Skip to main content
SearchLoginLogin or Signup

Evaluating Machine Learning Approaches for Detecting Coronal Mass Ejections in Images

Presentation #131.01 in the session Improving Understanding of the Sun-Earth System Through Advanced Statistical and Machine Learning Techniques.

Published onOct 20, 2022
Evaluating Machine Learning Approaches for Detecting Coronal Mass Ejections in Images

The analysis of heliophysics data continues to present new challenges and traditional tools alone lack the capability to unlock all the potential science in heliophysics observations. Scientific investigation is often hampered by limitations including automated detection and extraction of physical observables of a specific phenomenon from data, the consequent development of large catalogs and quantifying how closely numerical simulations reproduce observed behavior. Machine learning (ML) approaches can help address these challenges.

In this presentation, we report on our work aimed at evaluating common computer vision algorithms and data processing approaches for their effectiveness at detecting Coronal Mass Ejections (CMEs) in white light coronagraph data from SOHO/LASCO C2 and C3 images. We find that some of these algorithms provide good performance when detecting CMEs.

Comments
0
comment

No comments here