Multi-branch classification of diffuse cluster radio emission
📄 arXiv:2607.28349 · 📥 PDF · 2026-07-30 · astro-ph.CO
Authors: Markus Bredberg [arXiv · scholar] , Emma Tolley [arXiv · scholar]
🕰 Orloj analysis
Tato práce zkoumá techniky strojového učení (scattering transform a squeeze-excitation attention) pro zlepšení detekce difúzní rádiové emise v kupách galaxií. Autoři benchmarkují tyto metody proti základnímu CNN na datech z LOFAR/Planck, přičemž zjišťují, že multi-branch architektury s normalizovaným ořezem obrazu jsou slibné pro budoucí éru SKA.
💡 Tato práce aplikuje pokročilé techniky strojového učení k vylepšení detekce difúzní rádiové emise v kupách galaxií, což má praktickou hodnotu pro zpracování budoucích dat z SKA.
✓ falsifiable, modest_claims
⚠ small_labelled_datasets
📄 Abstract
Context. Galaxy clusters sometimes host synchrotron radiation on scales of ~100 kpc to ~1 Mpc, with surface brightness only a few times the image noise. This diffuse cluster radio emission is a sensitive probe of magnetic fields and intracluster medium dynamics, but disentangling the underlying physical processes requires statistically large samples spanning a wide range of cluster masses, dynamical states, and redshifts, together with sufficient sensitivity to low-surface-brightness emission. Aims. We explore two techniques for improving detection of diffuse emission in galaxy cluster images, relative to a baseline classifier: the scattering transform (ST) and squeeze-excitation (SE) attention. Methods. We integrate an ST encoder into a dual-branch classifier (DualSSN) and a scattering network (ScatterNet). We incorporate SE attention into the DualSSN and dual-branch convolutional neural network (DualCSN). These classifiers are then benchmarked against a simple CNN, across ten image preprocessing configurations and three cropping strategies. Performance is evaluated on small labelled datasets from the second data release of the LOFAR two-metre sky survey overlapping with the second Planck catalogue of Sunyaev-Zel'dovich sources (LoTSS-DR2/PSZ2). Results. Alongside the multi-branch approach with SE and ST, cropping the image to a fixed number of telescope beams and uv- tapering (smoothing to a coarser angular resolution) improve classification performance, while stacking multiple preprocessed ver- sions of an image does not. Conclusions. Scattering-transform-based multi-branch architectures with beam-normalised cropping are a promising direction for diffuse emission classification in the SKA era.