Master of Science Andreas Kvammen will on September 30 at 12.15 publically defend his thesis for the PhD degree in Science.
Title of the PhD thesis:
«Auroral Image Processing Techniques - Machine Learning Classification and Multi-Viewpoint Analysis»
Popular scientific abstract:
The Aurora Borealis, commonly called the Northern lights, is a spectacular display that can be observed in the Arctic sky. After the work of Kristian Birkeland and others, it is an established theory that the aurora is produced when charged particles, originating from the Sun, flow into the Earth's atmosphere. However, the physical processes that leads the solar particles towards the Earth is still an active field of research and millions of images are captured every year in order to study the auroral phenomena.
This large data set of auroral images is impossible to analyze by manual (human) inspection. This dissertation studies two branches of automatic or near-automatic computer processing techniques: machine learning classification and multi-viewpoint analysis.
The main conclusion of this work is that machine learning methods can be used to automatically classify auroral images and that multi-viewpoint analysis can be used to determine the height and distribution of the aurora.
Leader of the public defence and opposition ex auditorio:
Professor Torbjørn Eltoft, Department of Physics and Technology, will lead the public defence. If you have any questions for the candidate during the public defence, please send an e-mail to email@example.com. He will announce your questions during the defence.
Andreas Kvammen will hold his trial lecture for the PhD degree in Science on September 30 at 10.15 in the same auditorium. His given topic is:
«Recent advances in understanding auroral dynamics through the substorm cycle»