Auroral breakup detection in all-sky images by unsupervised learning
Due to a large number of automatic auroral camera systems on the ground, image data analysis requires more efficiency than what human expert visual inspection can provide. Furthermore, there is no solid consensus on how many different types or shapes exist in auroral displays. We report the first at...
Published in: | Annales Geophysicae |
---|---|
Main Authors: | , , , , , |
Format: | Text |
Language: | English |
Published: |
2024
|
Subjects: | |
Online Access: | https://doi.org/10.5194/angeo-42-103-2024 https://angeo.copernicus.org/articles/42/103/2024/ |
id |
ftcopernicus:oai:publications.copernicus.org:angeo116342 |
---|---|
record_format |
openpolar |
spelling |
ftcopernicus:oai:publications.copernicus.org:angeo116342 2024-06-23T07:50:46+00:00 Auroral breakup detection in all-sky images by unsupervised learning Partamies, Noora Dol, Bas Teissier, Vincent Juusola, Liisa Syrjäsuo, Mikko Mulders, Hjalmar 2024-04-25 application/pdf https://doi.org/10.5194/angeo-42-103-2024 https://angeo.copernicus.org/articles/42/103/2024/ eng eng doi:10.5194/angeo-42-103-2024 https://angeo.copernicus.org/articles/42/103/2024/ eISSN: 1432-0576 Text 2024 ftcopernicus https://doi.org/10.5194/angeo-42-103-2024 2024-06-13T01:24:45Z Due to a large number of automatic auroral camera systems on the ground, image data analysis requires more efficiency than what human expert visual inspection can provide. Furthermore, there is no solid consensus on how many different types or shapes exist in auroral displays. We report the first attempt to classify auroral morphological forms by an unsupervised learning method on an image set that contains both nightside and dayside aurora. We used 6 months of full-colour auroral all-sky images captured at a high-Arctic observatory on Svalbard, Norway, in 2019–2020. The selection of images containing aurora was performed manually. These images were then input into a convolutional neural network called SimCLR for feature extraction. The clustered and fused features resulted in 37 auroral morphological clusters. In the clustering of auroral image data with two different time resolutions, we found that the occurrence of 8 clusters strongly increased when the image cadence was high (24 s), while the occurrence of 14 clusters experienced little or no change with changes in input image cadence. We therefore investigated the temporal evolution of a group of eight “active aurora” clusters. Time periods for which this active aurora persisted for longer than two consecutive images with a maximum cadence of 6 min coincided with ground-magnetic deflections, and their occurrence was found to maximize around magnetic midnight. The active aurora onsets typically included vortical auroral structures and equivalent current patterns typical for substorms. Our findings therefore suggest that our unsupervised image clustering method can be used to detect auroral breakups in ground-based image datasets with a temporal accuracy determined by the image cadence. Text Arctic Svalbard Copernicus Publications: E-Journals Arctic Norway Svalbard Annales Geophysicae 42 1 103 115 |
institution |
Open Polar |
collection |
Copernicus Publications: E-Journals |
op_collection_id |
ftcopernicus |
language |
English |
description |
Due to a large number of automatic auroral camera systems on the ground, image data analysis requires more efficiency than what human expert visual inspection can provide. Furthermore, there is no solid consensus on how many different types or shapes exist in auroral displays. We report the first attempt to classify auroral morphological forms by an unsupervised learning method on an image set that contains both nightside and dayside aurora. We used 6 months of full-colour auroral all-sky images captured at a high-Arctic observatory on Svalbard, Norway, in 2019–2020. The selection of images containing aurora was performed manually. These images were then input into a convolutional neural network called SimCLR for feature extraction. The clustered and fused features resulted in 37 auroral morphological clusters. In the clustering of auroral image data with two different time resolutions, we found that the occurrence of 8 clusters strongly increased when the image cadence was high (24 s), while the occurrence of 14 clusters experienced little or no change with changes in input image cadence. We therefore investigated the temporal evolution of a group of eight “active aurora” clusters. Time periods for which this active aurora persisted for longer than two consecutive images with a maximum cadence of 6 min coincided with ground-magnetic deflections, and their occurrence was found to maximize around magnetic midnight. The active aurora onsets typically included vortical auroral structures and equivalent current patterns typical for substorms. Our findings therefore suggest that our unsupervised image clustering method can be used to detect auroral breakups in ground-based image datasets with a temporal accuracy determined by the image cadence. |
format |
Text |
author |
Partamies, Noora Dol, Bas Teissier, Vincent Juusola, Liisa Syrjäsuo, Mikko Mulders, Hjalmar |
spellingShingle |
Partamies, Noora Dol, Bas Teissier, Vincent Juusola, Liisa Syrjäsuo, Mikko Mulders, Hjalmar Auroral breakup detection in all-sky images by unsupervised learning |
author_facet |
Partamies, Noora Dol, Bas Teissier, Vincent Juusola, Liisa Syrjäsuo, Mikko Mulders, Hjalmar |
author_sort |
Partamies, Noora |
title |
Auroral breakup detection in all-sky images by unsupervised learning |
title_short |
Auroral breakup detection in all-sky images by unsupervised learning |
title_full |
Auroral breakup detection in all-sky images by unsupervised learning |
title_fullStr |
Auroral breakup detection in all-sky images by unsupervised learning |
title_full_unstemmed |
Auroral breakup detection in all-sky images by unsupervised learning |
title_sort |
auroral breakup detection in all-sky images by unsupervised learning |
publishDate |
2024 |
url |
https://doi.org/10.5194/angeo-42-103-2024 https://angeo.copernicus.org/articles/42/103/2024/ |
geographic |
Arctic Norway Svalbard |
geographic_facet |
Arctic Norway Svalbard |
genre |
Arctic Svalbard |
genre_facet |
Arctic Svalbard |
op_source |
eISSN: 1432-0576 |
op_relation |
doi:10.5194/angeo-42-103-2024 https://angeo.copernicus.org/articles/42/103/2024/ |
op_doi |
https://doi.org/10.5194/angeo-42-103-2024 |
container_title |
Annales Geophysicae |
container_volume |
42 |
container_issue |
1 |
container_start_page |
103 |
op_container_end_page |
115 |
_version_ |
1802641688654512128 |