{"id":10899,"identifier":"SSA38J","persistentUrl":"https://doi.org/10.18710/SSA38J","protocol":"doi","authority":"10.18710","publisher":"DataverseNO","publicationDate":"2020-01-15","storageIdentifier":"S3://10.18710/SSA38J","datasetVersion":{"id":3217,"datasetId":10899,"datasetPersistentId":"doi:10.18710/SSA38J","storageIdentifier":"S3://10.18710/SSA38J","versionNumber":3,"versionMinorNumber":1,"versionState":"RELEASED","lastUpdateTime":"2023-09-28T22:38:38Z","releaseTime":"2023-09-28T22:38:38Z","createTime":"2023-08-04T16:08:59Z","publicationDate":"2020-01-15","citationDate":"2020-01-15","termsOfUse":"The files in this dataset may be reused under a CC0 1.0 Universal (CC0 1.0)\nPublic Domain Dedication (https://creativecommons.org/publicdomain/zero/1.0/), except for the file \"data.npz\", which is restricted as stated in the license agreement \"kagolicens.pdf\" attached to the dataset and attached below.
\n
\nThe image data archive is freely accessible at http://www2.irf.se/allsky/data.html, however, the users are obliged to contact the Kiruna Atmospheric and Geophysical Observatory before usage.
\n
\n
\n------------------------------------
\nLicense agreement for data obtained from Kiruna Atmospheric and Geophysical Observatory at the Swedish Institute of Space Physics
\n
\nKiruna Atmospheric and Geophysical Observatory (KAGO) is an organisational unit within Swedish Institute of Space Physics (IRF). Observing sites include: the Kiruna area, Jamton, Lycksele Ionospheric Observatory (LJO), Uppsala Ionospheric Observatory (UJO) and Tormestorp.
\n
\n1. IRF welcomes and encourages the use of KAGO data. It should be noted that to collect, archive and provide observatory data is a considerable work effort. The purpose of this license agreement is to ensure the highest possible scientific quality and that IRF and the involved staff get proper credit for their work.
\n
\n2. This license agreement is valid for all data sets and products originating from KAGO at IRF. Exception: If the data is obtained through an international network where IRF has a signed written agreement (For example: IMAGE, Intermagnet, WDC, etc.), the data agreement for that network (if any) supersedes this license agreement. Contact the Head of KAGO for further advice.
\n
\n3. IRF grants to the user a non-exclusive license to retrieve and use data sets and products from KAGO in accordance with this license.
\n
\n4. When the data is redistributed, the licensee has to ensure the acceptance of the license agreement by any third parties.
\n
\n5. This license also applies for derivative works containing or depending on data from KAGO.
\n
\n6. Downloading and using of data sets and products from KAGO is free of charge for scientific and other non-profit usage. Please note that the use or reproduction of data for commercial purpose requires prior written permission.
\n
\n7. Regardless of whether the data are quality controlled or not, KAGO and IRF do not accept any liability for the correctness and/or appropriate interpretation of the data. Interpretation should follow best scientific practice and is always the user’s responsibility. Correct and appropriate data interpretation is solely the responsibility of data users.
\n
\n8. IRF requires permission from the Head of KAGO if data are to be published in any media.
\n
\n9. In the case of scientific publication or presentation of KAGO data at scientific conferences:
\n(a) Always first contact the Head of KAGO for advice how to proceed.
\n(b) The instrument PI will provide scientific support to aid proper interpretation of the data.
\n(c) The Head of KAGO, instrument PI, or other persons contributing to the scientific work at IRF require an offer for co-authorship according to normal scientific publication rules. For minor contributions the PI might decide that an acknowledgement is more appropriate.
\n(d) When a scientific publication that includes data from IRF is accepted for publication, a complete literature reference together with a manuscript in final form must be sent to the Head of KAGO, to ensure that the bibliography of the use of KAGO data can be kept up to date.
\n(e) If a generally accepted method of making data citations is available, it should be used.
\n
\n10. Before using information obtained from KAGO special attention should be given to the date and time of the data and products being displayed. This information shall not be modified in content and then presented as official material. Please note that time is usually given as coordinated universal time (UTC).
\n
\n11. Users are requested to inform the Head of KAGO of any problems encountered with the data itself or the data provision. Such feedback will increase the availability and quality of the data.
","fileAccessRequest":true,"metadataBlocks":{"citation":{"displayName":"Citation Metadata","name":"citation","fields":[{"typeName":"title","multiple":false,"typeClass":"primitive","value":"Replication Data for: Auroral Image Classification with Deep Neural Networks"},{"typeName":"author","multiple":true,"typeClass":"compound","value":[{"authorName":{"typeName":"authorName","multiple":false,"typeClass":"primitive","value":"Kvammen, Andreas"},"authorAffiliation":{"typeName":"authorAffiliation","multiple":false,"typeClass":"primitive","value":"UiT The Arctic University of Norway"},"authorIdentifierScheme":{"typeName":"authorIdentifierScheme","multiple":false,"typeClass":"controlledVocabulary","value":"ORCID"},"authorIdentifier":{"typeName":"authorIdentifier","multiple":false,"typeClass":"primitive","value":"0000-0002-5511-4473"}},{"authorName":{"typeName":"authorName","multiple":false,"typeClass":"primitive","value":"Wickstrøm, Kristoffer"},"authorAffiliation":{"typeName":"authorAffiliation","multiple":false,"typeClass":"primitive","value":"UiT The Arctic University of Norway"},"authorIdentifierScheme":{"typeName":"authorIdentifierScheme","multiple":false,"typeClass":"controlledVocabulary","value":"ORCID"},"authorIdentifier":{"typeName":"authorIdentifier","multiple":false,"typeClass":"primitive","value":"0000-0003-1395-7154"}},{"authorName":{"typeName":"authorName","multiple":false,"typeClass":"primitive","value":"McKay, Derek"},"authorAffiliation":{"typeName":"authorAffiliation","multiple":false,"typeClass":"primitive","value":"NORCE Norwegian Research Centre"},"authorIdentifierScheme":{"typeName":"authorIdentifierScheme","multiple":false,"typeClass":"controlledVocabulary","value":"ORCID"},"authorIdentifier":{"typeName":"authorIdentifier","multiple":false,"typeClass":"primitive","value":"0000-0003-1052-1929"}},{"authorName":{"typeName":"authorName","multiple":false,"typeClass":"primitive","value":"Partamies, Noora"},"authorAffiliation":{"typeName":"authorAffiliation","multiple":false,"typeClass":"primitive","value":"University Centre in Svalbard"},"authorIdentifierScheme":{"typeName":"authorIdentifierScheme","multiple":false,"typeClass":"controlledVocabulary","value":"ORCID"},"authorIdentifier":{"typeName":"authorIdentifier","multiple":false,"typeClass":"primitive","value":"0000-0003-2536-9341"}}]},{"typeName":"datasetContact","multiple":true,"typeClass":"compound","value":[{"datasetContactName":{"typeName":"datasetContactName","multiple":false,"typeClass":"primitive","value":"Kvammen, Andreas"},"datasetContactAffiliation":{"typeName":"datasetContactAffiliation","multiple":false,"typeClass":"primitive","value":"UiT The Arctic University of Norway"},"datasetContactEmail":{"typeName":"datasetContactEmail","multiple":false,"typeClass":"primitive","value":"andreas.kvammen@uit.no"}},{"datasetContactName":{"typeName":"datasetContactName","multiple":false,"typeClass":"primitive","value":"Wickstrøm, Kristoffer"},"datasetContactAffiliation":{"typeName":"datasetContactAffiliation","multiple":false,"typeClass":"primitive","value":"UiT The Arctic University of Norway"},"datasetContactEmail":{"typeName":"datasetContactEmail","multiple":false,"typeClass":"primitive","value":"kristoffer.k.wickstrom@uit.no"}}]},{"typeName":"dsDescription","multiple":true,"typeClass":"compound","value":[{"dsDescriptionValue":{"typeName":"dsDescriptionValue","multiple":false,"typeClass":"primitive","value":"Results from a study of automatic aurora classification using machine learning techniques are presented. The aurora is the manifestation of physical phenomena in the ionosphere magnetosphere environment. Automatic classification of millions of auroral images from the Arctic and Antarctic is therefore an attractive tool for developing auroral statistics and for supporting scientists to study auroral images in an objective, organized and repeatable manner. Although previous studies have presented tools for detecting aurora, there has been a lack of tools for classifying aurora into subclasses with a high precision (>90%). This work considers seven auroral subclasses; breakup, colored, arcs-bands, discrete, patchy, edge and clear-faint. Five different deep neural network architectures have been tested along with the well known classification methods; k nearest neighbor (KNN) and support vector machine (SVM). A set of clean nighttime color auroral images, without ambiguous auroral forms, moonlight, twilight, clouds etc., were used for training and testing. The deep neural networks generally outperformed the KNN and SVM methods, and the ResNet-50 architecture achieved the highest performance with an average classification precision of 92%. Although the results indicate that high precision aurora classification is an attainable objective using deep neural networks, it is stressed that a common consensus of the auroral morphology and the criteria for each class needs.
The authors would like to thank Urban Brändström and the Swedish Institute of Space Physics for providing the original auroral image data. The image data archive is freely accessible at http://www2.irf.se/allsky/data.html, however, the users are obliged to contact the Kiruna Atmospheric and Geophysical Observatory before usage"},"dsDescriptionDate":{"typeName":"dsDescriptionDate","multiple":false,"typeClass":"primitive","value":"2020-01-03"}}]},{"typeName":"subject","multiple":true,"typeClass":"controlledVocabulary","value":["Computer and Information Science","Mathematical Sciences","Physics"]},{"typeName":"keyword","multiple":true,"typeClass":"compound","value":[{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"auroral images"}},{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"auroral classification"}},{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"convolutional neural networks"}},{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"aurora dataset"}},{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"deep learning"}}]},{"typeName":"publication","multiple":true,"typeClass":"compound","value":[{"publicationCitation":{"typeName":"publicationCitation","multiple":false,"typeClass":"primitive","value":"Kvammen, A., Wickstrøm, K., McKay, D., & Partamies, N. (2020). Auroral image classification with deep neural networks. Journal of Geophysical Research: Space Physics, 125, e2020JA027808. https://doi.org/10.1029/2020JA027808"},"publicationIDType":{"typeName":"publicationIDType","multiple":false,"typeClass":"controlledVocabulary","value":"doi"},"publicationIDNumber":{"typeName":"publicationIDNumber","multiple":false,"typeClass":"primitive","value":"10.1029/2020JA027808"},"publicationURL":{"typeName":"publicationURL","multiple":false,"typeClass":"primitive","value":"https://doi.org/10.1029/2020JA027808"}},{"publicationCitation":{"typeName":"publicationCitation","multiple":false,"typeClass":"primitive","value":"McKay, D. and Kvammen, A.: Auroral classification ergonomics and the implications for machine learning, Geoscientific Instrumentation, Methods and Data Systems, 9, 267–273, https://doi.org/10.5194/gi-9-267-2020, 2020."},"publicationIDType":{"typeName":"publicationIDType","multiple":false,"typeClass":"controlledVocabulary","value":"doi"},"publicationIDNumber":{"typeName":"publicationIDNumber","multiple":false,"typeClass":"primitive","value":"10.5194/gi-9-267-2020"},"publicationURL":{"typeName":"publicationURL","multiple":false,"typeClass":"primitive","value":"https://doi.org/10.5194/gi-9-267-2020"}}]},{"typeName":"language","multiple":true,"typeClass":"controlledVocabulary","value":["English"]},{"typeName":"producer","multiple":true,"typeClass":"compound","value":[{"producerName":{"typeName":"producerName","multiple":false,"typeClass":"primitive","value":"UiT The Arctic University of Norway"},"producerAbbreviation":{"typeName":"producerAbbreviation","multiple":false,"typeClass":"primitive","value":"UiT"},"producerURL":{"typeName":"producerURL","multiple":false,"typeClass":"primitive","value":"https://en.uit.no/"}}]},{"typeName":"productionPlace","multiple":true,"typeClass":"primitive","value":["Kiruna, Sweden"]},{"typeName":"contributor","multiple":true,"typeClass":"compound","value":[{"contributorType":{"typeName":"contributorType","multiple":false,"typeClass":"controlledVocabulary","value":"Data Collector"},"contributorName":{"typeName":"contributorName","multiple":false,"typeClass":"primitive","value":"Swedish Institute of Space Physics"}}]},{"typeName":"distributor","multiple":true,"typeClass":"compound","value":[{"distributorName":{"typeName":"distributorName","multiple":false,"typeClass":"primitive","value":"UiT The Arctic University of Norway"},"distributorAffiliation":{"typeName":"distributorAffiliation","multiple":false,"typeClass":"primitive","value":"UiT The Arctic University of Norway"},"distributorURL":{"typeName":"distributorURL","multiple":false,"typeClass":"primitive","value":"https://dataverse.no/dataverse/uit"}}]},{"typeName":"depositor","multiple":false,"typeClass":"primitive","value":"Wickstrøm, Kristoffer Knutsen"},{"typeName":"dateOfDeposit","multiple":false,"typeClass":"primitive","value":"2020-01-03"},{"typeName":"timePeriodCovered","multiple":true,"typeClass":"compound","value":[{"timePeriodCoveredStart":{"typeName":"timePeriodCoveredStart","multiple":false,"typeClass":"primitive","value":"2009-01-01"},"timePeriodCoveredEnd":{"typeName":"timePeriodCoveredEnd","multiple":false,"typeClass":"primitive","value":"2020-01-01"}}]},{"typeName":"dateOfCollection","multiple":true,"typeClass":"compound","value":[{"dateOfCollectionStart":{"typeName":"dateOfCollectionStart","multiple":false,"typeClass":"primitive","value":"2010-01-01"},"dateOfCollectionEnd":{"typeName":"dateOfCollectionEnd","multiple":false,"typeClass":"primitive","value":"2019-01-01"}}]},{"typeName":"kindOfData","multiple":true,"typeClass":"primitive","value":["Images"]},{"typeName":"relatedDatasets","multiple":true,"typeClass":"primitive","value":["The dataset contained in this repository was extracted from the All-Sky camera in Kiruna, Sweden, available at http://www2.irf.se/allsky/data.html."]}]},"geospatial":{"displayName":"Geospatial Metadata","name":"geospatial","fields":[{"typeName":"geographicCoverage","multiple":true,"typeClass":"compound","value":[{"country":{"typeName":"country","multiple":false,"typeClass":"controlledVocabulary","value":"Sweden"},"city":{"typeName":"city","multiple":false,"typeClass":"primitive","value":"Kiruna"}}]}]}},"files":[{"description":"ReadMe file where dataset and the files in repository is described.","label":"00_ReadMe.txt","restricted":false,"version":1,"datasetVersionId":3217,"dataFile":{"id":18931,"persistentId":"doi:10.18710/SSA38J/JIXSEM","pidURL":"https://doi.org/10.18710/SSA38J/JIXSEM","filename":"00_ReadMe.txt","contentType":"text/plain","filesize":4709,"description":"ReadMe file where dataset and the files in repository is described.","storageIdentifier":"S3://2002-yellow-dataverseno:17174358222-f1c70da0b376","rootDataFileId":-1,"md5":"6c25a3fdae5cc8db0938a055991b217e","checksum":{"type":"MD5","value":"6c25a3fdae5cc8db0938a055991b217e"},"creationDate":"2020-04-13"}},{"description":"Google colab notebook to training ResNet 18 classifier on aurora dataset.","label":"Auroral Image Classification with Deep Neural Networks.ipynb","restricted":false,"version":1,"datasetVersionId":3217,"dataFile":{"id":10902,"persistentId":"doi:10.18710/SSA38J/LIFQOS","pidURL":"https://doi.org/10.18710/SSA38J/LIFQOS","filename":"Auroral Image Classification with Deep Neural Networks.ipynb","contentType":"application/x-ipynb+json","filesize":8761,"description":"Google colab notebook to training ResNet 18 classifier on aurora dataset.","storageIdentifier":"S3://2002-yellow-dataverseno:16f6bb14f67-46b601f8454b","rootDataFileId":-1,"md5":"ddd34b7f1d5d9754a71b45d70af39b93","checksum":{"type":"MD5","value":"ddd34b7f1d5d9754a71b45d70af39b93"},"creationDate":"2020-01-03"}},{"description":"Zipped archive containing the test images of the dataset in .PNG format. The class of the image is included in the image title.","label":"TestImages.zip","restricted":false,"version":1,"datasetVersionId":3217,"dataFile":{"id":18930,"persistentId":"doi:10.18710/SSA38J/UWZZOI","pidURL":"https://doi.org/10.18710/SSA38J/UWZZOI","filename":"TestImages.zip","contentType":"application/zip","filesize":10646108,"description":"Zipped archive containing the test images of the dataset in .PNG format. The class of the image is included in the image title.","storageIdentifier":"S3://2002-yellow-dataverseno:17174324a44-a965c8b46f3c","rootDataFileId":-1,"md5":"b1ff172b00f06c86dd82c021683d451e","checksum":{"type":"MD5","value":"b1ff172b00f06c86dd82c021683d451e"},"creationDate":"2020-04-13"}},{"description":"Zipped archive containing the training images of the dataset in .PNG format. The class of the image is included in the image title.","label":"TrainingImages.zip","restricted":false,"version":1,"datasetVersionId":3217,"dataFile":{"id":18929,"persistentId":"doi:10.18710/SSA38J/WLVZQ4","pidURL":"https://doi.org/10.18710/SSA38J/WLVZQ4","filename":"TrainingImages.zip","contentType":"application/zip","filesize":37562412,"description":"Zipped archive containing the training images of the dataset in .PNG format. The class of the image is included in the image title.","storageIdentifier":"S3://2002-yellow-dataverseno:17174300699-acb45dffb12f","rootDataFileId":-1,"md5":"b27e63cd55a51e756d0c387a1b7be893","checksum":{"type":"MD5","value":"b27e63cd55a51e756d0c387a1b7be893"},"creationDate":"2020-04-13"}},{"description":"File containing images and labels stored as numpy arrays.","label":"data.npz","restricted":false,"version":1,"datasetVersionId":3217,"categories":["Data"],"dataFile":{"id":10900,"persistentId":"doi:10.18710/SSA38J/WGWRRE","pidURL":"https://doi.org/10.18710/SSA38J/WGWRRE","filename":"data.npz","contentType":"application/octet-stream","filesize":76439416,"description":"File containing images and labels stored as numpy arrays.","categories":["Data"],"storageIdentifier":"S3://2002-yellow-dataverseno:16f6b923cc4-ac11b0cafaa2","rootDataFileId":-1,"md5":"ee2755c2c09b8f2cf8395eae849a1484","checksum":{"type":"MD5","value":"ee2755c2c09b8f2cf8395eae849a1484"},"creationDate":"2020-01-03"}},{"description":"Data license.","label":"kagolicens.pdf","restricted":false,"version":1,"datasetVersionId":3217,"dataFile":{"id":11161,"persistentId":"doi:10.18710/SSA38J/QQ6HKG","pidURL":"https://doi.org/10.18710/SSA38J/QQ6HKG","filename":"kagolicens.pdf","contentType":"application/pdf","filesize":161297,"description":"Data license.","storageIdentifier":"S3://2002-yellow-dataverseno:16f8e7c5785-20b9f8835723","rootDataFileId":-1,"md5":"d5d42b2e7cf58c748b78d9771ffc3010","checksum":{"type":"MD5","value":"d5d42b2e7cf58c748b78d9771ffc3010"},"creationDate":"2020-01-10"}}],"citation":"Kvammen, Andreas; Wickstrøm, Kristoffer; McKay, Derek; Partamies, Noora, 2020, \"Replication Data for: Auroral Image Classification with Deep Neural Networks\", https://doi.org/10.18710/SSA38J, DataverseNO, V3"}}