A Detection and Tracking Method for Active Regions Based on Deep Learning

Active regions (ARs) are the primary source of energy for various solar activities. The violent solar activities adversely affect human living environment. Therefore, accurate detection and tracking of ARs are very important for monitoring and forecasting the space weather. In this paper, we propose an AR detection and tracking method (ARDTM) based on the deep learning model comprising of the YOLOv3-spp and DeepSort. The method solves the problem that one AR is mis-detected as multiple ARs, or multiple ARs are mis-detected as an AR. Besides, it captures new ARs and terminates disappeared ARs in time. The method improves the precision of detecting and tracking ARs. It can be used for detecting and tracking ARs in the solar full-disk longitudinal magnetograms observed from different telescopes, or images of different time interval series.