HyperAIRI: a plug-and-play algorithm for precise hyperspectral image reconstruction in radio interferometry

2025-10-16

AI Summary

To address the challenge of high-resolution, high-dynamic-range 3D imaging for next-generation radio interferometers operating over wide bandwidths and generating massive data volumes, this paper proposes HyperAIRI—a novel framework integrating physics-informed deep priors with optimization-based reconstruction. Methodologically, HyperAIRI embeds a learnable power-law spectral denoiser—trained with nonexpansive constraints—into a multi-channel joint reconstruction pipeline, augmented by parallel update schemes and a dynamic-range-adaptive denoiser library. Jacobian regularization ensures algorithmic convergence, while spatial tiling enables scalable large-scale image reconstruction. The framework unifies plug-and-play (PnP) denoising, ℓ₂,₁-norm joint sparsity modeling, and nonexpansive training. Evaluated on both synthetic and real observational data, HyperAIRI consistently outperforms state-of-the-art methods—including Hyper-uSARA, WSClean (hyperspectral variant), and monochromatic AIRI/uSARA—delivering superior imaging fidelity, dynamic range, and robustness.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

The next-generation radio-interferometric (RI) telescopes require imaging algorithms capable of forming high-resolution high-dynamic-range images from large data volumes spanning wide frequency bands. Recently, AIRI, a plug-and-play (PnP) approach taking the forward-backward algorithmic structure (FB), has demonstrated state-of-the-art performance in monochromatic RI imaging by alternating a data-fidelity step with a regularisation step via learned denoisers. In this work, we introduce HyperAIRI, its hyperspectral extension, underpinned by learned hyperspectral denoisers enforcing a power-law spectral model. For each spectral channel, the HyperAIRI denoiser takes as input its current image estimate, alongside estimates of its two immediate neighbouring channels and the spectral index map, and provides as output its associated denoised image. To ensure convergence of HyperAIRI, the denoisers are trained with a Jacobian regularisation enforcing non-expansiveness. To accommodate varying dynamic ranges, we assemble a shelf of pre-trained denoisers, each tailored to a specific dynamic range. At each HyperAIRI iteration, the spectral channels of the target image cube are updated in parallel using dynamic-range-matched denoisers from the pre-trained shelf. The denoisers are also endowed with a spatial image faceting functionality, enabling scalability to varied image sizes. Additionally, we formally introduce Hyper-uSARA, a variant of the optimisation-based algorithm HyperSARA, promoting joint sparsity across spectral channels via the l2,1-norm, also adopting FB. We evaluate HyperAIRI's performance on simulated and real observations. We showcase its superior performance compared to its optimisation-based counterpart Hyper-uSARA, CLEAN's hyperspectral variant in WSClean, and the monochromatic imaging algorithms AIRI and uSARA.

Problem

• Develops hyperspectral image reconstruction for radio interferometry telescopes
• Enhances imaging precision using learned denoisers with spectral constraints
• Enables scalable high-dynamic-range imaging across wide frequency bands

Innovation

• Plug-and-play hyperspectral imaging with learned denoisers
• Spectral denoising using neighboring channels and power-law model
• Pre-trained denoiser shelf for varying dynamic ranges