Image reconstruction algorithms in radio interferometry: from handcrafted to learned regularization denoisers

Abstract

We introduce a new class of iterative image reconstruction algorithms for radio interferometry, at the interface of convex optimization and deep learning, inspired by plug-and-play methods. The approach consists in learning a prior image model by training a deep neural network (DNN) as a denoiser, and substituting it for the handcrafted proximal regularization operator of an optimization algorithm. The proposed AIRI (``AI for Regularization in radio-interferometric Imaging'') framework, for imaging complex intensity structure with diffuse and faint emission from visibility data, inherits the robustness and interpretability of optimization, and the learning power and speed of networks. Our approach relies on three steps. Firstly, we design a low dynamic range training database from optical intensity images. Secondly, we train a DNN denoiser at a noise level inferred from the signal-to-noise ratio of the data. We use training losses enhanced with a nonexpansiveness term ensuring algorithm convergence, and including on-the-fly database dynamic range enhancement via exponentiation. Thirdly, we plug the learned denoiser into the forward-backward optimization algorithm, resulting in a simple iterative structure alternating a denoising step with a gradient-descent data-fidelity step. We have validated AIRI against CLEAN, optimization algorithms of the SARA family, and a DNN trained to reconstruct the image directly from visibility data. Simulation results show that AIRI is competitive in imaging quality with SARA and its unconstrained forward-backward-based version uSARA, while providing significant acceleration. CLEAN remains faster but offers lower quality. The end-to-end DNN offers further acceleration, but with far lower quality than AIRI.

arXiv commentTo match revision for MNRAS publication. The new version includes the development of, and benchmarking with, a pure end-to-end DNN approach

References in corpus (33)

• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• First M87 Event Horizon Telescope Results. IV. Imaging the Central Supermassive Black Hole
• Enzo: An Adaptive Mesh Refinement Code for Astrophysics
• First Sagittarius A* Event Horizon Telescope Results. III: Imaging of the Galactic Center Supermassive Black Hole
• CNN-Based Projected Gradient Descent for Consistent Image Reconstruction
• The non-coplanar baselines effect in radio interferometry: The W-Projection algorithm
• An optimized algorithm for multi-scale wideband deconvolution of radio astronomical images
• Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction
• Compressed sensing imaging techniques for radio interferometry
• Plug and play methods for magnetic resonance imaging (long version)
• Sparsity Averaging Reweighted Analysis (SARA): a novel algorithm for radio-interferometric imaging
• PURIFY: a new approach to radio-interferometric imaging
• The application of compressive sampling to radio astronomy I: Deconvolution
• LOFAR Sparse Image Reconstruction
• Scale sensitive deconvolution of interferometric images I. Adaptive Scale Pixel (Asp) decomposition
• Scalable splitting algorithms for big-data interferometric imaging in the SKA era
• Uncertainty quantification for radio interferometric imaging: I. proximal MCMC methods
• MORESANE: MOdel REconstruction by Synthesis-ANalysis Estimators. A sparse deconvolution algorithm for radio interferometric imaging
• Robust sparse image reconstruction of radio interferometric observations with purify
• Principles of image reconstruction in optical interferometry: tutorial
• Convolutional Proximal Neural Networks and Plug-and-Play Algorithms
• Higher-order total variation approaches and generalisations
• Deep Radio Interferometric Imaging with POLISH: DSA-2000 and weak lensing
• Unified Radio Interferometric Calibration and Imaging with Joint Uncertainty Quantification
• An accelerated splitting algorithm for radio-interferometric imaging: when natural and uniform weighting meet
• Non-convex optimization for self-calibration of direction-dependent effects in radio interferometric imaging
• Cygnus A super-resolved via convex optimisation from VLA data
• Wideband Super-resolution Imaging in Radio Interferometry via Low Rankness and Joint Average Sparsity Models (HyperSARA)
• Sparse interferometric Stokes imaging under polarization constraint (Polarized SARA)
• Polca SARA - Full polarization, direction-dependent calibration and sparse imaging for radio interferometry
• Cygnus A jointly calibrated and imaged via non-convex optimisation from VLA data
• The w-effect in interferometric imaging: from a fast sparse measurement operator to super-resolution

Cited by in corpus (15)

• First AI for deep super-resolution wide-field imaging in radio astronomy: unveiling structure in ESO 137--006
• Scalable precision wide-field imaging in radio interferometry: II. AIRI validated on ASKAP data
• Scalable precision wide-field imaging in radio interferometry: I. uSARA validated on ASKAP data
• Bayesian self-calibration and imaging in Very Long Baseline Interferometry
• Regularized Maximum Likelihood Image Synthesis and Validation for ALMA Continuum Observations of Protoplanetary Disks
• fast-resolve: Fast Bayesian Radio Interferometric Imaging
• PolyCLEAN: Atomic Optimization for Super-Resolution Imaging and Uncertainty Estimation in Radio Interferometry
• Deep network series for large-scale high-dynamic range imaging
• Deep learning-based deconvolution for interferometric radio transient reconstruction
• Toward a Robust R2D2 Paradigm for Radio-interferometric Imaging: Revisiting Deep Neural Network Training and Architecture
• A Decentralized Framework for Radio-interferometric Image Reconstruction
• An Unsupervised Learning Method for Radio Interferometry Deconvolution
• Equivariant plug-and-play image reconstruction
• Very-Long Baseline Interferometry Imaging with Closure Invariants using Conditional Image Diffusion
• Can AI Dream of Unseen Galaxies? Conditional Diffusion Model for Galaxy Morphology Augmentation