Radio astronomical images object detection and segmentation: a benchmark on deep learning methods

In recent years, deep learning has been successfully applied in various scientific domains. Following these promising results and performances, it recently also started being evaluated the domain of radio astronomy. particular, since astronomy is entering Big Data era, with advent largest telescope world - Square Kilometre Array (SKA), task automatic object detection instance segmentation crucial for source finding analysis. this work, we explore performance most affirmed approaches, to astronomical images obtained by interferometric instrumentation, solve detection. This carried out applying models designed accomplish two different kinds tasks: semantic segmentation. The goal provide an overview existing techniques, terms prediction computational efficiency, scientists astrophysics community who would like employ machine their research.