Poisson Noise Reduction with Higher-Order Natural Image Prior Model

DOI10.1137/16M1072930MaRDI QIDQ3179613zbMATH OpenOpenAlexFDO

Authors Wensen Feng, Yunjin Chen, H. Qiao

Publication date 19 December 2016

Published in SIAM Journal on Imaging Sciences (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1609.05722

zbMATH Keywords

nonconvex optimizationPoisson denoisingfields of expertsanscombe root transformation

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Learning and adaptive systems in artificial intelligence (68T05) Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) (68T20) Nonconvex programming, global optimization (90C26) Computing methodologies for image processing (68U10) Variational inequalities (49J40) Inverse problems in optimal control (49N45)

Abstract: Poisson denoising is an essential issue for various imaging applications, such as night vision, medical imaging and microscopy. State-of-the-art approaches are clearly dominated by patch-based non-local methods in recent years. In this paper, we aim to propose a local Poisson denoising model with both structure simplicity and good performance. To this end, we consider a variational modeling to integrate the so-called Fields of Experts (FoE) image prior, that has proven an effective higher-order Markov Random Fields (MRF) model for many classic image restoration problems. We exploit several feasible variational variants for this task. We start with a direct modeling in the original image domain by taking into account the Poisson noise statistics, which performs generally well for the cases of high SNR. However, this strategy encounters problem in cases of low SNR. Then we turn to an alternative modeling strategy by using the Anscombe transform and Gaussian statistics derived data term. We retrain the FoE prior model directly in the transform domain. With the newly trained FoE model, we end up with a local variational model providing strongly competitive results against state-of-the-art non-local approaches, meanwhile bearing the property of simple structure. Furthermore, our proposed model comes along with an additional advantage, that the inference is very efficient as it is well-suited for parallel computation on GPUs. For images of size $512imes512$, our GPU implementation takes less than 1 second to produce state-of-the-art Poisson denoising performance.

Recommendations

• Improved Denoising via Poisson Mixture Modeling of Image Sensor Noise
• A nonlocal low rank model for Poisson noise removal
• Image Denoising in Mixed Poisson–Gaussian Noise
• Poisson noise reduction with non-local PCA
• Poisson image denoising based on fractional-order total variation
• Practical Poissonian-Gaussian Noise Modeling and Fitting for Single-Image Raw-Data
• Poisson shot noise removal by an oracular non-local algorithm
• Iterative reweighted total generalized variation based Poisson noise removal model
• Multilevel algorithm for a Poisson noise removal model with total-variation regularization
• Sparse Poisson Noisy Image Deblurring

Cites work

• scientific article; zbMATH DE number 515890 (Why is no real title available?)
• A Closed-Form Approximation of the Exact Unbiased Inverse of the Anscombe Variance-Stabilizing Transformation
• A discrepancy principle for Poisson data
• A novel sparsity reconstruction method from Poisson data for 3D bioluminescence tomography
• BM3D Frames and Variational Image Deblurring
• Bayesian Inference on Multiscale Models for Poisson Intensity Estimation: Applications to Photon-Limited Image Denoising
• Fast interscale wavelet denoising of Poisson-corrupted images
• Fields of experts
• Insights Into Analysis Operator Learning: From Patch-Based Sparse Models to Higher Order MRFs
• Introductory lectures on convex optimization. A basic course.
• Multiscale Poisson Intensity and Density Estimation
• Multiscale likelihood analysis and complexity penalized estimation.
• On the limited memory BFGS method for large scale optimization
• Optimal Inversion of the Anscombe Transformation in Low-Count Poisson Image Denoising
• Poisson noise reduction with non-local PCA
• Restoration of Poissonian Images Using Alternating Direction Optimization
• Solving Inverse Problems With Piecewise Linear Estimators: From Gaussian Mixture Models to Structured Sparsity
• Sparsity-Based Poisson Denoising With Dictionary Learning
• THE TRANSFORMATION OF POISSON, BINOMIAL AND NEGATIVE-BINOMIAL DATA
• The limiting distribution of a function of two independent random variables and its statistical application
• This is SPIRAL-TAP: Sparse Poisson Intensity Reconstruction ALgorithms—Theory and Practice
• Why least squares and maximum entropy? An axiomatic approach to inference for linear inverse problems
• iPiano: inertial proximal algorithm for nonconvex optimization

Cited in

(7)

• A noise-robust online convolutional coding model and its applications to Poisson denoising and image fusion
• Field of experts regularized nonlocal low rank matrix approximation for image denoising
• Medical image denoising using simple form of MMSE estimation in Poisson-Gaussian noise model
• A nonlocal low rank model for Poisson noise removal
• Poisson shot noise removal by an oracular non-local algorithm
• Poisson noise reduction with non-local PCA
• Image denoising via multiscale nonlinear diffusion models

Describes a project that uses

Uses Software

• BM3D
• SPIRAL
• iPiano
• L-BFGS

This page was built for publication: Poisson Noise Reduction with Higher-Order Natural Image Prior Model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3179613)