Solar rotation speed by detecting and tracking of Coronal Bright Points

• Main Author: Coelho, André Manuel Fernandes
• Publication Date: 2017
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10362/34371
• topic: coronal bright points; solar image; image processing; segmentation algorithms; object tracking; GPL; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

Coronal Bright Points are one of many Solar manifestations that provide scientists evi-dences of its activity and are usually recognized by being small light dots, like scattered jewels. For many years these Bright Points have been overlooked due to another element of solar activity, sunspots, which drawn scientists full attention mainly because they were easier to detect. Never-theless, CBPs as a result of a clear distribution across all latitudes, provide better tracers to study Solar corona rotation. A literature review on CBPs detection and tracking unveiled limitations both in detection accuracy and lacking an automated image processing feature. The purpose of this dissertation was to present an alternative method for detecting CBPs using advanced image processing techniques and provide an automatic recognition software. The proposed methodology is divided into pre-processing methods, a segmentation section, post processing and a data evaluation approach to increase the CBP detection efficiency. As iden-tified by the study of the available data, pre-processing transformations were needed to ensure each image met certain specifications for future detection. The detection process includes a gra-dient based segmentation algorithm, previously developed for retinal image analysis, which is now successfully applied to this CBP case study. The outcome is the CBP list obtained by the detection algorithm which is then filtered and evaluated to remove false positives. To validate the proposed methodology, CBPs need to be tracked along time, to obtain the rotation of the Solar corona. Therefore, the images used in this study were taken from 19.3nm wavelength by the AIA 193 instrument on board of the Solar Dynamics Observatory (SDO) sat-ellite over 3 days during august 2010. These images allowed the perception of how CBPs angular rotation velocity not only depends on heliographic latitude, but also on other factors such as time. From the results obtained it was clear that the proposed methodology is an effective method to detect and track CBPs providing a consistent method for its detection.