GHELLER, ClaudioClaudioGHELLERANGELINELLI, MatteoMatteoANGELINELLIDolag, KlausKlausDolagSANVITALE, NicolettaNicolettaSANVITALEVAZZA, FrancoFrancoVAZZA2025-03-102025-03-10202397830313416631570-6591http://hdl.handle.net/20.500.12386/36620Deep Learning represents a promising, general purpose solution to process and analyse large, complex datasets in an automated and efficient fashion. We present the experience accomplished on radioastronomy and X-ray simulated astrophysical data, adopting two different approaches, Autoencoders and Fully Convolutional Networks. The former aims at denoising the data, the latter at detecting and identifying faint sources in noisy images. We give an overview of the main outcomes.STAMPAenConference paper10.1007/978-3-031-34167-0_262-s2.0-85175952815https://link.springer.com/chapter/10.1007/978-3-031-34167-0_26https://api.elsevier.com/content/abstract/scopus_id/851759528152023ASSP...60..129GFIS/05 - ASTRONOMIA E ASTROFISICA