Pith. sign in

REVIEW 4 major objections 4 minor 33 references

Bayesian Despeckling of Structured Sources

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proposes BD-QMAP, a Bayesian despeckler for multiplicative Gaussian speckle, and proves that for piecewise-constant first-order Markov sources it reduces to jump-penalized segmentation with a guaranteed MSE lower bound.

desk verdict A clean empirical extension of Q-MAP to speckle noise, but the theoretical lower bound is not proved and the paper's own equations don't agree. read the letter →

arxiv 2501.11860 v2 pith:VDLCLSNK submitted 2025-01-21 cs.IT cs.LGmath.ITstat.AP

Ali Zafari , Shirin Jalali This is my paper · ORCID
classification cs.ITcs.LGmath.ITstat.AP MSC 94A1262F1562M05
keywords specklenoisedespecklingBayesianestimationmultiplicativepiecewiseconstantMarkovsourcequantizedmaximumaposterioriMSElowerbound
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Speckle noise corrupts images in radar, holography, and optical coherence tomography, and many existing despecklers are heuristics without performance guarantees. This paper proposes BD-QMAP, a Bayesian estimator built from the negative log-likelihood of the multiplicative Gaussian noise model plus a regularizer that penalizes disagreement between the quantized empirical distribution of a candidate reconstruction and the known source statistics. For a piecewise-constant first-order Markov source, the regularizer becomes a penalty on the number of jumps, so the optimization reduces to finding a segmentation and averaging $Y_i^2$ over each segment. The paper then proves a lower bound on the mean squared error of such an estimator, giving a theoretical performance certificate that most existing despecklers lack.

What carries the argument

The driving object is the regularizer $c_w(u^n)=\sum_{a^k} w_{a^k}\hat{p}_k(a^k|u^n)$, with weights $w_{a^k}=-\log P([X^k]_b=a^k)$; it summarizes the source's $n$-dimensional distribution in $|X_b|^k$ quantized weights. In BD-QMAP this regularizer is added to the negative log-likelihood of the multiplicative Gaussian model $Y_i=X_iW_i$, $W_i\sim\mathcal{N}(0,1)$. The reduction to Eq. (7) is carried by the observation that, for the piecewise-constant first-order Markov source with $k=2$, the regularizer is proportional to the number of jumps $N_J([u^n]_b)$. The proof of Theorem 1 then combines a concentration bound for sums of geometric run lengths (Lemma 2) with a Gamma-ratio estimate to control within-segment estimation error.

What would settle it

Evaluate the full BD-QMAP objective of Eq. (5) and the simplified objective of Eq. (7) for a fixed piecewise-constant sequence with chosen $q_0$, $b$, and $\lambda$; if their minimizers differ for any such instance, the claimed reduction in Lemma 1 is false and Theorem 1 does not bound BD-QMAP as defined.

Watch

Extended reading notes

Core claim

The paper's central claim is that BD-QMAP, defined in Eq. (3) as the minimizer of $\frac{1}{n}\sum_{i=1}^n \left(\log u_i^2 + \frac{Y_i^2}{u_i^2}\right) + \frac{\lambda}{b} c_w(u^n)$, is a theoretically grounded despeckler for structured stationary sources. For a piecewise-constant first-order Markov source with jump probability $q_0$ and $k=2$, it simplifies to the jump-penalized segmentation of Eq. (7): choose segment lengths $n_1, \ldots, n_{k+1}$, minimize $\sum_j n_j \log\left(\frac{1}{n_j}\sum_{l\in I_j} Y_l^2\right) + \frac{n}{n-1}(\lambda+\frac{1}{b})k$, and reconstruct each segment as the square root of the average of the squared observations. Theorem 1 lower-bounds the expected per-sample MSE of this estimator, given the number of jumps $k=k(n)$, by an expression involving $q_0$, the second moment $\eta_2=E[X_i^2]$, and a geometric run length $T$; Corollary 1 shows the remainder term is $O(n^{-1/4})$.

Load-bearing premise

The theoretical results assume that the BD-QMAP regularizer for a piecewise-constant first-order Markov source is exactly equivalent to the simple jump-count penalty in Eq. (7); the proof uses the coefficient $\lambda+1/b$, while the stated simplification in Eq. (5) has $\lambda+\eta/b$, so if that equivalence is not exact, Theorem 1 bounds a different optimization than the one BD-QMAP defines.

Editorial extensions

If this is right

  • BD-QMAP turns despeckling of piecewise-constant sources into a tractable jump-detection problem solvable by the Viterbi algorithm, followed by per-segment averaging of squared observations.
  • The lower bound of Theorem 1 gives a benchmark against which practical despecklers can be measured, including the genie-aided ML estimator used in the experiments.
  • The method does not require an upper bound on speckle power or local stationarity assumptions, so it can operate in high-noise regimes where linear adaptive filters saturate or bypass the input.
  • Because the regularizer is learned from quantized source statistics rather than a parametric prior, the approach is stated for general structured stationary sources, not only the piecewise-constant case tested here.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same regularizer-to-sparsity reduction might be attempted for higher-order Markov or tree-structured sources; whether the quantized weights collapse to a jump count in those settings is a direct testable extension.
  • The bound's dependence on geometric run lengths suggests that any source with identically distributed segment lengths will exhibit the same qualitative floor, roughly $1/4$ per segment as $T$ grows.
  • The two-stage pipeline---coarse quantized jump detection followed by continuous ML refinement---suggests a transferable recipe for coherent imaging: discrete structure search first, then continuous parameter estimation.
  • A numerical check of the $\eta$ coefficient in Eq. (5) for small $b$ would either close the gap between the stated objective and the proof or show that the theorem applies to a slightly different estimator.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes BD-QMAP, a Bayesian despeckling method for multiplicative Gaussian speckle, defined in Eq. (3) as the minimization of a negative log-likelihood fidelity term plus a regularizer built from quantized empirical statistics. For piecewise-constant first-order Markov sources, the paper claims in Lemma 1 that the method reduces to a jump-penalized segmentation problem, and it derives a lower bound on reconstruction MSE in Theorem 1 and Corollary 1. Experiments on synthetic piecewise-constant sources compare BD-QMAP with classical filters and with a genie-aided ML estimator, reporting improved PSNR. The practical algorithm is clearly implementable via Viterbi decoding, but the theoretical claims contain several unresolved inconsistencies.

Significance. If the theoretical results were valid, the paper would make a useful contribution: a tractable Bayesian treatment of multiplicative noise for structured sources, a concrete Viterbi implementation, and a certified performance benchmark. The experimental comparison is informative and includes a parameter study of the regularization weight and quantization level. However, the advertised lower bound and the claimed equivalence between BD-QMAP and a jump-penalized segmentation are not established: the regularizer coefficient in Eq. (5) is inconsistent with Lemma 1, the proof of Theorem 1 uses a false genie-aided optimality assertion, and the theorem and corollary disagree in signs and constants. These issues are load-bearing for the central theoretical claims.

major comments (4)
  1. [Section 3.2, Lemma 1, Eq. (5) and Eq. (10)] The jump penalty in Eq. (5) has coefficient λ + η/b, with η = -λ log q0 - log(1 - q0 + q0 2^{-b}), while Lemma 1 in Eq. (7) and the proof in Eq. (10) use the coefficient λ + 1/b with no factor η. Since η is not identically equal to 1 (for example, with q0 = 0.1, λ = 1, b = 2, one obtains η ≈ 0.31), Lemma 1 does not describe the optimization in Eq. (5). The simplification of the BD-QMAP regularizer to a jump count is therefore unsupported as stated, and any theorem built on that simplification does not apply to BD-QMAP as defined.
  2. [Section 4.2, Eq. (13) and following paragraph] The proof of Theorem 1 asserts that 'the expected error is minimized when the locations of the jumps are detected correctly.' This is not a property of the objective in Eq. (13), which minimizes negative log-likelihood, not squared-error risk. A segmentation that merges two close adjacent levels, or splits a long constant block, can have lower MSE than the true-location genie estimator while having a similar or even larger likelihood. Consequently, the inequality E[||X^n - hat X^n||^2] ≥ E[||X^n - hat X^{n,ML}||^2] used in Eqs. (16)-(17) is not justified. Without this inequality, Theorem 1 only computes the MSE of a genie-aided estimator with known jump locations; it does not lower-bound the MSE of the fixed-k estimator in Eq. (13), let alone the actual BD-QMAP optimization in Eq. (3).
  3. [Theorem 1, Corollary 1, and Eq. (23)] There are two independent discrepancies in the constant terms. First, Theorem 1 defines c1 = 3 - sqrt(2/π) - 2√π, but the proof in Eq. (23) evaluates the T = 1 and T = 2 contributions as q0(3 - sqrt(2/π) - √π - q0(2 - √π)), which requires c1 = 3 - sqrt(2/π) - √π rather than the printed value. Second, Corollary 1 changes the + q0^2 c2 term of Theorem 1 into - q0^2 c2. These are not cosmetic differences, since the resulting bound can change sign for small q0; as printed, the theorem, its proof, and the corollary cannot all be correct.
  4. [Section 5.3 and Theorem 1 assumptions] The simulations use a uniform source distribution on [0,1], but Theorem 1 and the BD-QMAP objective in Eq. (3) are stated under the assumption x_m > 0. Moreover, for the b-bit quantized search space X_b^n, the value 0 is a valid grid point, and at u_i = 0 the fidelity term log u_i^2 is undefined (it diverges to -∞). The paper neither imposes x_m > 0 in the experiments nor specifies how the zero quantization bin is handled, so the comparison in Figure 2 and Table 1 between the empirical PSNR and the Theorem 1 bound is not covered by the stated assumptions.
minor comments (4)
  1. [Abstract] The abstract contains a typo: 'depseckler' should be 'despeckler'.
  2. [Section 4.1, Eq. (10)] In Eq. (10), the sum is written with index j but the summand uses i; this should be indexed consistently.
  3. [Section 3.2, Eq. (5) and Section 5.1] The derivation of the regularizer simplification is imported from [27, Sec. 3.2] without being reproduced. Given that the coefficient mismatch in Lemma 1 depends on that derivation, the relevant argument from [27] should be stated or at least sketched.
  4. [Table 1] Table 1 reports mean PSNR values without standard deviations or per-trial statistics; with 100 test signals, reporting only single numbers makes it difficult to judge whether the gains are stable across realizations.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the lower bound is derived from the stated Markov and speckle model, and the Q-MAP regularizer is self-cited background rather than a fitted prediction.

full rationale

I cannot exhibit any step in which a claimed prediction or first-principles result reduces by construction to its inputs. BD-QMAP is defined in Eq. (3), and the piecewise-constant simplification in Eq. (7) is obtained by an analytical minimization over segment lengths and per-segment ML amplitude estimates, not by fitting to the quantities the paper later reports. Theorem 1's lower bound is a distributional calculation over the stated first-order Markov source and Gaussian speckle; it does not use the experimental PSNR values, and the genie-aided comparison is a theoretical benchmark, not a fitted output renamed as a prediction. The regularizer weights in Eq. (2) are estimated from samples of the same source model used for testing, which is self-referential in a statistical sense but is the standard Bayesian prior-construction setting and does not make the theory circular. The main load-bearing citation, [27], is the authors' own Q-MAP framework, so the method is not fully self-contained; however, that prior work supplies a regularizer construction rather than the despeckling theorem, and the lower-bound proof is carried out here. I do note two serious correctness risks that are not circularity: the proof of Theorem 1 in Section 4.2 asserts without proof that 'the expected error is minimized when the locations of the jumps are detected correctly,' which is an unjustified genie-optimality claim, and the proof in Eq. (10) silently replaces the coefficient lambda + eta/b with lambda + 1/b, so Lemma 1 and Theorem 1 may describe a different optimization than Eq. (5). These are mathematical derivation concerns, not equivalence-by-construction or fitted-input circularity. The score of 2 reflects the presence of a self-citation in the method's derivation, not a finding that the central theoretical claim is circular.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central derivation rides on the assumed Markov model and on the prior Q-MAP framework, with lambda and b as tuned hyperparameters. No new physical entities are introduced.

free parameters (3)
  • lambda (regularization weight) = not reported; swept per q0/b in Figure 3
    Controls the jump penalty in Eq. (5) and is tuned on the test data; no validation split is described.
  • b (quantization level) = 2 or 3 bits
    Selected per experiment in Table 1 and Figure 3; affects state count and accuracy.
  • Filter window size for linear baselines = 1/(2 q0)
    Set using the true source parameter q0 in Section 5.3.3; baselines are given information a real system may not have.
assumptions (6)
  • domain assumption The speckle noise W_i is i.i.d. standard normal and independent of X^n
    Section 2; this is the fully developed speckle model (Goodman).
  • domain assumption The source is stationary first-order Markov with transition kernel (1-q0) delta_xi + q0 pi_c(xm, xM), support [xm, xM], xm > 0
    Section 3.2; the lower bound and experiments depend on this exact model.
  • domain assumption Weights computed from 10^7 sampled sequences with fixed q0 approximate the true quantized distribution P([X^k]_b = a^k)
    Section 5.3; training and test data come from the same parameter setting.
  • ad hoc to paper Restricting the search to the quantized space X_b^n and applying Viterbi yields a faithful approximation of BD-QMAP
    Remark 1 and Section 5.1; the paper acknowledges this is suboptimal but relies on it for the experiments.
  • standard math The simplification results of [27, Sec. 3.1 and 3.2] extend from additive to multiplicative noise
    Sections 3.1 and 3.2 invoke them without reproving; this is the main unverified transfer.
  • standard math Concentration of geometric sums (Lemma 2, from Janson) and the log bounds (from Topsoe)
    Used in the proof of Theorem 1 and Corollary 1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bayesian Despeckling of Structured Sources." pith.science (2026). https://pith.science/paper/VDLCLSNK

@misc{pith2026250111860,
  author       = {Pith},
  title        = {Pith review of: Bayesian Despeckling of Structured Sources},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VDLCLSNK}},
  note         = {Machine review of arXiv:2501.11860}
}
read the original abstract

Speckle noise is a fundamental challenge in coherent imaging systems, significantly degrading image quality. Over the past decades, numerous despeckling algorithms have been developed for applications such as Synthetic Aperture Radar (SAR) and digital holography. In this paper, we aim to establish a theoretically grounded approach to despeckling. We propose a method applicable to general structured stationary stochastic sources. We demonstrate the effectiveness of the proposed method on piecewise constant sources. Additionally, we theoretically derive a lower bound on the despeckling performance for such sources. The proposed depseckler applied to the 1-Markov structured sources achieves better reconstruction performance with no strong simplification of the ground truth signal model or speckle noise.

Figures

Figures reproduced from arXiv: 2501.11860 by the authors.

Figure 1
Figure 1. Piecewise constant source (X) with parameter q0 = 0.001 sampled under speckle noise (Y ), enhanced Lee and BD-QMAPb despeckled reconstructions, as discussed in Section 5. associated with state si = u i i−k+1 is updated recursively as: Ci(si) = min s ′∈S(si)  log u 2 i + y 2 i u 2 i + λ b wu i i−k+1 + Ci−1(s ′ )  , for i ∈ {k, . . . , n}, where S(si) denotes the set of all states that transition into si , and |S(si… view at source ↗
Figure 2
Figure 2. MSE lower bound of the genie-aided ML despeckler as derived in Theorem [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Effect of hyperparameter λ in BD-QMAPb and BD-QMAP optimization for different choices of bits (b) and source structure (q0). Acknowledgment AZ and SJ were supported by NSF grant CCF-2237538. References [1] F. Argenti, A. Lapini, T. Bianchi, and L. Alparone, “A tutorial on speckle reduction in synthetic aperture radar images,” IEEE Geoscience and Remote Sensing Magazine, vol. 1, no. 3, pp. 6–35, 2013. 1 [2] V. Bianco… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

  1. [1]

    A tutorial on speckle reduction in synthetic aperture radar images,

    F. Argenti, A. Lapini, T. Bianchi, and L. Alparone, “A tutorial on speckle reduction in synthetic aperture radar images,” IEEE Geoscience and Remote Sensing Magazine , vol. 1, no. 3, pp. 6–35, 2013. 1

    work page 2013

  2. [2]

    Strategies for reducing speckle noise in digital holography,

    V. Bianco, P. Memmolo, M. Leo, S. Montresor, C. Distante, M. Paturzo, P. Picart, B. Javidi, and P. Ferraro, “Strategies for reducing speckle noise in digital holography,”Light: Science & Applications , vol. 7, no. 1, p. 48, 2018. 1

    work page 2018

  3. [3]

    Speckle in optical coherence tomography,

    J. M. Schmitt, S. Xiang, and K. M. Yung, “Speckle in optical coherence tomography,” Journal of biomedical optics, vol. 4, no. 1, pp. 95–105, 1999. 1

    work page 1999

  4. [4]

    A review of speckle filtering in the context of estimation theory,

    R. Touzi, “A review of speckle filtering in the context of estimation theory,”IEEE Transactions on Geoscience and Remote Sensing , vol. 40, no. 11, pp. 2392–2404, 2002. 1

    work page 2002

  5. [5]

    J. W. Goodman, Speckle phenomena in optics: theory and applications . Roberts and Company Pub- lishers, 2007. 1

    work page 2007

  6. [6]

    Digital image enhancement and noise filtering by use of local statistics,

    J.-S. Lee, “Digital image enhancement and noise filtering by use of local statistics,”IEEE Transactions on Pattern Analysis and Machine Intelligence , no. 2, pp. 165–168, 1980. 1, 13, 14 15

    work page 1980

  7. [7]

    Adaptive noise smoothing filter for images with signal-dependent noise,

    D. T. Kuan, A. A. Sawchuk, T. C. Strand, and P. Chavel, “Adaptive noise smoothing filter for images with signal-dependent noise,”IEEE Transactions on Pattern Analysis and Machine Intelligence , no. 2, pp. 165–177, 1985. 2, 13, 14

    work page 1985

  8. [8]

    Adaptive speckle filters and scene heterogeneity,

    A. Lopes, R. Touzi, and E. Nezry, “Adaptive speckle filters and scene heterogeneity,”IEEE transactions on Geoscience and Remote Sensing , vol. 28, no. 6, pp. 992–1000, 1990. 2, 13, 14

    work page 1990

Show all 33 references
  1. [9]

    Refined filtering of image noise using local statistics,

    J.-S. Lee, “Refined filtering of image noise using local statistics,”Computer graphics and image process- ing, vol. 15, no. 4, pp. 380–389, 1981. 2

    1981

  2. [10]

    Maximumaposteriorispecklefilteringandfirstordertexture models in sar images,

    A.Lopes, E.Nezry, R.Touzi, andH.Laur, “Maximumaposteriorispecklefilteringandfirstordertexture models in sar images,” in10th annual international symposium on geoscience and remote sensing . Ieee, 1990, pp. 2409–2412. 2, 13

    1990

  3. [11]

    Adaptive restoration of images with speckle,

    D. Kuan, A. Sawchuk, T. Strand, and P. Chavel, “Adaptive restoration of images with speckle,”IEEE Transactions on Acoustics, Speech, and Signal Processing , vol. 35, no. 3, pp. 373–383, 1987. 2

    1987

  4. [12]

    Nonlinear total variation based noise removal algorithms,

    L. I. Rudin, S. Osher, and E. Fatemi, “Nonlinear total variation based noise removal algorithms,”Physica D: nonlinear phenomena , vol. 60, no. 1-4, pp. 259–268, 1992. 2

    1992

  5. [13]

    A variational approach to removing multiplicative noise,

    G. Aubert and J.-F. Aujol, “A variational approach to removing multiplicative noise,”SIAM Journal on Applied Mathematics , vol. 68, no. 4, pp. 925–946, 2008. 2, 13

    2008

  6. [14]

    A nonlinear inverse scale space method for a convex multiplicative noise model,

    J. Shi and S. Osher, “A nonlinear inverse scale space method for a convex multiplicative noise model,” SIAM Journal on Imaging Sciences , vol. 1, no. 3, pp. 294–321, 2008. 2, 13, 14

    2008

  7. [15]

    Iterative weighted maximum likelihood denoising with prob- abilistic patch-based weights,

    C.-A. Deledalle, L. Denis, and F. Tupin, “Iterative weighted maximum likelihood denoising with prob- abilistic patch-based weights,”IEEE transactions on Image Processing , vol. 18, no. 12, pp. 2661–2672,

  8. [16]

    A nonlocal sar image denoising algorithm based on llmmse wavelet shrinkage,

    S. Parrilli, M. Poderico, C. V. Angelino, and L. Verdoliva, “A nonlocal sar image denoising algorithm based on llmmse wavelet shrinkage,”IEEE Transactions on Geoscience and Remote Sensing , vol. 50, no. 2, pp. 606–616, 2011. 2

    2011

  9. [17]

    Deep learn- ing meets SAR: Concepts, models, pitfalls, and perspectives,

    X. X. Zhu, S. Montazeri, M. Ali, Y. Hua, Y. Wang, L. Mou, Y. Shi, F. Xu, and R. Bamler, “Deep learn- ing meets SAR: Concepts, models, pitfalls, and perspectives,”IEEE Geoscience and Remote Sensing Magazine, vol. 9, no. 4, pp. 143–172, 2021. 2

    2021

  10. [18]

    Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising,

    K. Zhang, W. Zuo, Y. Chen, D. Meng, and L. Zhang, “Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising,”IEEE transactions on image processing , vol. 26, no. 7, pp. 3142–3155,

  11. [19]

    Sar image despeckling through convolutional neural networks,

    G. Chierchia, D. Cozzolino, G. Poggi, and L. Verdoliva, “Sar image despeckling through convolutional neural networks,” in 2017 IEEE international geoscience and remote sensing symposium (IGARSS) . IEEE, 2017, pp. 5438–5441. 2

    2017

  12. [20]

    Sar image despeckling using a convolutional neural network,

    P. Wang, H. Zhang, and V. M. Patel, “Sar image despeckling using a convolutional neural network,” IEEE Signal Processing Letters , vol. 24, no. 12, pp. 1763–1767, 2017. 2

    2017

  13. [21]

    Noise2Noise: Learning image restoration without clean data,

    J. Lehtinen, J. Munkberg, J. Hasselgren, S. Laine, T. Karras, M. Aittala, and T. Aila, “Noise2Noise: Learning image restoration without clean data,” inProceedings of the 35th International Conference on Machine Learning, vol. 80. PMLR, 2018, pp. 2965–2974. 2

    2018

  14. [22]

    Noise2void-learning denoising from single noisy images,

    A. Krull, T.-O. Buchholz, and F. Jug, “Noise2void-learning denoising from single noisy images,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2019, pp. 2129–

    2019

  15. [23]

    Sar2sar: A semi-supervised despeckling algorithm for sar images,

    E. Dalsasso, L. Denis, and F. Tupin, “Sar2sar: A semi-supervised despeckling algorithm for sar images,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , vol. 14, pp. 4321– 4329, 2021. 2

    2021

  16. [24]

    Speckle2void: Deep self-supervised sar despeck- ling with blind-spot convolutional neural networks,

    A. B. Molini, D. Valsesia, G. Fracastoro, and E. Magli, “Speckle2void: Deep self-supervised sar despeck- ling with blind-spot convolutional neural networks,” IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–17, 2021. 2

    2021

  17. [25]

    Deeplsr: a deep learning approach for laser speckle reduction,

    T. L. Bobrow, F. Mahmood, M. Inserni, and N. J. Durr, “Deeplsr: a deep learning approach for laser speckle reduction,” Biomedical optics express, vol. 10, no. 6, pp. 2869–2882, 2019. 2

    2019

  18. [26]

    SDDPM: Speckle denoising diffusion probabilistic models,

    S. Guha and S. T. Acton, “SDDPM: Speckle denoising diffusion probabilistic models,”arXiv preprint arXiv:2311.10868, 2023. 2

    2023 arXiv

  19. [27]

    Bayesian denoising of structured sources and its implications on learning-based denoising,

    W. Zhou, J. Wabnig, and S. Jalali, “Bayesian denoising of structured sources and its implications on learning-based denoising,” Information and Inference: A Journal of the IMA , vol. 12, no. 4, pp. 2503– 2545, 2023. 2, 3, 4, 5

    2023

  20. [28]

    Rényi information dimension: Fundamental limits of almost lossless analog compression,

    Y. Wu and S. Verdú, “Rényi information dimension: Fundamental limits of almost lossless analog compression,” IEEE Transactions on Information Theory , vol. 56, no. 8, pp. 3721–3748, 2010. 4

    2010

  21. [29]

    Universal compressed sensing for almost lossless recovery,

    S. Jalali and H. V. Poor, “Universal compressed sensing for almost lossless recovery,”IEEE Transactions on Information Theory , vol. 63, no. 5, pp. 2933–2953, 2017. 4, 12

    2017

  22. [30]

    Tailboundsforsumsofgeometricandexponentialvariables,

    S.Janson, “Tailboundsforsumsofgeometricandexponentialvariables,” Statistics & Probability Letters, vol. 135, pp. 1–6, 2018. 6

    2018

  23. [31]

    Topsøe,Some Bounds for the Logarithmic Function

    F. Topsøe,Some Bounds for the Logarithmic Function . United States: Nova Science Publishers, 2007, vol. 4, pp. 137–151. 10

    2007

  24. [32]

    The viterbi algorithm,

    G. D. Forney, “The viterbi algorithm,”Proceedings of the IEEE, vol. 61, no. 3, pp. 268–278, 1973. 11

    1973

  25. [33]

    A model for radar images and its application to adaptive digital filtering of multiplicative noise,

    V. S. Frost, J. A. Stiles, K. S. Shanmugan, and J. C. Holtzman, “A model for radar images and its application to adaptive digital filtering of multiplicative noise,”IEEE Transactions on Pattern Analysis and Machine Intelligence , no. 2, pp. 157–166, 1982. 14 17

    1982

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.