Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada
Arctic amplification has many impacts on sea-ice extent, thickness, and flux. It becomes critical to monitor sea-ice conditions at finer spatio-temporal resolution. We used a simple convolutional neural network (CNN) on the RADARSAT-2 dual-polarized ScanSAR wide archive available over Coronation Gul...
Main Authors: | , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
Zenodo
2023
|
Subjects: | |
Online Access: | https://doi.org/10.5281/zenodo.8350643 |
id |
ftzenodo:oai:zenodo.org:8350643 |
---|---|
record_format |
openpolar |
spelling |
ftzenodo:oai:zenodo.org:8350643 2024-09-15T18:02:51+00:00 Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada Montpetit, Benoit Deschamps, Benjamin King, Joshua Duffe, Jason 2023-09-19 https://doi.org/10.5281/zenodo.8350643 unknown Zenodo https://doi.org/10.5281/zenodo.8350642 https://doi.org/10.5281/zenodo.8350643 oai:zenodo.org:8350643 info:eu-repo/semantics/openAccess Open Government Licence - Canada https://open.canada.ca/en/open-government-licence-canada Canadian Journal of Remote Sensing, 49(1), (2023-09-19) Sea Ice Convolutional Neural Network Deep Learning Synthetic Aperture Radar info:eu-repo/semantics/article 2023 ftzenodo https://doi.org/10.5281/zenodo.835064310.5281/zenodo.8350642 2024-07-25T20:39:14Z Arctic amplification has many impacts on sea-ice extent, thickness, and flux. It becomes critical to monitor sea-ice conditions at finer spatio-temporal resolution. We used a simple convolutional neural network (CNN) on the RADARSAT-2 dual-polarized ScanSAR wide archive available over Coronation Gulf, Canada, to assess which SAR parameter improves model performances to classify sea ice from water on a large volume of data covering 11 years of ice and surface water conditions. An overall accuracy of 90.1% was achieved on 989 scenes of 100% ice cover or ice-free conditions. An accuracy of 86.3% was achieved on the last year of data (134 scenes) which was kept out of the training process to test the model on an independent dataset. A better accuracy is obtained at lower incidence angles and the HH polarization provides the most information to classify ice from water. To achieve the best accuracy, the incidence angle and the noise equivalent sigma-nought had to be included as input to the model. A comparison done with the ASI passive microwave product shows similar errors in total sea ice concentration when using the Canadian Ice Service regional charts as reference. Nonetheless, errors from both datasets differ and the CNN outputs show greater potential to reduce masked areas, given the better spatial resolution, enabling data classification closer to land and identify features not captured by the ASI dataset. This dataset has been processed from RADARSAT-2 image products and saved into a stacked Python Numpy array. These arrays are analysis ready data to train/test the CNN model used in the referenced publication. Original RADARSAT-2 image products could not be shared directly since they are government by a End-User Licence Agreement (EULA). "RADARSAT-2 Data and Products © MacDONALD, DETTWILER and \n ASSOCIATES LTD (2023) - All Rights Reserved" and " RADARSAT is an official mark of the Canadian Space Agency" Article in Journal/Newspaper Coronation Gulf Sea ice Zenodo |
institution |
Open Polar |
collection |
Zenodo |
op_collection_id |
ftzenodo |
language |
unknown |
topic |
Sea Ice Convolutional Neural Network Deep Learning Synthetic Aperture Radar |
spellingShingle |
Sea Ice Convolutional Neural Network Deep Learning Synthetic Aperture Radar Montpetit, Benoit Deschamps, Benjamin King, Joshua Duffe, Jason Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada |
topic_facet |
Sea Ice Convolutional Neural Network Deep Learning Synthetic Aperture Radar |
description |
Arctic amplification has many impacts on sea-ice extent, thickness, and flux. It becomes critical to monitor sea-ice conditions at finer spatio-temporal resolution. We used a simple convolutional neural network (CNN) on the RADARSAT-2 dual-polarized ScanSAR wide archive available over Coronation Gulf, Canada, to assess which SAR parameter improves model performances to classify sea ice from water on a large volume of data covering 11 years of ice and surface water conditions. An overall accuracy of 90.1% was achieved on 989 scenes of 100% ice cover or ice-free conditions. An accuracy of 86.3% was achieved on the last year of data (134 scenes) which was kept out of the training process to test the model on an independent dataset. A better accuracy is obtained at lower incidence angles and the HH polarization provides the most information to classify ice from water. To achieve the best accuracy, the incidence angle and the noise equivalent sigma-nought had to be included as input to the model. A comparison done with the ASI passive microwave product shows similar errors in total sea ice concentration when using the Canadian Ice Service regional charts as reference. Nonetheless, errors from both datasets differ and the CNN outputs show greater potential to reduce masked areas, given the better spatial resolution, enabling data classification closer to land and identify features not captured by the ASI dataset. This dataset has been processed from RADARSAT-2 image products and saved into a stacked Python Numpy array. These arrays are analysis ready data to train/test the CNN model used in the referenced publication. Original RADARSAT-2 image products could not be shared directly since they are government by a End-User Licence Agreement (EULA). "RADARSAT-2 Data and Products © MacDONALD, DETTWILER and \n ASSOCIATES LTD (2023) - All Rights Reserved" and " RADARSAT is an official mark of the Canadian Space Agency" |
format |
Article in Journal/Newspaper |
author |
Montpetit, Benoit Deschamps, Benjamin King, Joshua Duffe, Jason |
author_facet |
Montpetit, Benoit Deschamps, Benjamin King, Joshua Duffe, Jason |
author_sort |
Montpetit, Benoit |
title |
Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada |
title_short |
Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada |
title_full |
Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada |
title_fullStr |
Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada |
title_full_unstemmed |
Assessing the Parameterization of RADARSAT-2 Dual-polarized ScanSAR Scenes on the Accuracy of a Convolutional Neural Network for Sea Ice Classification: Case Study over Coronation Gulf, Canada |
title_sort |
assessing the parameterization of radarsat-2 dual-polarized scansar scenes on the accuracy of a convolutional neural network for sea ice classification: case study over coronation gulf, canada |
publisher |
Zenodo |
publishDate |
2023 |
url |
https://doi.org/10.5281/zenodo.8350643 |
genre |
Coronation Gulf Sea ice |
genre_facet |
Coronation Gulf Sea ice |
op_source |
Canadian Journal of Remote Sensing, 49(1), (2023-09-19) |
op_relation |
https://doi.org/10.5281/zenodo.8350642 https://doi.org/10.5281/zenodo.8350643 oai:zenodo.org:8350643 |
op_rights |
info:eu-repo/semantics/openAccess Open Government Licence - Canada https://open.canada.ca/en/open-government-licence-canada |
op_doi |
https://doi.org/10.5281/zenodo.835064310.5281/zenodo.8350642 |
_version_ |
1810440263604109312 |