Deep learning VLBI image reconstruction with closure invariants
Monthly Notices of the Royal Astronomical Society · 04.12.2024 · 536 (1) · 446-461
DOI: 10.1093/mnras/stae2607 ↗Kurz gefasst
Fachartikel in der Zeitschrift Monthly Notices of the Royal Astronomical Society, online erschienen am 22.11.2024. Verfasst von Samuel Lai und 4 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 83 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Interferometric closure invariants, constructed from triangular loops of mixed Fourier components, capture calibration-independent information on source morphology. While a complete set of closure invariants is directly obtainable from measured visibilities, the inverse transformation from closure invariants to the source intensity distribution is not established. In this work, we demonstrate a deep learning approach, Deep learning Image Reconstruction with Closure Terms (DIReCT), to directly reconstruct the image from closure invariants. Trained on both well-defined mathematical shapes (two-dimensional Gaussians, discs, ellipses, m-rings) and natural images (CIFAR-10), the results from our specially designed model are insensitive to station-based corruptions and thermal noise. The median fidelity score between the reconstruction and the blurred ground truth achieved is $\gtrsim 0.9$ even for untrained morphologies, where a unit score denotes perfect reconstruction. In our validation tests, DIReCT’s results are comparable to other state-of-the-art deconvolution and regularized maximum-likelihood image reconstruction algorithms, with the advantage that DIR eCT does not require hand-tuned hyperparameters for each individual prediction. This independent approach shows promising results and offers a calibration-independent constraint on source morphology, ultimately complementing and improving the reliability of sparse very long baseline interferometry imaging results.
Zitieren
APA
Lai, S., Thyagarajan, N., Wong, O. I., Diakogiannis, F., & Hoefs, L. (2024). Deep learning VLBI image reconstruction with closure invariants. Monthly Notices of the Royal Astronomical Society, 536(1), 446-461. https://doi.org/10.1093/mnras/stae2607
Vancouver
Lai S, Thyagarajan N, Wong OI, Diakogiannis F, Hoefs L. Deep learning VLBI image reconstruction with closure invariants. Monthly Notices of the Royal Astronomical Society. 2024;536(1):446-461. doi:10.1093/mnras/stae2607
BibTeX
@article{Lai2024,
author = {Lai, Samuel and Thyagarajan, Nithyanandan and Wong, O Ivy and Diakogiannis, Foivos and Hoefs, Lucas},
title = {Deep learning VLBI image reconstruction with closure invariants},
journal = {Monthly Notices of the Royal Astronomical Society},
year = {2024},
volume = {536},
number = {1},
pages = {446--461},
doi = {10.1093/mnras/stae2607}
}
EUVIMED