Mapping Diffuse Radio Sources Using TUNA: A Transformer-Based Deep Learning Approach
arXiv:2507.11320 · doi:10.1093/mnras/staf1174
Abstract
Vision Transformers are used via a customized TransUNet architecture, which is a hybrid model combining Transformers into a U-Net backbone, to achieve precise, automated, and fast segmentation of radio astronomy data affected by calibration and imaging artifacts, addressing the identification of faint, diffuse radio sources. Trained on mock radio observations from numerical simulations, the network is applied to the LOFAR Two-meter Sky Survey data. It is then evaluated on key use cases, specifically megahalos and bridges between galaxy clusters, to assess its performance in targeting sources at different resolutions and at the sensitivity limits of the telescope. The network is capable of detecting low surface brightness radio emission without manual source subtraction or re-imaging. The results demonstrate its groundbreaking capability to identify sources that typically require reprocessing at resolutions 4-6 times lower than that of the input image, accurately capturing their morphology and ensuring detection completeness. This approach represents a significant advancement in accelerating discovery within the large datasets generated by next-generation radio telescopes.
16 pages, 10 figures, accepted for publication in Monthly Notices of the Royal Astronomical Society
References in corpus (26)
- Astropy: A Community Python Package for Astronomy
- LOFAR: The LOw-Frequency ARray
- The Murchison Widefield Array: the Square Kilometre Array Precursor at low radio frequencies
- WSClean: an implementation of a fast, generic wide-field imager for radio astronomy
- Enzo: An Adaptive Mesh Refinement Code for Astrophysics
- Gemini: A Family of Highly Capable Multimodal Models
- Clusters of galaxies : observational properties of the diffuse radio emission
- Cosmic rays in galaxy clusters and their non-thermal emission
- The LOFAR Two-metre Sky Survey -- V. Second data release
- Turbulence and Magnetic Fields in the Large Scale Structure of the Universe
- An optimized algorithm for multi-scale wideband deconvolution of radio astronomical images
- Australian Square Kilometre Array Pathfinder: I. System Description
- DeepSeek-V3 Technical Report
- Detecting shock waves in cosmological smoothed particle hydrodynamics simulations
- Radio signature of cosmological structure formation shocks
- A radio ridge connecting two galaxy clusters in a filament of the cosmic web
- The Planck clusters in the LOFAR sky. I. LoTSS-DR2: new detections and sample overview
- Second order Fermi reacceleration mechanisms and large scale synchrotron radio emission in intra-cluster bridges
- Galaxy clusters enveloped by vast volumes of relativistic electrons
- Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs
- Deep Learning Based Detection of Cosmological Diffuse Radio Sources
- The Planck clusters in the LOFAR sky VI. LoTSS-DR2: Properties of radio relics
- Simulations and observational tests of primordial magnetic fields from Cosmic Microwave Background constraints
- Synchrotron Emission on the Largest Scales: Radio Detection of the Cosmic-Web
- FIGARO Simulation: FIlaments & GAlactic RadiO Simulation
- Giant radio halos in galaxy clusters as probes of particle acceleration in turbulent regions