Catalogue of inspected PRGs - CIPRG

Our training sample is extremely small - only 87 objects where are strong and good PRGs candidates. For the first time, we applied deep learning to search for PRGs. Augmentation, image segmentation, ensemble, and transfer learning techniques were applied to mitigate the problem of an insufficient dataset (87 strong and good PRGs).

To search for PRGs, we applied our models to the SDSS catalogue of galaxies at z<0.1 (Vavilova et al., 2022, 2022KNIT...28....3V). Transfer learning with synthetic GALFIT images demonstrated that, despite overtraining, we were able to find galaxies with a ring pattern. Our deep learning approach has resulted in the discovery of three PRGs (SDSS J140644.42+471602.0; SDSS J133650.48+492745.3; SDSS J095717.30+364953.5).

Four PRGs were discovered through the visual inspection of ~2200 galaxies from the Catalogue of the SDSS Ring galaxies at z<0.1 (SDSS J095851.32+320422.9; SDSS J104211.05+234448.2; SDSS J162212.63+272032.2; SDSS J104600.10+090627.2). One galaxy discovered with transfer learning, SDSS J140644.42+471602.0, was studied using CIGALE software to determine its spectral energy distribution from IR to UV bands. Finally, we present a catalogue of 179 galaxies, among them 87 objects are strong and good PRG candidates, 79 objects are weak candidates, seven objects are from our current research and six objects from new research of Skrybina et al., 2024 (arXiv:2406.13496), Akhil et al., 2024MNRAS.530.2907A, Freitas-Lemes et al. 2024Ap&SS.369...93F, and Stanonik et al., 2009ApJ...696L...6S.

This catalogue will be useful for the CNN approach and theoretical studies. Our strategies represent valuable opportunities for future development of deep learning models to increase PRG identification in sky surveys.