Math> &Lt;/inline-Formula>-norm Minimization for Image Denoising and Background Subtraction

Impulsive Noise and Gaussian-Impulsive Mixed Noise 1 Removal by Low Dimensional Manifold Model ∗

An efficient method to remove mixed Gaussian and random-valued impulse noise

A New Noise Removing Algorithm

Hyperspectral Image Mixed Noise Removal via Nonlinear Transform-Based Block-Term Tensor Decomposition

REMOVING RANDOM-VALUED IMPULSE NOISE WITH RELIABLE WEIGHT