Quantifying the fine structures of disk galaxies with deep learning:Segmentation of spiral arms in different Hubble types

Abstract

Spatial correlations between spiral arms and other galactic components such as giant molecular clouds and massive OB stars suggest that spiral arms can play vital roles in various aspects of disk galaxy evolution. Segmentation of spiral arms in disk galaxies is therefore a key task to investigate these correlations. We here try to decompose disk galaxies into spiral and non-spiral regions by using U-net, which is based on deep learning algorithms and has been invented for segmentation tasks in biology.