View: |
Part 1: Document Description
|
Citation |
|
---|---|
Title: |
Replication Data for: Auroral Image Classification with Deep Neural Networks |
Identification Number: |
doi:10.18710/SSA38J |
Distributor: |
DataverseNO |
Date of Distribution: |
2020-01-15 |
Version: |
3 |
Bibliographic 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 |
Citation |
|
Title: |
Replication Data for: Auroral Image Classification with Deep Neural Networks |
Identification Number: |
doi:10.18710/SSA38J |
Authoring Entity: |
Kvammen, Andreas (UiT The Arctic University of Norway) |
Wickstrøm, Kristoffer (UiT The Arctic University of Norway) |
|
McKay, Derek (NORCE Norwegian Research Centre) |
|
Partamies, Noora (University Centre in Svalbard) |
|
Other identifications and acknowledgements: |
Swedish Institute of Space Physics |
Producer: |
UiT The Arctic University of Norway |
Distributor: |
DataverseNO |
Distributor: |
UiT The Arctic University of Norway |
Access Authority: |
Kvammen, Andreas |
Access Authority: |
Wickstrøm, Kristoffer |
Depositor: |
Wickstrøm, Kristoffer Knutsen |
Date of Deposit: |
2020-01-03 |
Holdings Information: |
https://doi.org/10.18710/SSA38J |
Study Scope |
|
Keywords: |
Computer and Information Science, Mathematical Sciences, Physics, auroral images, auroral classification, convolutional neural networks, aurora dataset, deep learning |
Abstract: |
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.<br><br>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 |
Time Period: |
2009-01-01-2020-01-01 |
Date of Collection: |
2010-01-01-2019-01-01 |
Country: |
Sweden |
Geographic Coverage: |
Kiruna |
Kind of Data: |
Images |
Methodology and Processing |
|
Sources Statement |
|
Data Access |
|
Other Study Description Materials |
|
Related Studies |
|
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. |
|
Related Publications |
|
Citation |
|
Title: |
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 |
Identification Number: |
10.1029/2020JA027808 |
Bibliographic Citation: |
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 |
Citation |
|
Title: |
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. |
Identification Number: |
10.5194/gi-9-267-2020 |
Bibliographic Citation: |
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. |
Label: |
00_ReadMe.txt |
Text: |
ReadMe file where dataset and the files in repository is described. |
Notes: |
text/plain |
Label: |
Auroral Image Classification with Deep Neural Networks.ipynb |
Text: |
Google colab notebook to training ResNet 18 classifier on aurora dataset. |
Notes: |
application/x-ipynb+json |
Label: |
data.npz |
Text: |
File containing images and labels stored as numpy arrays. |
Notes: |
application/octet-stream |
Label: |
kagolicens.pdf |
Text: |
Data license. |
Notes: |
application/pdf |
Label: |
TestImages.zip |
Text: |
Zipped archive containing the test images of the dataset in .PNG format. The class of the image is included in the image title. |
Notes: |
application/zip |
Label: |
TrainingImages.zip |
Text: |
Zipped archive containing the training images of the dataset in .PNG format. The class of the image is included in the image title. |
Notes: |
application/zip |