LAUSR

This paper presents an automatic multi-band source cross-identification method based on deep learning to identify the hosts of extragalactic radio emission structures. The aim is to satisfy the increased demand for automatic radio source identification and analysis of large-scale survey data from the next generation radio facilities such as the Square Kilometre Array (SKA) and the Next Generation Very Large Array (ngVLA). We demonstrate a 98% overall accuracy in distinguishing QSOs, galaxies, and stars using their optical morphologies plus their corresponding mid-infrared color information by training and testing a convolutional neural network (CNN) on the Pan-STARRS imaging and WISE photometry data. Compared with the expert-evaluated sample, we show the effectiveness of our approach in multi-band cross-matching with 95% of the extended radio components correctly identified with their hosts. We find that a better radio core localization with methods such as the geodesic center can further increase the accuracy of locating the cores of systems with a complex radio structure such as the C-shaped radio galaxies and significantly boost the efficiency for host identification. The framework developed in this work can be used for analyzing data from future large-scale radio surveys.