Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR
Automated solutions for sea ice type classification from synthetic aperture (SAR) imagery offer an opportunity to monitor sea ice, unimpeded by cloud cover or the arctic night. However, there is a common struggle to obtain accurate classifications year round; particularly in the melt and freeze-up s...
Published in: | IEEE Transactions on Geoscience and Remote Sensing |
---|---|
Main Authors: | , , |
Format: | Other Non-Article Part of Journal/Newspaper |
Language: | English |
Published: |
IEEE - Institute of Electrical and Electronics Engineers
2022
|
Subjects: | |
Online Access: | https://elib.dlr.de/148457/ https://elib.dlr.de/148457/1/2002%20_IEEE_TGRS_Kortum_paper_SAR_revised.pdf https://doi.org/10.1109/TGRS.2022.3144731 |
id |
ftdlr:oai:elib.dlr.de:148457 |
---|---|
record_format |
openpolar |
spelling |
ftdlr:oai:elib.dlr.de:148457 2023-05-15T14:57:43+02:00 Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR Kortum, Karl Singha, Suman Spreen, Gunnar 2022-01-20 application/pdf https://elib.dlr.de/148457/ https://elib.dlr.de/148457/1/2002%20_IEEE_TGRS_Kortum_paper_SAR_revised.pdf https://doi.org/10.1109/TGRS.2022.3144731 en eng IEEE - Institute of Electrical and Electronics Engineers https://elib.dlr.de/148457/1/2002%20_IEEE_TGRS_Kortum_paper_SAR_revised.pdf Kortum, Karl und Singha, Suman und Spreen, Gunnar (2022) Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR. IEEE Transactions on Geoscience and Remote Sensing, 60, Seite 4408512. IEEE - Institute of Electrical and Electronics Engineers. doi:10.1109/TGRS.2022.3144731 <https://doi.org/10.1109/TGRS.2022.3144731>. ISSN 0196-2892. cc_by SAR-Signalverarbeitung Zeitschriftenbeitrag PeerReviewed 2022 ftdlr https://doi.org/10.1109/TGRS.2022.3144731 2023-03-20T00:16:29Z Automated solutions for sea ice type classification from synthetic aperture (SAR) imagery offer an opportunity to monitor sea ice, unimpeded by cloud cover or the arctic night. However, there is a common struggle to obtain accurate classifications year round; particularly in the melt and freeze-up seasons. During these seasons, the radar backscatter signal is affected by wet snow cover, obscuring information about underlying ice types. By using additional spatiotemporal contextual data and a combination of convolutional neural networks and a dense conditional random field, we can mitigate these problems and obtain a single classifier which is able to classify accurately at 3.5 m spatial resolution for five different classes of sea ice surface from October to May. During the near year-long drift of the MOSAiC expedition we collected satellite scenes of the same patch of Arctic pack ice with X-Band SAR with a revisit-time of less than a day on average. Combined with in-situ observations of the local ice properties this offers up the unprecedented opportunity to perform a detailed and quantitative assessment of the robustness of our classifier for level, deformed and heavily deformed ice. For these three classes, we can perform accurate classification with a probability > 95% and calculate a lower bound for the robustness between 85% and 88%. Other Non-Article Part of Journal/Newspaper Arctic Sea ice German Aerospace Center: elib - DLR electronic library Arctic IEEE Transactions on Geoscience and Remote Sensing 60 1 12 |
institution |
Open Polar |
collection |
German Aerospace Center: elib - DLR electronic library |
op_collection_id |
ftdlr |
language |
English |
topic |
SAR-Signalverarbeitung |
spellingShingle |
SAR-Signalverarbeitung Kortum, Karl Singha, Suman Spreen, Gunnar Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR |
topic_facet |
SAR-Signalverarbeitung |
description |
Automated solutions for sea ice type classification from synthetic aperture (SAR) imagery offer an opportunity to monitor sea ice, unimpeded by cloud cover or the arctic night. However, there is a common struggle to obtain accurate classifications year round; particularly in the melt and freeze-up seasons. During these seasons, the radar backscatter signal is affected by wet snow cover, obscuring information about underlying ice types. By using additional spatiotemporal contextual data and a combination of convolutional neural networks and a dense conditional random field, we can mitigate these problems and obtain a single classifier which is able to classify accurately at 3.5 m spatial resolution for five different classes of sea ice surface from October to May. During the near year-long drift of the MOSAiC expedition we collected satellite scenes of the same patch of Arctic pack ice with X-Band SAR with a revisit-time of less than a day on average. Combined with in-situ observations of the local ice properties this offers up the unprecedented opportunity to perform a detailed and quantitative assessment of the robustness of our classifier for level, deformed and heavily deformed ice. For these three classes, we can perform accurate classification with a probability > 95% and calculate a lower bound for the robustness between 85% and 88%. |
format |
Other Non-Article Part of Journal/Newspaper |
author |
Kortum, Karl Singha, Suman Spreen, Gunnar |
author_facet |
Kortum, Karl Singha, Suman Spreen, Gunnar |
author_sort |
Kortum, Karl |
title |
Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR |
title_short |
Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR |
title_full |
Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR |
title_fullStr |
Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR |
title_full_unstemmed |
Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR |
title_sort |
robust multi-seasonal ice classification from high resolution x-band sar |
publisher |
IEEE - Institute of Electrical and Electronics Engineers |
publishDate |
2022 |
url |
https://elib.dlr.de/148457/ https://elib.dlr.de/148457/1/2002%20_IEEE_TGRS_Kortum_paper_SAR_revised.pdf https://doi.org/10.1109/TGRS.2022.3144731 |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic Sea ice |
genre_facet |
Arctic Sea ice |
op_relation |
https://elib.dlr.de/148457/1/2002%20_IEEE_TGRS_Kortum_paper_SAR_revised.pdf Kortum, Karl und Singha, Suman und Spreen, Gunnar (2022) Robust Multi-Seasonal Ice Classification from High Resolution X-Band SAR. IEEE Transactions on Geoscience and Remote Sensing, 60, Seite 4408512. IEEE - Institute of Electrical and Electronics Engineers. doi:10.1109/TGRS.2022.3144731 <https://doi.org/10.1109/TGRS.2022.3144731>. ISSN 0196-2892. |
op_rights |
cc_by |
op_doi |
https://doi.org/10.1109/TGRS.2022.3144731 |
container_title |
IEEE Transactions on Geoscience and Remote Sensing |
container_volume |
60 |
container_start_page |
1 |
op_container_end_page |
12 |
_version_ |
1766329842866847744 |