This is SPIRAL-TAP: Sparse Poisson Intensity Reconstruction ALgorithms - Theory and Practice

Abstract

The observations in many applications consist of counts of discrete events, such as photons hitting a detector, which cannot be effectively modeled using an additive bounded or Gaussian noise model, and instead require a Poisson noise model. As a result, accurate reconstruction of a spatially or temporally distributed phenomenon (f*) from Poisson data (y) cannot be effectively accomplished by minimizing a conventional penalized least-squares objective function. The problem addressed in this paper is the estimation of f* from y in an inverse problem setting, where (a) the number of unknowns may potentially be larger than the number of observations and (b) f* admits a sparse approximation. The optimization formulation considered in this paper uses a penalized negative Poisson log-likelihood objective function with nonnegativity constraints (since Poisson intensities are naturally nonnegative). In particular, the proposed approach incorporates key ideas of using separable quadratic approximations to the objective function at each iteration and penalization terms related to l1 norms of coefficient vectors, total variation seminorms, and partition-based multiscale estimation methods.

11 pages, 7 figures, IEEE Transactions on Image Processing (2011), in press

References in corpus (7)

• Model-Based Compressive Sensing
• A Proximal Decomposition Method for Solving Convex Variational Inverse Problems
• A proximal iteration for deconvolving Poisson noisy images using sparse representations
• Compressed sensing performance bounds under Poisson noise
• Multiscale likelihood analysis and complexity penalized estimation
• Sparse Poisson Intensity Reconstruction Algorithms
• Data-driven wavelet-Fisz methodology for nonparametric function estimation

Cited by in corpus (26)

• Photon-Efficient Computational 3D and Reflectivity Imaging with Single-Photon Detectors
• Single-photon computational 3D imaging at 45 km
• Super-resolution single-photon imaging at 8.2 kilometers
• Sparsity Based Poisson Denoising with Dictionary Learning
• Bayesian 3D Reconstruction of Complex Scenes from Single-Photon Lidar Data
• Revealing hidden scenes by photon-efficient occlusion-based opportunistic active imaging
• Faster PET Reconstruction with Non-Smooth Priors by Randomization and Preconditioning
• A new steplength selection for scaled gradient methods with application to image deblurring
• Ultraviolet photon-counting single-pixel imaging
• Projected Nesterov's Proximal-Gradient Algorithm for Sparse Signal Reconstruction with a Convex Constraint
• Plug-and-Play Quantum Adaptive Denoiser for Deconvolving Poisson Noisy Images
• A Hierarchical Bayesian Approach to Neutron Spectrum Unfolding with Organic Scintillators
• Reconstructing the Image Scanning Microscopy Dataset: an Inverse Problem
• Joint Reconstruction via Coupled Bregman Iterations with Applications to PET-MR Imaging
• Line-scan Compressive Raman imaging with spatio-spectral encoding
• Non-convex optimization for 3D point source localization using a rotating point spread function
• Signal retrieval with measurement system knowledge using variational generative model
• Few-photon computed x-ray imaging
• String-Averaging Incremental Subgradients for Constrained Convex Optimization with Applications to Reconstruction of Tomographic Images
• Poisson2Sparse: Self-Supervised Poisson Denoising From a Single Image
• Photon allocation strategy in region-of-interest tomographic imaging
• Extreme Compressed Sensing of Poisson Rates from Multiple Measurements
• Composite Convex Optimization with Global and Local Inexact Oracles
• Interference in Fuzzy Dark Matter Filaments: Idealised Models and Statistics
• Imaging with SPADs and DMDs: Seeing through Diffraction-Photons
• Sparse Signal Reconstruction for Overdispersed Low-photon Count Biomedical Imaging Using ${\ell }_p$ Total Variation