Development of an Image Processing Technique Using Satellite Data to Study the Nature and Behaviour of Solar Coronal Mass Ejections

Abstract Coronal mass ejection (CME) is one mysterious property of the sun. Practically studying CMEs directed to earth, known as Halo CMEs is very important since these CMEs can affect space weather around the earth and cause major problems to human activities. This research aims to develop an image processing technique to analyze halo CME (HCME) events that can influence the earth by classifying associated solar flares based on coronagraph images. LASCO coronagraph telescope and EIT telescope of the “Solar and Heliospheric Observatory” (SOHO) spacecraft are the main instruments used for research. Coronagraph images and other additional data of 723 HCME events were collected from the database of the SOHO spacecraft of NASA & the database of the NOAA space environment services centre. A MATLAB script was developed to extract the detail of coronagraph images as pixel values using image processing methods such as image transformation, segmentation, Enhancement & morphological Processing. Finally, a classification algorithm was developed by using four K-Nearest Neighbor (KNN) classifiers to classify solar flares of each HCME event. The final accuracy of the classification algorithm was 86%. The developed method can be used as an alternative method to study HCMEs since the results of this study were agreeable with previous studies on CMEs. It is possible to improve this image processing method to calculate the effect of HCME events on earth, such as magnetic and space weather effects.