Automatic Search for Giant Low-Surface-Brightness Galaxies

Giant galaxies of low surface brightness (gLSB) are difficult to detect due to the presence of a faint extended periphery, which is barely visible against the sky. At the same time, the study of such galaxies is necessary to clarify the mechanisms of formation and evolution of giant disk galaxies. To increase the number of known low surface brightness giant galaxies, it is planned to use machine learning models to solve the binary classification problem. To reduce the dimension of the problem, it is proposed to use as training data the radial brightness profiles of galaxies obtained by processing photometric fits images of galaxies in the HSC2 survey. For this purpose, a system for in-line processing of galaxy images was developed and an isophot analysis of 26008 galaxies in the visually inspected early quadrant of the sky, including 27 gLSB and 13 giant disk galaxies, was carried out. All obtained intensity profiles were visualized on one graph, as a result of which a criterion for selecting galaxies was formulated: for potentially interesting objects, the signal-to-noise ratio at a distance of 30 kpc from the center must be at least 2 in the g and r filters in order for the extended periphery to be clearly detected against the sky.