Speaker
Description
Solar eruptive events are complex phenomena that typically include filament eruptions, flares, coronal mass ejections (CMEs) and compressive/shock waves. Coronal mass ejections (CMEs) are large expulsions of plasma and magnetic fields from the solar corona into the heliosphere. The dynamic behavior of solar prominences and filaments serves as a precursor to CMEs, which can disrupt Earth's magnetosphere and affect satellite communications. Analysis of the morphological changes and destabilization processes of filaments and prominences, captured in observational datasets, can help identify early warning signs of potential eruptions, while more precise CME identification and tracking can enhance existing Solar Energetic Particle (SEP) models. Previously, we developed algorithmic multi-instrument solar eruptive feature recognition and tracking methods and applied them to tracking coronal bright fronts and CMEs from the low corona out to 30 solar radii, using ground- (COSMO K-Coronagraph) and space-based (SDO/AIA, SOHO/LASCO C2 and C3) telescopic observations, followed by demonstration of data-driven image segmentation of CBF's based on SDO AIA data.
In this work, we extend this approach for on disk image segmentation of filaments using H-alha Kanzelhohe Observatory data. We discuss our approach to engineering training sets on real and synthetic data and present a few models for data-driven segmentation and tracking of solar filaments.
