Abstract Details
| Name: Ravindra Pawase Affiliation: Maven Systems Pvt Ltd Conference ID: ASI2018_1620 Title : Automatic detection and classification of solar radio bursts using machine learning and deep learning techniques Authors and Co-Authors : K.Sasikumar Raja(1), Ravindra Pawase(2), Tomin James(1), Prasad Subramanian(1), Christian Monstein(3). 1 Indian Institute of Science Education and Research, Pashan, Pune - 411 008, India. 2 Maven Systems Pvt. Ltd., Pune - 411 021, India. 3 Institute for Particle Physics and Astrophysics, ETH Zurich, Switzerland. Abstract Type : Poster Abstract Category : Instrumentation and Techniques Abstract : The solar radio transient emissions or bursts are powerful diagnostics to probe the dynamical processes that occur in the solar corona. Radio bursts are classified based on their drifting speeds and morphology in the dynamic spectrum. Some of these bursts can be used as proxies for the space weather hazards. Statistical analysis of radio bursts provide clues in resolving the long-standing mysteries in the solar corona. The physical properties and their association with the solar flares and coronal mass ejections have to be studied thoroughly. The e-CALLISTO is the network of radio spectrometers distributed around the globe to monitor radio bursts from the solar corona. Using the archival data (observed 24 hours a day) and by making use of machine learning and deep learning techniques, our aim is to automatically identify and classify the type of radio bursts by pattern recognition and extract their physical properties. A statistical study of such plasma parameters plays a crucial role in addressing the above mentioned issues. We will present the features of the already developed image processing library called ‘pycallisto’ and demonstrate the developed algorithms with the preliminary results. |