Discovery of the polar ring galaxies with deep learning

Polar ring galaxies (PRGs) play an important role in understanding the evolution of galaxies, especially as unique cases of gas accretion and merging process between early and late morphological galaxy types. Regardless of their spectacular shape, these objects are very few in number and hard to find. Most of them were visually discovered, and then several of them were photometrically validated and kinematically confirmed. The aim of our research is to create a catalogue of strong and good candidates for PRGs using existing catalogues of PRGs; develop an image-based approach with machine learning methods for the search and discovery of PRGs in a big sky survey; explore the capability of the CIGALE software for determining their multiwavelength properties. For the first time, we applied a deep learning method to the search for PRGs. We visually inspected galaxies from existing catalogues of PRGs to create a training sample based on high-quality SDSS images. Since the resulting training sample was extremely small (87 strong and good PRGs), we applied augmentation, image segmentation, and ensemble learning techniques. However, the most effective method was transfer learning, with its ability to enlarge the training sample by synthetic images generated by GALFIT. To examine the deep learning approach for finding new PRGs, we used the SDSS catalogue of galaxies at z < 0.1. The method with synthetic images showed that even with overtraining, we can 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). We also visually inspected the catalogue of the SDSS ring galaxies at z < 0.1 and discovered four PRGs among ∼2,200 ring galaxies (SDSS J095851.32+320422.9; SDSS J104211.05+234448.2; SDSS J162212.63+272032.2; SDSS J104600.10+090627.2). One of the discovered galaxies with transfer learning, SDSS J140644.42+471602.0, was studied with CIGALE software to determine its spectral energy distribution from IR to UV bands. The current SFR is 71 M_⊙ per year, although the lack of FUV data limits this estimate. The total stellar mass is 8.34times 10^ M_⊙. The predominance of an old stellar population (two-thirds of the total mass) suggests that this PRG is undergoing an interaction process. We also explored the nuclear activity and isolation neighbourhood of several PRGs. Finally, we present a catalogue of 179 visually inspected PRGs. Supplemented with new discovered objects, this catalogue will become quite useful for the convolutional neural network approach and theoretical studies. Our strategies represent valuable opportunities for future development of deep learning models to achieve more quantitative PRG identification in sky surveys.