Efficient Bayesian Computation for Low-Photon Imaging Problems
From MaRDI portal
Publication:6168337
Bayesian inferencecomputational imaginginverse problemslow-photon imagingMarkov chain Monte Carlo methodsproximal algorithmsuncertainty quantification
Approximations to statistical distributions (nonasymptotic) (62E17) Bayesian inference (62F15) Parametric inference under constraints (62F30) Numerical analysis or methods applied to Markov chains (65C40) Numerical solution to inverse problems in abstract spaces (65J22) Computing methodologies for image processing (68U10)
Abstract: This paper studies a new and highly efficient Markov chain Monte Carlo (MCMC) methodology to perform Bayesian inference in low-photon imaging problems, with particular attention to situations involving observation noise processes that deviate significantly from Gaussian noise, such as binomial, geometric and low-intensity Poisson noise. These problems are challenging for many reasons. From an inferential viewpoint, low-photon numbers lead to severe identifiability issues, poor stability and high uncertainty about the solution. Moreover, low-photon models often exhibit poor regularity properties that make efficient Bayesian computation difficult; e.g., hard non-negativity constraints, non-smooth priors, and log-likelihood terms with exploding gradients. More precisely, the lack of suitable regularity properties hinders the use of state-of-the-art Monte Carlo methods based on numerical approximations of the Langevin stochastic differential equation (SDE), as both the SDE and its numerical approximations behave poorly. We address this difficulty by proposing an MCMC methodology based on a reflected and regularised Langevin SDE, which is shown to be well-posed and exponentially ergodic under mild and easily verifiable conditions. This then allows us to derive four reflected proximal Langevin MCMC algorithms to perform Bayesian computation in low-photon imaging problems. The proposed approach is demonstrated with a range of experiments related to image deblurring, denoising, and inpainting under binomial, geometric and Poisson noise.
Recommendations
- Bayesian Estimation for Optimized Structured Illumination Microscopy
- On the Bayesian approach to image reconstruction
- Fully Bayesian spectral methods for imaging data
- Bayesian Imaging Using Plug & Play Priors: When Langevin Meets Tweedie
- Fully Bayesian Reconstructions From Single-Photon Emission Computed Tomography Data
- Bayesian image analysis
- Scalable Bayesian uncertainty quantification in imaging inverse problems via convex optimization
- Bayesian Inference on Multiscale Models for Poisson Intensity Estimation: Applications to Photon-Limited Image Denoising
Cites work
- A convex approach for image restoration with exact Poisson-Gaussian likelihood
- A Proximal Iteration for Deconvolving Poisson Noisy Images Using Sparse Representations
- A simple proof of the Poincaré inequality for a large class of probability measures
- A symmetrized Euler scheme for an efficient approximation of reflected diffusions
- Accelerating Proximal Markov Chain Monte Carlo by Using an Explicit Stabilized Method
- An introduction to continuous optimization for imaging
- An Unbiased Risk Estimator for Image Denoising in the Presence of Mixed Poisson–Gaussian Noise
- Bayesian computation: a summary of the current state, and samples backwards and forwards
- Bayesian Imaging Using Plug & Play Priors: When Langevin Meets Tweedie
- Bayesian Imaging with Data-Driven Priors Encoded by Neural Networks
- Bayesian inference and uncertainty quantification for medical image reconstruction with Poisson data
- Bismut-Elworthy's formula and random walk representation for SDEs with reflection
- Brownian models of open queueing networks with homogeneous customer populations∗
- Collaborative sparse regression using spatially correlated supports - Application to hyperspectral unmixing
- Efficient Bayesian computation by proximal Markov chain Monte Carlo: when Langevin meets Moreau
- Ergodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise.
- Euler schemes and half-space approximation for the simulation of diffusion in a domain
- Expectation propagation for Poisson data
- Exponential convergence of Langevin distributions and their discrete approximations
- Exponential integrators
- Functional Integration and Partial Differential Equations. (AM-109)
- Generalized SURE for Exponential Families: Applications to Regularization
- Gradient-Based Markov Chain Monte Carlo for Bayesian Inference With Non-differentiable Priors
- scientific article; zbMATH DE number 3951715 (Why is no real title available?)
- scientific article; zbMATH DE number 2117879 (Why is no real title available?)
- scientific article; zbMATH DE number 7626757 (Why is no real title available?)
- Image Denoising in Mixed Poisson–Gaussian Noise
- Invariant density estimation for a reflected diffusion using an Euler scheme
- Inverse problems with Poisson data: statistical regularization theory, applications and algorithms
- Irreversible samplers from jump and continuous Markov processes
- Lidar Waveform-Based Analysis of Depth Images Constructed Using Sparse Single-Photon Data
- Long-time behaviour of degenerate diffusions: UFG-type SDEs and time-inhomogeneous hypoelliptic processes
- Markov chains and stochastic stability
- Maximum likelihood estimation of regularization parameters in high-dimensional inverse problems: an empirical Bayesian approach. I: Methodology and experiments
- Maximum-a-posteriori estimation with Bayesian confidence regions
- MCMC Algorithms for Computational UQ of Nonnegativity Constrained Linear Inverse Problems
- Numerical Approximation for Functionals of Reflecting Diffusion Processes
- Optimal explicit stabilized integrator of weak order 1 for stiff and ergodic stochastic differential equations
- Optimal Inversion of the Anscombe Transformation in Low-Count Poisson Image Denoising
- Patch-based image restoration using expectation propagation
- Penalization schemes for reflecting stochastic differential equations
- Probability with Martingales
- Proximal Markov chain Monte Carlo algorithms
- Proximal splitting methods in signal processing
- PURE-LET Image Deconvolution
- Restoration of Poissonian Images Using Alternating Direction Optimization
- Reversible and non-reversible Markov chain Monte Carlo algorithms for reservoir simulation problems
- Sampling from a log-concave distribution with projected Langevin Monte Carlo
- Scalable Bayesian uncertainty quantification in imaging inverse problems via convex optimization
- Simplest random walk for approximating Robin boundary value problems and ergodic limits of reflected diffusions
- Split-and-Augmented Gibbs Sampler—Application to Large-Scale Inference Problems
- Splitting methods
- Stability of Markovian processes III: Foster–Lyapunov criteria for continuous-time processes
- Stochastic differential equations with reflecting boundary condition in convex regions
- The Bayesian Choice
- This is SPIRAL-TAP: Sparse Poisson Intensity Reconstruction ALgorithms—Theory and Practice
- Variable metric forward-backward algorithm for minimizing the sum of a differentiable function and a convex function
- Variational Analysis
Cited in
(14)- Bayesian Smoothing of Photon-Limited Images with Applications in Astronomy
- Accuracy of the Bethe approximation for hyperparameter estimation in probabilistic image processing
- Efficient Bayesian computation by proximal Markov chain Monte Carlo: when Langevin meets Moreau
- A proximal Markov chain Monte Carlo method for Bayesian inference in imaging inverse problems: when Langevin meets Moreau
- Bayesian Imaging Using Plug & Play Priors: When Langevin Meets Tweedie
- Quick and energy-efficient Bayesian computing of binocular disparity using stochastic digital signals
- The Split Gibbs Sampler Revisited: Improvements to Its Algorithmic Structure and Augmented Target Distribution
- Accelerated Bayesian imaging by relaxed proximal-point Langevin sampling
- Marginal likelihood estimation in semiblind image deconvolution: a stochastic approximation approach
- Proximal Langevin sampling with inexact proximal mapping
- Empirical Bayesian image restoration by Langevin sampling with a denoising diffusion implicit prior
- Efficient Bayesian computation using plug-and-play priors for Poisson inverse problems
- Mixed-Noise Plug-and-Play with Infimal Convolution Fidelities and Multiple Priors
- Piecewise deterministic sampling for constrained distributions
This page was built for publication: Efficient Bayesian Computation for Low-Photon Imaging Problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6168337)
