Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation
Permafrost tundra contains more than twice as much carbon as is currently in the atmosphere, and it is warming six times as fast as the global mean. Tundra lakes dynamics is a robust indicator of global climate processes, and is still not well understood. Satellite data, particularly, from synthetic...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , , |
Other Authors: | , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
MDPI AG
2023
|
Subjects: | |
Online Access: | http://hdl.handle.net/10754/689840 https://doi.org/10.3390/rs15051298 |
id |
ftkingabdullahun:oai:repository.kaust.edu.sa:10754/689840 |
---|---|
record_format |
openpolar |
spelling |
ftkingabdullahun:oai:repository.kaust.edu.sa:10754/689840 2024-01-07T09:45:59+01:00 Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation Demchev, Denis Sudakow, Ivan Khodos, Alexander Abramova, Irina Lyakhov, Dmitry Michels, Dominik L. Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division Computer Science Program Visual Computing Center (VCC) Department of Space, Earth and Environment, Chalmers University of Technology, 412 96 Gothenburg, Sweden School of Mathematics and Statistics, The Open University, Milton Keynes MK7 6AA, UK The Center for Research and Invention, Veliky Novgorod 173008, Russia Arctic and Antarctic Research Institute, Saint Petersburg 199397, Russia 2023-03-01T06:02:57Z application/pdf http://hdl.handle.net/10754/689840 https://doi.org/10.3390/rs15051298 unknown MDPI AG https://www.mdpi.com/2072-4292/15/5/1298 Demchev, D., Sudakow, I., Khodos, A., Abramova, I., Lyakhov, D., & Michels, D. (2023). Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation. Remote Sensing, 15(5), 1298. https://doi.org/10.3390/rs15051298 doi:10.3390/rs15051298 2072-4292 5 Remote Sensing 1298 http://hdl.handle.net/10754/689840 15 Archived with thanks to Remote Sensing under a Creative Commons license, details at: https://creativecommons.org/licenses/by/4.0/ https://creativecommons.org/licenses/by/4.0/ Article 2023 ftkingabdullahun https://doi.org/10.3390/rs15051298 2023-12-09T20:18:30Z Permafrost tundra contains more than twice as much carbon as is currently in the atmosphere, and it is warming six times as fast as the global mean. Tundra lakes dynamics is a robust indicator of global climate processes, and is still not well understood. Satellite data, particularly, from synthetic aperture radar (SAR) is a suitable tool for tundra lakes recognition and monitoring of their changes. However, manual analysis of lake boundaries can be slow and inefficient; therefore, reliable automated algorithms are required. To address this issue, we propose a two-stage approach, comprising instance deep-learning-based segmentation by U-Net, followed by semantic segmentation based on a watershed algorithm for separating touching and overlapping lakes. Implementation of this concept is essential for accurate sizes and shapes estimation of an individual lake. Here, we evaluated the performance of the proposed approach on lakes, manually extracted from tens of C-band SAR images from Sentinel-1, which were collected in the Yamal Peninsula and Alaska areas in the summer months of 2015–2022. An accuracy of 0.73, in terms of the Jaccard similarity index, was achieved. The lake’s perimeter, area and fractal sizes were estimated, based on the algorithm framework output from hundreds of SAR images. It was recognized as lognormal distributed. The evaluation of the results indicated the efficiency of the proposed approach for accurate automatic estimation of tundra lake shapes and sizes, and its potential to be used for further studies on tundra lake dynamics, in the context of global climate change, aimed at revealing new factors that could cause the planet to warm or cool. This research was funded by the Russian Science Foundation (RSF), project # 21-71-10052. D.L. and D.M. were partially supported by KAUST baseline funding. The authors thank Viktoria Kharchenko for manual labeling of part of the Sentinel-1 training and validation images Article in Journal/Newspaper permafrost Tundra Yamal Peninsula Alaska King Abdullah University of Science and Technology: KAUST Repository The Sentinel ENVELOPE(73.317,73.317,-52.983,-52.983) Yamal Peninsula ENVELOPE(69.873,69.873,70.816,70.816) Remote Sensing 15 5 1298 |
institution |
Open Polar |
collection |
King Abdullah University of Science and Technology: KAUST Repository |
op_collection_id |
ftkingabdullahun |
language |
unknown |
description |
Permafrost tundra contains more than twice as much carbon as is currently in the atmosphere, and it is warming six times as fast as the global mean. Tundra lakes dynamics is a robust indicator of global climate processes, and is still not well understood. Satellite data, particularly, from synthetic aperture radar (SAR) is a suitable tool for tundra lakes recognition and monitoring of their changes. However, manual analysis of lake boundaries can be slow and inefficient; therefore, reliable automated algorithms are required. To address this issue, we propose a two-stage approach, comprising instance deep-learning-based segmentation by U-Net, followed by semantic segmentation based on a watershed algorithm for separating touching and overlapping lakes. Implementation of this concept is essential for accurate sizes and shapes estimation of an individual lake. Here, we evaluated the performance of the proposed approach on lakes, manually extracted from tens of C-band SAR images from Sentinel-1, which were collected in the Yamal Peninsula and Alaska areas in the summer months of 2015–2022. An accuracy of 0.73, in terms of the Jaccard similarity index, was achieved. The lake’s perimeter, area and fractal sizes were estimated, based on the algorithm framework output from hundreds of SAR images. It was recognized as lognormal distributed. The evaluation of the results indicated the efficiency of the proposed approach for accurate automatic estimation of tundra lake shapes and sizes, and its potential to be used for further studies on tundra lake dynamics, in the context of global climate change, aimed at revealing new factors that could cause the planet to warm or cool. This research was funded by the Russian Science Foundation (RSF), project # 21-71-10052. D.L. and D.M. were partially supported by KAUST baseline funding. The authors thank Viktoria Kharchenko for manual labeling of part of the Sentinel-1 training and validation images |
author2 |
Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division Computer Science Program Visual Computing Center (VCC) Department of Space, Earth and Environment, Chalmers University of Technology, 412 96 Gothenburg, Sweden School of Mathematics and Statistics, The Open University, Milton Keynes MK7 6AA, UK The Center for Research and Invention, Veliky Novgorod 173008, Russia Arctic and Antarctic Research Institute, Saint Petersburg 199397, Russia |
format |
Article in Journal/Newspaper |
author |
Demchev, Denis Sudakow, Ivan Khodos, Alexander Abramova, Irina Lyakhov, Dmitry Michels, Dominik L. |
spellingShingle |
Demchev, Denis Sudakow, Ivan Khodos, Alexander Abramova, Irina Lyakhov, Dmitry Michels, Dominik L. Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation |
author_facet |
Demchev, Denis Sudakow, Ivan Khodos, Alexander Abramova, Irina Lyakhov, Dmitry Michels, Dominik L. |
author_sort |
Demchev, Denis |
title |
Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation |
title_short |
Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation |
title_full |
Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation |
title_fullStr |
Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation |
title_full_unstemmed |
Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation |
title_sort |
recognizing the shape and size of tundra lakes in synthetic aperture radar (sar) images using deep learning segmentation |
publisher |
MDPI AG |
publishDate |
2023 |
url |
http://hdl.handle.net/10754/689840 https://doi.org/10.3390/rs15051298 |
long_lat |
ENVELOPE(73.317,73.317,-52.983,-52.983) ENVELOPE(69.873,69.873,70.816,70.816) |
geographic |
The Sentinel Yamal Peninsula |
geographic_facet |
The Sentinel Yamal Peninsula |
genre |
permafrost Tundra Yamal Peninsula Alaska |
genre_facet |
permafrost Tundra Yamal Peninsula Alaska |
op_relation |
https://www.mdpi.com/2072-4292/15/5/1298 Demchev, D., Sudakow, I., Khodos, A., Abramova, I., Lyakhov, D., & Michels, D. (2023). Recognizing the Shape and Size of Tundra Lakes in Synthetic Aperture Radar (SAR) Images Using Deep Learning Segmentation. Remote Sensing, 15(5), 1298. https://doi.org/10.3390/rs15051298 doi:10.3390/rs15051298 2072-4292 5 Remote Sensing 1298 http://hdl.handle.net/10754/689840 15 |
op_rights |
Archived with thanks to Remote Sensing under a Creative Commons license, details at: https://creativecommons.org/licenses/by/4.0/ https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs15051298 |
container_title |
Remote Sensing |
container_volume |
15 |
container_issue |
5 |
container_start_page |
1298 |
_version_ |
1787427658845388800 |