Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images
Abstract We report a novel machine‐learning algorithm for automatically detecting and classifying aurora in all–sky images (ASI) that is largely trained without requiring ground–truth labels. By including a small number of labeled images, we are able to automatically label all of the approximately 7...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-12-01
|
| Series: | Journal of Geophysical Research: Machine Learning and Computation |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2024JH000292 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850066456684789760 |
|---|---|
| author | Jeremiah W. Johnson Doğacan Su Öztürk Donald Hampton Hyunju K. Connor Matthew Blandin Amy Keesee |
| author_facet | Jeremiah W. Johnson Doğacan Su Öztürk Donald Hampton Hyunju K. Connor Matthew Blandin Amy Keesee |
| author_sort | Jeremiah W. Johnson |
| collection | DOAJ |
| description | Abstract We report a novel machine‐learning algorithm for automatically detecting and classifying aurora in all–sky images (ASI) that is largely trained without requiring ground–truth labels. By including a small number of labeled images, we are able to automatically label all of the approximately 700 million images in the Time History of Events and Macroscale Interactions during Substorms (THEMIS) ASI data set from 2008 to 2022. We use a two–stage approach. In the first stage, we adapt the Simple framework for Contrastive Learning of Representations (SimCLR) algorithm to learn latent representations of THEMIS all–sky images. We then finetune a classifier network on the latent representations our model learns of the manually labeled Oslo aurora THEMIS (OATH) data set. We demonstrate that this two–stage approach achieves excellent classification results on data for which there is no current ML classification benchmark. The outcome of this work will facilitate efficient information retrieval for researchers interested in specific categories of aurora and will enable large scale statistical studies and machine learning analyses of THEMIS all–sky images that have not previously been possible. To demonstrate possible ways to utilize this database, we performed a statistical analysis of the occurrence rates of auroral labels with respect to solar wind parameters, interplanetary magnetic field vector, and geomagnetic indices. We further investigate the occurrence rates of auroral phenomena in the annotated data set and their geoeffectiveness by utilizing the co–located THEMIS ground magnetometer data set. |
| format | Article |
| id | doaj-art-2bc09a8f3ce4404f979fbfaeb719d176 |
| institution | DOAJ |
| issn | 2993-5210 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Wiley |
| record_format | Article |
| series | Journal of Geophysical Research: Machine Learning and Computation |
| spelling | doaj-art-2bc09a8f3ce4404f979fbfaeb719d1762025-08-20T02:48:45ZengWileyJournal of Geophysical Research: Machine Learning and Computation2993-52102024-12-0114n/an/a10.1029/2024JH000292Automatic Detection and Classification of Aurora in THEMIS All‐Sky ImagesJeremiah W. Johnson0Doğacan Su Öztürk1Donald Hampton2Hyunju K. Connor3Matthew Blandin4Amy Keesee5Department of Applied Engineering & Sciences University of New Hampshire Manchester NH USAThe Geophysical Institute University of Alaska‐Fairbanks Fairbanks AK USAThe Geophysical Institute University of Alaska‐Fairbanks Fairbanks AK USANASA Goddard Space Flight Center Greenbelt MD USAThe Geophysical Institute University of Alaska‐Fairbanks Fairbanks AK USADepartment of Physics & Astronomy University of New Hampshire Durham NH USAAbstract We report a novel machine‐learning algorithm for automatically detecting and classifying aurora in all–sky images (ASI) that is largely trained without requiring ground–truth labels. By including a small number of labeled images, we are able to automatically label all of the approximately 700 million images in the Time History of Events and Macroscale Interactions during Substorms (THEMIS) ASI data set from 2008 to 2022. We use a two–stage approach. In the first stage, we adapt the Simple framework for Contrastive Learning of Representations (SimCLR) algorithm to learn latent representations of THEMIS all–sky images. We then finetune a classifier network on the latent representations our model learns of the manually labeled Oslo aurora THEMIS (OATH) data set. We demonstrate that this two–stage approach achieves excellent classification results on data for which there is no current ML classification benchmark. The outcome of this work will facilitate efficient information retrieval for researchers interested in specific categories of aurora and will enable large scale statistical studies and machine learning analyses of THEMIS all–sky images that have not previously been possible. To demonstrate possible ways to utilize this database, we performed a statistical analysis of the occurrence rates of auroral labels with respect to solar wind parameters, interplanetary magnetic field vector, and geomagnetic indices. We further investigate the occurrence rates of auroral phenomena in the annotated data set and their geoeffectiveness by utilizing the co–located THEMIS ground magnetometer data set.https://doi.org/10.1029/2024JH000292auroraannotationmachine learningTHEMISall‐sky images |
| spellingShingle | Jeremiah W. Johnson Doğacan Su Öztürk Donald Hampton Hyunju K. Connor Matthew Blandin Amy Keesee Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images Journal of Geophysical Research: Machine Learning and Computation aurora annotation machine learning THEMIS all‐sky images |
| title | Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images |
| title_full | Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images |
| title_fullStr | Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images |
| title_full_unstemmed | Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images |
| title_short | Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images |
| title_sort | automatic detection and classification of aurora in themis all sky images |
| topic | aurora annotation machine learning THEMIS all‐sky images |
| url | https://doi.org/10.1029/2024JH000292 |
| work_keys_str_mv | AT jeremiahwjohnson automaticdetectionandclassificationofaurorainthemisallskyimages AT dogacansuozturk automaticdetectionandclassificationofaurorainthemisallskyimages AT donaldhampton automaticdetectionandclassificationofaurorainthemisallskyimages AT hyunjukconnor automaticdetectionandclassificationofaurorainthemisallskyimages AT matthewblandin automaticdetectionandclassificationofaurorainthemisallskyimages AT amykeesee automaticdetectionandclassificationofaurorainthemisallskyimages |