Using Deep Learning to Track Solar Magnetic Flux Elements

We present a new deep learning tool, called SolarUnet, to identify and track solar magnetic flux elements in observed vector magnetograms. This tool is applied to data from the 1.6 meter Goode Solar Telescope at the Big Bear Solar Observatory. In this poster we present the architecture of SolarUnet, which can be run in Jupyter Notebook, and some experimental results. To our knowledge, this is the first time that deep learning has been used for solar magnetic tracking.