Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data
In this paper, the performance of C-band synthetic aperture radar (SAR) Gaofen-3 (GF-3) quad-polarization Stripmap (QPS) data is assessed for classifying late spring and summer sea ice types. The investigation is based on 18 scenes of GF-3 QPS data acquired in the Arctic Ocean in 2017. In this study...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2021
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs13081452 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/13/8/1452/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/13/8/1452/ 2023-08-20T04:05:01+02:00 Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data Tianyu Zhang Ying Yang Mohammed Shokr Chunlei Mi Xiao-Ming Li Xiao Cheng Fengming Hui agris 2021-04-09 application/pdf https://doi.org/10.3390/rs13081452 EN eng Multidisciplinary Digital Publishing Institute https://dx.doi.org/10.3390/rs13081452 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 13; Issue 8; Pages: 1452 Gaofen-3 polarimetric data sea ice classification residual convolutional network Text 2021 ftmdpi https://doi.org/10.3390/rs13081452 2023-08-01T01:27:56Z In this paper, the performance of C-band synthetic aperture radar (SAR) Gaofen-3 (GF-3) quad-polarization Stripmap (QPS) data is assessed for classifying late spring and summer sea ice types. The investigation is based on 18 scenes of GF-3 QPS data acquired in the Arctic Ocean in 2017. In this study, floe ice (FI), brash ice (BI) between floes and open water (OW, ice-free area) were classified based on a mini sea ice residual convolutional network, which we call MSI-ResNet. While investigating the optimal patch size for MSI-ResNet, we found that, as the patch size continues to grow, the classification accuracy first increases and then decreases. A patch size of 31 × 31 was found to achieve the best performance. The performance of classification using different polarization combinations from the QPS data was also assessed. The vertical-vertical (VV) polarization input overestimates the FI category while incorrectly identifying most of the BI as FI. The VH polarization produces a synchronous improvement in FI, BI, and OW discrimination, with a higher overall accuracy and kappa coefficient (91.09% and 0.85, respectively) than the VV polarization (83.37% and 0.70, respectively). The combination of VV and vertical-horizontal (VH) polarizations presents a modest precision improvement for BI and OW together with a slight overestimation for FI. With VV, VH, and horizontal-horizontal (HH) polarization data as the inputs, the user’s accuracy improves to 95.12%, 93.42%, and 95.17% for FI, BI, and OW, respectively. The accuracy was assessed against visual interpretation of the sea ice classes in the images using a stratified sampling method. The application of the MSI-ResNet method to data covering the Beaufort Sea and the north of the Severnaya Zemlya archipelago was found to achieve a high overall accuracy (kappa) of 94.62% (±0.92) and 94.23% (±0.90), respectively. This is similar to the classification accuracy obtained in the Fram Strait. From the results of this study, it is shown that the MSI-ResNet method performs ... Text Arctic Arctic Ocean Beaufort Sea Fram Strait Sea ice Severnaya Zemlya MDPI Open Access Publishing Arctic Arctic Ocean Severnaya Zemlya ENVELOPE(98.000,98.000,79.500,79.500) Remote Sensing 13 8 1452 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
Gaofen-3 polarimetric data sea ice classification residual convolutional network |
spellingShingle |
Gaofen-3 polarimetric data sea ice classification residual convolutional network Tianyu Zhang Ying Yang Mohammed Shokr Chunlei Mi Xiao-Ming Li Xiao Cheng Fengming Hui Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data |
topic_facet |
Gaofen-3 polarimetric data sea ice classification residual convolutional network |
description |
In this paper, the performance of C-band synthetic aperture radar (SAR) Gaofen-3 (GF-3) quad-polarization Stripmap (QPS) data is assessed for classifying late spring and summer sea ice types. The investigation is based on 18 scenes of GF-3 QPS data acquired in the Arctic Ocean in 2017. In this study, floe ice (FI), brash ice (BI) between floes and open water (OW, ice-free area) were classified based on a mini sea ice residual convolutional network, which we call MSI-ResNet. While investigating the optimal patch size for MSI-ResNet, we found that, as the patch size continues to grow, the classification accuracy first increases and then decreases. A patch size of 31 × 31 was found to achieve the best performance. The performance of classification using different polarization combinations from the QPS data was also assessed. The vertical-vertical (VV) polarization input overestimates the FI category while incorrectly identifying most of the BI as FI. The VH polarization produces a synchronous improvement in FI, BI, and OW discrimination, with a higher overall accuracy and kappa coefficient (91.09% and 0.85, respectively) than the VV polarization (83.37% and 0.70, respectively). The combination of VV and vertical-horizontal (VH) polarizations presents a modest precision improvement for BI and OW together with a slight overestimation for FI. With VV, VH, and horizontal-horizontal (HH) polarization data as the inputs, the user’s accuracy improves to 95.12%, 93.42%, and 95.17% for FI, BI, and OW, respectively. The accuracy was assessed against visual interpretation of the sea ice classes in the images using a stratified sampling method. The application of the MSI-ResNet method to data covering the Beaufort Sea and the north of the Severnaya Zemlya archipelago was found to achieve a high overall accuracy (kappa) of 94.62% (±0.92) and 94.23% (±0.90), respectively. This is similar to the classification accuracy obtained in the Fram Strait. From the results of this study, it is shown that the MSI-ResNet method performs ... |
format |
Text |
author |
Tianyu Zhang Ying Yang Mohammed Shokr Chunlei Mi Xiao-Ming Li Xiao Cheng Fengming Hui |
author_facet |
Tianyu Zhang Ying Yang Mohammed Shokr Chunlei Mi Xiao-Ming Li Xiao Cheng Fengming Hui |
author_sort |
Tianyu Zhang |
title |
Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data |
title_short |
Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data |
title_full |
Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data |
title_fullStr |
Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data |
title_full_unstemmed |
Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data |
title_sort |
deep learning based sea ice classification with gaofen-3 fully polarimetric sar data |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2021 |
url |
https://doi.org/10.3390/rs13081452 |
op_coverage |
agris |
long_lat |
ENVELOPE(98.000,98.000,79.500,79.500) |
geographic |
Arctic Arctic Ocean Severnaya Zemlya |
geographic_facet |
Arctic Arctic Ocean Severnaya Zemlya |
genre |
Arctic Arctic Ocean Beaufort Sea Fram Strait Sea ice Severnaya Zemlya |
genre_facet |
Arctic Arctic Ocean Beaufort Sea Fram Strait Sea ice Severnaya Zemlya |
op_source |
Remote Sensing; Volume 13; Issue 8; Pages: 1452 |
op_relation |
https://dx.doi.org/10.3390/rs13081452 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs13081452 |
container_title |
Remote Sensing |
container_volume |
13 |
container_issue |
8 |
container_start_page |
1452 |
_version_ |
1774715445429927936 |