pith. sign in

arxiv: 2604.12600 · v1 · submitted 2026-04-14 · 💻 cs.CV · cs.NA· math.NA

Spatial-Spectral Adaptive Fidelity and Noise Prior Reduction Guided Hyperspectral Image Denoising

Xuelin Xie , Xiliang Lu , Zhengshan Wang , Yang Zhang , Long Chen This is my paper

Pith reviewed 2026-05-10 14:46 UTC · model grok-4.3

classification 💻 cs.CV cs.NAmath.NA
keywords hyperspectral image denoisingmixed noiseadaptive fidelitynoise prior reductiontotal variation regularizationADMM optimizationspatial-spectral processing
0
0 comments X

The pith

A denoising framework for hyperspectral images uses an adaptive weight tensor to dynamically balance data fidelity against noise prior modeling.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The core challenge of hyperspectral image denoising is striking the right balance between data fidelity and noise prior modeling. Most existing methods place too much emphasis on the intrinsic priors of the image while overlooking diverse noise assumptions and the dynamic trade-off between fidelity and priors. The paper proposes a framework that integrates noise prior reduction with a spatial-spectral adaptive fidelity term controlled by an adaptive weight tensor. Within this, it develops a fast pixel-wise model paired with representative coefficient total variation regularization to remove mixed noise while capturing spectral low-rank structure and local smoothness. An ADMM-based solver delivers stable convergence, and experiments on simulated and real datasets show superior denoising with competitive speed.

Core claim

The paper establishes that integrating comprehensive noise prior reduction with fewer parameters and a spatial-spectral adaptive fidelity term, using an adaptive weight tensor to dynamically balance fidelity and prior regularization, enables a pixel-wise model with representative coefficient total variation to accurately remove mixed noise in hyperspectral images while preserving their spectral low-rank structure and local smoothness.

What carries the argument

The adaptive weight tensor that dynamically balances the fidelity and prior regularization terms, combined with the representative coefficient total variation regularizer in a pixel-wise model.

If this is right

  • The method efficiently handles various types of mixed noise in hyperspectral images.
  • It accurately captures the spectral low-rank structure and local smoothness of the images.
  • The ADMM-based optimization algorithm ensures stable and fast convergence.
  • Superior denoising performance is achieved on both simulated and real-world datasets.
  • Competitive computational efficiency is maintained relative to other methods.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The dynamic balancing approach might reduce reliance on manual parameter tuning when noise statistics change across different sensors or scenes.
  • If the noise prior reduction with fewer parameters holds, the framework could be tested as a drop-in preprocessor for downstream hyperspectral tasks such as classification or unmixing.
  • The same adaptive fidelity idea could be examined in related domains that process multi-channel data under mixed noise, such as multispectral video or medical imaging stacks.

Load-bearing premise

The adaptive weight tensor can dynamically and appropriately balance the fidelity and prior regularization terms across diverse mixed-noise scenarios without introducing bias or requiring post-hoc adjustments.

What would settle it

New experiments on additional real-world hyperspectral datasets containing unseen combinations of mixed noise in which the method fails to exceed leading alternatives on standard metrics such as PSNR, SSIM, or visual quality would falsify the performance claims.

Figures

Figures reproduced from arXiv: 2604.12600 by Long Chen, Xiliang Lu, Xuelin Xie, Yang Zhang, Zhengshan Wang.

Figure 1
Figure 1. Figure 1: Description of the three different types of noise [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Flowchart of the proposed framework and FRHD method. where  = ×3𝐕, 𝐕 ∈ ℝ𝐵×𝑅 is an orthogonal matrix. The explicit mathematical formulation of RCTV regularization is as follows: ‖𝐗‖RCTV ∶= ‖∇1 (𝐔)‖1 + ‖∇2 (𝐔)‖1 , (6) where 𝐗 ∈ ℝ𝑀𝑁×𝐵 is the matrix obtained after unfolding the HSI, 𝐗 = 𝐔𝐕⊤ represents the low-rank decomposition, 𝐔 ∈ ℝ𝑀𝑁×𝐵 is the matrix of representative coefficients. 2.4. Overview and compar… view at source ↗
Figure 3
Figure 3. Figure 3: Explanation of the adaptive pixel-wise weighting strategy. Theorem 1. In the ADMM framework, with proper parameter scaling, the denoising frameworks in Definitions 1 and 2 yield identical updates for , , and . This result of Theorem 1 establishes an equivalence at the level of ADMM update rules, rather than a strict equivalence between the two optimization problems. As an immediate consequence, we also … view at source ↗
Figure 4
Figure 4. Figure 4: Five different levels of mixed noise pollution on the CAVE datasets, the R, G, and B channels correspond to spectral bands 14, 23, and 16, respectively. (a) Case 1; (b) Case 2; (c) Case 3; (d) Case 4; (e) Case 5. 6) 𝐖 Subproblems: Extracting all terms related to , we obtain the following subproblem: 𝐖= arg min 𝐖 𝜌 2 ‖ ‖ ‖ ‖ 𝐖⊙(𝐘−𝐔𝐕⊤ )−𝐒−𝐃+ 𝚲3 𝜌 ‖ ‖ ‖ ‖ 2 𝐹 + 𝛼 2 ‖𝕿 − 𝐖‖ 2 𝐹 , 𝑠.𝑡. 𝐖 ∈ [0, 1]𝑀𝑁×𝐵 . (30) Si… view at source ↗
Figure 5
Figure 5. Figure 5: Denoising results of all competing methods on the simulated datasets under Case 5. The pseudo-color images are visualized using spectral bands 14/23/16 (CAVE), 6/20/50 (PaC), and 152/106/20 (WDC) for the R-G-B channels, respectively. nm with 10 nm intervals. The original dataset size is 512 × 512 × 31. For computational efficiency, we cropped the dataset to 200 × 200 × 31. • Harvard PaC Dataset3 : The PaC … view at source ↗
Figure 6
Figure 6. Figure 6: Denoising results of all competing methods on the HYDICE Urban city dataset. The R-G-B channels of the color image are composed of spectral bands 2, 150, and 206, respectively. 3) Quantitative Analysis of results [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Sensitivity analysis of FRHD model parameter 𝑟 under different noise conditions for three datasets. 10-3 10-2 10-1 100 101 102 = 18 20 22 24 26 28 30 32 34 P S N R Case 1 Case 2 Case 3 Case 4 Case 5 (a) CAVE dataset 10-2 100 102 = 22 24 26 28 30 32 34 36 38 P S N R Case 1 Case 2 Case 3 Case 4 Case 5 (b) PaC dataset 10-2 100 102 = 20 25 30 35 P S N R Case 1 Case 2 Case 3 Case 4 Case 5 (c) WDC dataset [PITH… view at source ↗
Figure 8
Figure 8. Figure 8: Sensitivity analysis of FRHD model parameter 𝜏 under different noise conditions for three datasets. (a) (b) (c) (d) (e) [PITH_FULL_IMAGE:figures/full_fig_p016_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Sensitivity of FRHD model parameters 𝛽 and 𝛾 to CAVE dataset under different noise conditions. (a) Case 1; (b) Case 2; (c) Case 3; (d) Case 4; (e) Case 5. In summary, the proposed FRHD method demonstrates excellent performance and computational efficiency on both simulated and real-world datasets. 5.3. Parameter analysis and setting rationale This subsection analyzes the impact of key hyperparameters and e… view at source ↗
Figure 10
Figure 10. Figure 10: Gradient Experiment (Gradually Intensifying Salt-and-Pepper Noise). 0 0.1 0.2 0.3 0.4 0.5 Gaussian Noise Level < 24 26 28 30 32 34 36 38 P S N R - R C T V / F R H D 10 15 20 25 P S N R - N oisy (a) PSNR Comparison RCTV FRHD Noisy 0 0.1 0.2 0.3 0.4 0.5 Gaussian Noise Level < 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 S SIM - R C T V / F R H D 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 S SIM - N oisy (b) SSIM Comparison RCTV… view at source ↗
Figure 11
Figure 11. Figure 11: Gradient Experiment (5% Salt-and-Pepper + 5% Stripes + Gradually Intensifying Gaussian Noise) [PITH_FULL_IMAGE:figures/full_fig_p019_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Iteration results of FRHD method on different simulated datasets. (a) CAVE dataset; (b) PaC dataset; (c) WDC dataset. Regarding potential integration of INR into FRHD, our current framework does not explicitly incorporate INR modules. Integrating INR presents several challenges: (i) training INR networks on high-dimensional HSIs is typically iterative and memory-intensive, which may undermine FRHD’s real-… view at source ↗
read the original abstract

The core challenge of hyperspectral image denoising is striking the right balance between data fidelity and noise prior modeling. Most existing methods place too much emphasis on the intrinsic priors of the image while overlooking diverse noise assumptions and the dynamic trade-off between fidelity and priors. To address these issues, we propose a denoising framework that integrates noise prior reduction and a spatial-spectral adaptive fidelity term. This framework considers comprehensive noise priors with fewer parameters and introduces an adaptive weight tensor to dynamically balance the fidelity and prior regularization terms. Within this framework, we further develop a fast and robust pixel-wise model combined with the representative coefficient total variation regularizer to accurately remove mixed noise in HSIs. The proposed method not only efficiently handles various types of noise but also accurately captures the spectral low-rank structure and local smoothness of HSIs. An efficient optimization algorithm based on the alternating direction method of multipliers is designed to ensure stable and fast convergence. Extensive experiments on simulated and real-world datasets demonstrate that the proposed model achieves superior denoising performance while maintaining competitive computational efficiency.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 4 minor

Summary. The paper proposes a hyperspectral image denoising framework integrating noise prior reduction with a spatial-spectral adaptive fidelity term. An adaptive weight tensor dynamically balances fidelity and prior regularization terms, and a pixel-wise model using representative coefficient total variation (RCTV) regularizer is developed for mixed noise removal. An ADMM-based optimization algorithm ensures convergence, and experiments on simulated and real-world datasets claim superior denoising performance with competitive computational efficiency.

Significance. If the empirical results hold, the approach advances HSI denoising by addressing mixed noise types through adaptive balancing with a modest parameter count, which is relevant for remote sensing applications. Positive aspects include the detailed ADMM procedure with convergence analysis, quantitative metric comparisons to multiple baselines, and evaluation on both simulated and real datasets. These elements support the central claim of effective noise handling while preserving spectral low-rank structure and local smoothness.

major comments (1)
  1. [Experiments section, Table 1] Experiments section, Table 1 (simulated data): the reported PSNR/SSIM gains over the strongest baseline are modest (typically <1 dB) and lack standard deviations or statistical significance tests across repeated noise realizations; this weakens the unqualified claim of 'superior' performance for the central mixed-noise scenario.
minor comments (4)
  1. [Abstract and Section 3.1] Abstract and Section 3.1: the term 'representative coefficient total variation regularizer' is used before a clear definition or reference is provided; a brief inline explanation would improve readability.
  2. [Figure 4] Figure 4 (real-world results): zoomed insets on spectral bands would better illustrate the claimed preservation of fine details and low-rank structure.
  3. [Section 2] Section 2 (related work): the review of prior HSI denoising methods could explicitly contrast the parameter count of the proposed adaptive tensor against the cited deep-learning baselines.
  4. [Section 3] Notation: the adaptive weight tensor W is referenced in the optimization before its full construction (including dependence on noise estimates) is given; moving the definition earlier would reduce forward references.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback and the recommendation for minor revision. We appreciate the recognition of the method's relevance for remote sensing and the positive aspects noted regarding the ADMM procedure and evaluations. We address the major comment on the experiments below.

read point-by-point responses

  1. Referee: Experiments section, Table 1 (simulated data): the reported PSNR/SSIM gains over the strongest baseline are modest (typically <1 dB) and lack standard deviations or statistical significance tests across repeated noise realizations; this weakens the unqualified claim of 'superior' performance for the central mixed-noise scenario.

    Authors: We acknowledge that the PSNR/SSIM improvements in Table 1 are often modest (typically below 1 dB) relative to the strongest baseline. Such incremental gains are common in HSI denoising, where competing methods already perform well on standard metrics, and our contributions lie in the adaptive framework's ability to handle mixed noise with fewer parameters, better preservation of spectral structure, and substantially improved runtime. The overall superiority is further evidenced by results on real-world data, additional metrics (e.g., SAM, ERGAS), and visual comparisons. To address the lack of variability reporting, we will update the revised manuscript to include standard deviations over multiple independent noise realizations for the simulated experiments in Table 1. We will also add a brief statistical analysis (e.g., paired t-tests) to confirm consistency of the improvements. revision: partial

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper introduces a spatial-spectral adaptive fidelity term and noise prior reduction framework, with an adaptive weight tensor presented as a model component balanced via ADMM optimization and convergence analysis. Experiments on simulated and real-world datasets provide external validation through quantitative metrics and baseline comparisons. No quoted step reduces a claimed prediction or uniqueness result to a fitted input by construction, nor does any load-bearing premise collapse to self-citation or definitional equivalence. The adaptive tensor is part of the proposed model design rather than a renamed fit, and the overall argument remains self-contained against the reported benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

Review is based solely on the abstract; full details on parameters, assumptions, and derivations are unavailable. The central claim rests on domain assumptions about HSI structure and the effectiveness of the adaptive balancing mechanism.

free parameters (2)
  • adaptive weight tensor parameters
    Dynamically balances fidelity and prior terms; values are data-dependent or optimized within the model.
  • regularization parameters
    Control the strength of the total variation and low-rank terms; typically tuned to data.
axioms (1)
  • domain assumption Hyperspectral images possess spectral low-rank structure and local smoothness
    Invoked to justify the use of low-rank modeling and total variation regularization on representative coefficients.

pith-pipeline@v0.9.0 · 5491 in / 1367 out tokens · 49526 ms · 2026-05-10T14:46:34.316393+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

50 extracted references · 50 canonical work pages

  1. [1]

    Zhang, Q

    Q. Zhang, Q. Yuan, M. Song, H. Yu, L. Zhang, Cooperated spectral low-rankness prior and deep spatial prior for hsi unsupervised denoising, IEEE Transactions on Image Processing 31 (2022) 6356–6368.doi:10.1109/TIP.2022.3211471

    work page doi:10.1109/tip.2022.3211471 2022

  2. [2]

    IEEE Access9, 143661–143676 (2021) https://doi.org/10.1109/access.2021.3121762

    B. Rasti, Y. Chang, E. Dalsasso, L. Denis, P. Ghamisi, Image restoration for remote sensing: Overview and toolbox, IEEE Geoscience and Remote Sensing Magazine 10 (2) (2021) 201–230.doi:10.1109/MGRS.2021.3121761

    work page doi:10.1109/mgrs.2021.3121761 2021

  3. [3]

    C.Li,S.Li,X.Chen,H.Zheng,Deepbidirectionalhierarchicalmatrixfactorizationmodelforhyperspectralunmixing,AppliedMathematical Modelling 137 (2025) 115736

    work page 2025

  4. [4]

    Dremin, Z

    V. Dremin, Z. Marcinkevics, E. Zherebtsov, A. Popov, A. Grabovskis, H. Kronberga, K. Geldnere, A. Doronin, I. Meglinski, A. Bykov, Skin complicationsofdiabetesmellitusrevealedbypolarizedhyperspectralimagingandmachinelearning,IEEETransactionsonMedicalImaging 40 (4) (2021) 1207–1216

    work page 2021

  5. [5]

    M. Wang, Y. Xu, Z. Wang, C. Xing, Deep margin cosine autoencoder-based medical hyperspectral image classification for tumor diagnosis, IEEE Transactions on Instrumentation and Measurement 72 (2023) 1–12.doi:10.1109/TIM.2023.3293548

    work page doi:10.1109/tim.2023.3293548 2023

  6. [6]

    Zhang, H

    X. Zhang, H. Zhao, Hyperspectral-cube-based mobile face recognition: A comprehensive review, Information Fusion 74 (2021) 132–150

    work page 2021

  7. [7]

    Okujeni, C

    A. Okujeni, C. Jänicke, S. Cooper, D. Frantz, P. Hostert, M. Clark, K. Segl, S. van der Linden, Multi-season unmixing of vegetation class fractions across diverse californian ecoregions using simulated spaceborne imaging spectroscopy data, Remote Sensing of Environment 264 (2021) 112558

    work page 2021

  8. [8]

    J. Peng, W. Sun, H.-C. Li, W. Li, X. Meng, C. Ge, Q. Du, Low-rank and sparse representation for hyperspectral image processing: A review, IEEE Geoscience and Remote Sensing Magazine 10 (1) (2021) 10–43

    work page 2021

  9. [9]

    W. He, Q. Yao, C. Li, N. Yokoya, Q. Zhao, H. Zhang, L. Zhang, Non-local meets global: An iterative paradigm for hyperspectral image restoration, IEEE Transactions on Pattern Analysis and Machine Intelligence 44 (4) (2022) 2089–2107.doi:10.1109/TPAMI.2020. 3027563

    work page doi:10.1109/tpami.2020 2022

  10. [10]

    Y. Chen, H. Zhang, Y. Wang, Y. Yang, J. Wu, Flex-dld: Deep low-rank decomposition model with flexible priors for hyperspectral image denoising and restoration, IEEE Transactions on Image Processing 33 (2024) 1211–1226.doi:10.1109/TIP.2024.3360902

    work page doi:10.1109/tip.2024.3360902 2024

  11. [11]

    Q.Zhang,Y.Zheng,Q.Yuan,M.Song,H.Yu,Y.Xiao,Hyperspectralimagedenoising:Frommodel-driven,data-driven,tomodel-data-driven, IEEE Transactions on Neural Networks and Learning Systems 35 (10) (2024) 13143–13163.doi:10.1109/TNNLS.2023.3278866

    work page doi:10.1109/tnnls.2023.3278866 2024

  12. [12]

    A.Maffei,J.M.Haut,M.E.Paoletti,J.Plaza,L.Bruzzone,A.Plaza,Asinglemodelcnnforhyperspectralimagedenoising,IEEETransactions on Geoscience and Remote Sensing 58 (4) (2020) 2516–2529.doi:10.1109/TGRS.2019.2952062

    work page doi:10.1109/tgrs.2019.2952062 2020

  13. [13]

    L.Zhuang,M.K.Ng,L.Gao,Z.Wang,Eigen-cnn:Eigenimagespluseigennoiselevelmapsguidednetworkforhyperspectralimagedenoising, IEEE Transactions on Geoscience and Remote Sensing 62 (2024) 1–18.doi:10.1109/TGRS.2024.3379199

    work page doi:10.1109/tgrs.2024.3379199 2024

  14. [14]

    Zhang, Y

    Q. Zhang, Y. Dong, Y. Zheng, H. Yu, M. Song, L. Zhang, Q. Yuan, Three-dimension spatial–spectral attention transformer for hyperspectral image denoising, IEEE Transactions on Geoscience and Remote Sensing 62 (2024) 1–13.doi:10.1109/TGRS.2024.3458174

    work page doi:10.1109/tgrs.2024.3458174 2024

  15. [15]

    M. Li, J. Liu, Y. Fu, Y. Zhang, D. Dou, Spectral enhanced rectangle transformer for hyperspectral image denoising, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5805–5814

    work page 2023

  16. [16]

    Y.Chang,L.Yan,H.Fang,S.Zhong,W.Liao,Hsi-denet:Hyperspectralimagerestorationviaconvolutionalneuralnetwork,IEEETransactions on Geoscience and Remote Sensing 57 (2) (2018) 667–682

    work page 2018

  17. [17]

    Q. Shi, X. Tang, T. Yang, R. Liu, L. Zhang, Hyperspectral image denoising using a 3-d attention denoising network, IEEE Transactions on Geoscience and Remote Sensing 59 (12) (2021) 10348–10363

    work page 2021

  18. [18]

    Xiong, J

    F. Xiong, J. Zhou, J. Zhou, J. Lu, Y. Qian, Deep parameterized neural networks for hyperspectral image denoising, IEEE Transactions on Geoscience and Remote Sensing 61 (2023) 1–15. Xuelin Xie et al.:Preprint submitted to Applied Mathematical Modelling Page 23 of 24

    work page 2023

  19. [19]

    J. Li, W. Zhao, T. Xu, Region-based spectral-spatial mutual induction network for hyperspectral image reconstruction, IEEE Transactions on Computational Imaging 10 (2024) 1139–1151.doi:10.1109/TCI.2024.3430478

    work page doi:10.1109/tci.2024.3430478 2024

  20. [20]

    Xiong, J

    F. Xiong, J. Zhou, S. Tao, J. Lu, J. Zhou, Y. Qian, Smds-net: Model guided spectral-spatial network for hyperspectral image denoising, IEEE Trans. Image Process. 31 (2022) 5469–5483

    work page 2022

  21. [21]

    Zhuang, M

    L. Zhuang, M. K. Ng, Fasthymix: Fast and parameter-free hyperspectral image mixed noise removal, IEEE Transactions on Neural Networks and Learning Systems 34 (8) (2023) 4702–4716.doi:10.1109/TNNLS.2021.3112577

    work page doi:10.1109/tnnls.2021.3112577 2023

  22. [22]

    X.Xie,X.Zhang,J.Shen,K.Du,Poplar’swaterloggingresistancemodelingandevaluating:exploringandperfectingthefeasibilityofmachine learning methods in plant science, Frontiers in Plant Science 13 (2022) 821365

    work page 2022

  23. [23]

    Q. Yuan, L. Zhang, H. Shen, Hyperspectral image denoising employing a spectral–spatial adaptive total variation model, IEEE Transactions on Geoscience and Remote Sensing 50 (10) (2012) 3660–3677

    work page 2012

  24. [24]

    H. Tao, Z. Li, An alternative minimization method for tv-image deblurring in tensor space, Applied Mathematical Modelling 145 (2025) 116102

    work page 2025

  25. [25]

    Dabov, A

    K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian, Image denoising by sparse 3-d transform-domain collaborative filtering, IEEE Transactions on Image Processing 16 (8) (2007) 2080–2095.doi:10.1109/TIP.2007.901238

    work page doi:10.1109/tip.2007.901238 2007

  26. [26]

    Sarkar, R

    S. Sarkar, R. R. Sahay, A non-local superpatch-based algorithm exploiting low rank prior for restoration of hyperspectral images, IEEE Transactions on Image Processing 30 (2021) 6335–6348

    work page 2021

  27. [27]

    J. Zhao, J. Wang, T. Zhang, J. Gao, Y. Wan, Tensor low-tubal-rank and sparse representations via dual-learnable transforms for multidimen- sional image recovery, Applied Mathematical Modelling (2025) 116406

    work page 2025

  28. [28]

    Chang, L

    Y. Chang, L. Yan, X.-L. Zhao, H. Fang, Z. Zhang, S. Zhong, Weighted low-rank tensor recovery for hyperspectral image restoration, IEEE transactions on Cybernetics 50 (11) (2020) 4558–4572

    work page 2020

  29. [29]

    Z.-Y.Zhu,T.-Z.Huang,J.Huang,Projectionsubspacebasedlow-rankrepresentationforsparsehyperspectralunmixing,AppliedMathematical Modelling 125 (2024) 463–481

    work page 2024

  30. [30]

    J. Peng, H. Wang, X. Cao, X. Liu, X. Rui, D. Meng, Fast noise removal in hyperspectral images via representative coefficient total variation, IEEE Transactions on Geoscience and Remote Sensing 60 (2022) 1–17

    work page 2022

  31. [31]

    Y. Chen, W. Cao, L. Pang, X. Cao, Hyperspectral image denoising with weighted nonlocal low-rank model and adaptive total variation regularization, IEEE Transactions on Geoscience and Remote Sensing 60 (2022) 1–15

    work page 2022

  32. [32]

    Y.Chen,W.Cao,L.Pang,J.Peng,X.Cao,Hyperspectralimagedenoisingviatexture-preservedtotalvariationregularizer,IEEETransactions on Geoscience and Remote Sensing 61 (2023) 1–14

    work page 2023

  33. [33]

    J. Peng, H. Wang, X. Cao, X. Jia, H. Zhang, D. Meng, Stable local-smooth principal component pursuit, SIAM Journal on Imaging Sciences 17 (2) (2024) 1182–1205

    work page 2024

  34. [34]

    Y. Su, H. Zhu, K.-C. Wong, Y. Chang, X. Li, Hyperspectral image denoising via weighted multidirectional low-rank tensor recovery, IEEE Transactions on Cybernetics 53 (5) (2023) 2753–2766

    work page 2023

  35. [35]

    W. He, H. Zhang, L. Zhang, H. Shen, Total-variation-regularized low-rank matrix factorization for hyperspectral image restoration, IEEE transactions on geoscience and remote sensing 54 (1) (2015) 178–188

    work page 2015

  36. [36]

    L.Zhuang,J.M.Bioucas-Dias,Fasthyperspectralimagedenoisingandinpaintingbasedonlow-rankandsparserepresentations,IEEEJournal of Selected Topics in Applied Earth Observations and Remote Sensing 11 (3) (2018) 730–742

    work page 2018

  37. [37]

    Q. Wang, Z. Wu, J. Jin, T. Wang, Y. Shen, Low rank constraint and spatial spectral total variation for hyperspectral image mixed denoising, Signal Processing 142 (2018) 11–26

    work page 2018

  38. [38]

    S. Xu, X. Cao, J. Peng, Q. Ke, C. Ma, D. Meng, Hyperspectral image denoising by asymmetric noise modeling, IEEE Transactions on Geoscience and Remote Sensing 60 (2022) 1–14.doi:10.1109/TGRS.2022.3227735

    work page doi:10.1109/tgrs.2022.3227735 2022

  39. [39]

    T.Zhang,J.Zhao,S.Fang,Z.Li,Q.Zhang,M.Gong,Hyperspectralimagerestorationviathecollaborationoflow-ranktensordenoisingand completion, Pattern Recognition 165 (2025) 111629

    work page 2025

  40. [40]

    X. Xie, X. Lu, Theta-regularized kriging: Modeling and algorithms, Applied Mathematical Modelling 136 (2024) 115627

    work page 2024

  41. [41]

    X. Rui, X. Cao, Q. Xie, Z. Yue, Q. Zhao, D. Meng, Learning an explicit weighting scheme for adapting complex hsi noise, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 6739–6748

    work page 2021

  42. [42]

    D. L. Donoho, De-noising by soft-thresholding, IEEE transactions on information theory 41 (3) (1995) 613–627

    work page 1995

  43. [43]

    Krishnan, R

    D. Krishnan, R. Fergus, Fast image deconvolution using hyper-laplacian priors, in: Y. Bengio, D. Schuurmans, J. Lafferty, C. Williams, A. Culotta (Eds.), Advances in Neural Information Processing Systems, Vol. 22, Curran Associates, Inc., 2009

    work page 2009

  44. [44]

    G.Liu,Z.Lin,S.Yan,J.Sun,Y.Yu,Y.Ma,Robustrecoveryofsubspacestructuresbylow-rankrepresentation,IEEETransactionsonPattern Analysis and Machine Intelligence 35 (1) (2013) 171–184.doi:10.1109/TPAMI.2012.88

    work page doi:10.1109/tpami.2012.88 2013

  45. [45]

    S.Boyd,N.Parikh,E.Chu,B.Peleato,J.Eckstein,etal.,Distributedoptimizationandstatisticallearningviathealternatingdirectionmethod of multipliers, Foundations and Trends® in Machine learning 3 (1) (2011) 1–122

    work page 2011

  46. [46]

    W.Liang,Z.Tu,J.Lu,K.Tu,M.K.Ng,C.Xu,Fixed-pointconvergenceofmulti-blockpnpadmmanditsapplicationtohyperspectralimage restoration, IEEE Transactions on Computational Imaging 10 (2024) 1571–1587.doi:10.1109/TCI.2024.3485467

    work page doi:10.1109/tci.2024.3485467 2024

  47. [47]

    Y.Chen,J.Zeng,W.He,X.-L.Zhao,T.-X.Jiang,Q.Huang,Fastlarge-scalehyperspectralimagedenoisingvianoniterativelow-ranksubspace representation, IEEE Transactions on Geoscience and Remote Sensing 62 (2024) 1–14.doi:10.1109/TGRS.2024.3458395

    work page doi:10.1109/tgrs.2024.3458395 2024

  48. [48]

    19303–19312

    Y.Luo,X.-L.Zhao,D.Meng,T.-X.Jiang,Hlrtf:Hierarchicallow-ranktensorfactorizationforinverseproblemsinmulti-dimensionalimaging, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 19303–19312

    work page 2022

  49. [49]

    B. He, X. Yuan, On the o(1/n) convergence rate of the douglas–rachford alternating direction method, SIAM Journal on Numerical Analysis 50 (2) (2012) 700–709

    work page 2012

  50. [50]

    Xuelin Xie et al.:Preprint submitted to Applied Mathematical Modelling Page 24 of 24

    M.Hong,Z.-Q.Luo,M.Razaviyayn,Convergenceanalysisofalternatingdirectionmethodofmultipliersforafamilyofnonconvexproblems, SIAM Journal on Optimization 26 (1) (2016) 337–364. Xuelin Xie et al.:Preprint submitted to Applied Mathematical Modelling Page 24 of 24

    work page 2016