Application of Image Processing on Segmentation of the EUV Solar Images

Recent trends in image processing are more focused on new technologies involved in the field of Astronomy, Astrophysics and Medical field applications. Image processing in astronomy applies different algorithms to extract scientific results from the observed data. Through image processing, coronal features of Sun are extracted and studied. The extraction of different coronal features and determination of their contributions to EUV irradiance variability are important in heliophysics, the Earth's climate and space weather applications. The objective of this research is to extract the coronal Active Regions (ARs), Coronal Holes (CHs) and Quiet Sun (QS) from the fulldisk calibrated daily images of 174 A observed from PROBA2/SWAP instrument for period of 6 months (October 2016 -March 2017). In image processing the enhancement and different segmentation techniques have been applied. Based on these feature extraction the Support Vector Machine (SVM) classification is implemented. The cumulative intensity values of all the features are derived. The contribution of these features to full-disk integrated EUV irradiance variability are estimated and found that the active regions contribute up to 60% whereas the quiet sun and coronal holes will be around 38% and 2% respectively. In addition, these results suggest that the variations in intensity of the various coronal features have to be taken into account in irradiance models. Furthermore the SVM classification method has been estimated and the result presents a 90 percent accuracy.