Detection and Tracking of Coronal Holes in Solar Images

Abstract

The study of solar activity and its effects on space weather is of great interest to humankind. Whether to study the dynamic of the star itself or the resulting phenomena and associated con-sequences from it, every different feature of the Sun provides valuable data to perform these studies. Features of the Sun are, for the most part, studied individually. However, studying differ-ent events collectively may result in new conclusions and findings that can be of as much interest as the individual studies. The objectives for this dissertation is to complement a Coronal Bright Points (CBPs) tracking algorithm, previously developed by (Pires, 2018), with an additional feature: detection of Coronal Holes (CHs) and classification of CBPs regarding whether they are located inside or outside of CHs. The proposed methodology is fully performed in Python language. Different image pro-cessing operations are applied in order to obtain a good detection result. The pre-processing stage involves an automatic image intensity normalization. The CHs detection uses a simple blur-ring before a fixed-value threshold segmentation. A last post-processing step includes performing adjustments to the detection results, using a closing morphologic operator, filling holes and an object detection. The data gathered by both tools is at the end consolidated, so that a result on the classifi-cation of each CBP is obtained and lastly added to the database.