Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning
Sea ice has a significant effect on climate change and ship navigation. Hence, it is crucial to draw sea-ice maps that reflect the geographical distribution of different types of sea ice. Many automatic sea-ice classification methods using synthetic aperture radar (SAR) images are based on the polar...
Published in: | IEEE Transactions on Geoscience and Remote Sensing |
---|---|
Main Authors: | , , , , , , |
Other Authors: | , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2021
|
Subjects: | |
Online Access: | http://hdl.handle.net/11584/321731 https://doi.org/10.1109/TGRS.2020.3049031 https://ieeexplore.ieee.org/document/9332239 |
id |
ftunicagliariris:oai:iris.unica.it:11584/321731 |
---|---|
record_format |
openpolar |
spelling |
ftunicagliariris:oai:iris.unica.it:11584/321731 2024-02-11T10:08:28+01:00 Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning Song W. Li M. Gao W. Huang D. Ma Z. Liotta A. Perra C. Song, W. Li, M. Gao, W. Huang, D. Ma, Z. Liotta, A. Perra, C. 2021 http://hdl.handle.net/11584/321731 https://doi.org/10.1109/TGRS.2020.3049031 https://ieeexplore.ieee.org/document/9332239 eng eng info:eu-repo/semantics/altIdentifier/wos/WOS:000722170500011 firstpage:1 lastpage:15 numberofpages:15 journal:IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING http://hdl.handle.net/11584/321731 doi:10.1109/TGRS.2020.3049031 info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85100464451 https://ieeexplore.ieee.org/document/9332239 info:eu-repo/semantics/closedAccess Ice chart long short-term memory (LSTM) residual convolution network sea-ice classification synthetic aperture radar (SAR) surface roughne radar polarimetry sea surface info:eu-repo/semantics/article 2021 ftunicagliariris https://doi.org/10.1109/TGRS.2020.3049031 2024-01-24T17:54:39Z Sea ice has a significant effect on climate change and ship navigation. Hence, it is crucial to draw sea-ice maps that reflect the geographical distribution of different types of sea ice. Many automatic sea-ice classification methods using synthetic aperture radar (SAR) images are based on the polarimetric characteristics or image texture features of sea ice. They either require professional knowledge to design the parameters and features or are sensitive to noise and condition changes. Moreover, ice changes over time are often ignored. In this article, we propose a new SAR sea-ice image classification method based on a combined learning of spatial and temporal features, derived from residual convolutional neural networks (ResNet) and long short-term memory (LSTM) networks. In this way, we achieve automatic and refined classification of sea-ice types. First, we construct a seven-type ice data set according to the Canadian Ice Service ice charts. We extract spatial feature vectors of a time series of sea-ice samples using a trained ResNet network. Then, using the feature vectors as inputs, the LSTM network further learns the variation of the set of sea-ice samples with time. Finally, the extracted high-level features are fed into a softmax classifier to output the most recent ice type. Taking both spatial features and time variation into consideration, our method can achieve a high classification accuracy of 95.7% for seven ice types. Our method can automatically produce more objective sea-ice interpretation maps, allowing detailed sea-ice distribution and improving the efficiency of sea-ice monitoring tasks. Article in Journal/Newspaper Sea ice Università degli Studi di Cagliari: UNICA IRIS IEEE Transactions on Geoscience and Remote Sensing 59 12 9887 9901 |
institution |
Open Polar |
collection |
Università degli Studi di Cagliari: UNICA IRIS |
op_collection_id |
ftunicagliariris |
language |
English |
topic |
Ice chart long short-term memory (LSTM) residual convolution network sea-ice classification synthetic aperture radar (SAR) surface roughne radar polarimetry sea surface |
spellingShingle |
Ice chart long short-term memory (LSTM) residual convolution network sea-ice classification synthetic aperture radar (SAR) surface roughne radar polarimetry sea surface Song W. Li M. Gao W. Huang D. Ma Z. Liotta A. Perra C. Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning |
topic_facet |
Ice chart long short-term memory (LSTM) residual convolution network sea-ice classification synthetic aperture radar (SAR) surface roughne radar polarimetry sea surface |
description |
Sea ice has a significant effect on climate change and ship navigation. Hence, it is crucial to draw sea-ice maps that reflect the geographical distribution of different types of sea ice. Many automatic sea-ice classification methods using synthetic aperture radar (SAR) images are based on the polarimetric characteristics or image texture features of sea ice. They either require professional knowledge to design the parameters and features or are sensitive to noise and condition changes. Moreover, ice changes over time are often ignored. In this article, we propose a new SAR sea-ice image classification method based on a combined learning of spatial and temporal features, derived from residual convolutional neural networks (ResNet) and long short-term memory (LSTM) networks. In this way, we achieve automatic and refined classification of sea-ice types. First, we construct a seven-type ice data set according to the Canadian Ice Service ice charts. We extract spatial feature vectors of a time series of sea-ice samples using a trained ResNet network. Then, using the feature vectors as inputs, the LSTM network further learns the variation of the set of sea-ice samples with time. Finally, the extracted high-level features are fed into a softmax classifier to output the most recent ice type. Taking both spatial features and time variation into consideration, our method can achieve a high classification accuracy of 95.7% for seven ice types. Our method can automatically produce more objective sea-ice interpretation maps, allowing detailed sea-ice distribution and improving the efficiency of sea-ice monitoring tasks. |
author2 |
Song, W. Li, M. Gao, W. Huang, D. Ma, Z. Liotta, A. Perra, C. |
format |
Article in Journal/Newspaper |
author |
Song W. Li M. Gao W. Huang D. Ma Z. Liotta A. Perra C. |
author_facet |
Song W. Li M. Gao W. Huang D. Ma Z. Liotta A. Perra C. |
author_sort |
Song W. |
title |
Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning |
title_short |
Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning |
title_full |
Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning |
title_fullStr |
Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning |
title_full_unstemmed |
Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning |
title_sort |
automatic sea-ice classification of sar images based on spatial and temporal features learning |
publishDate |
2021 |
url |
http://hdl.handle.net/11584/321731 https://doi.org/10.1109/TGRS.2020.3049031 https://ieeexplore.ieee.org/document/9332239 |
genre |
Sea ice |
genre_facet |
Sea ice |
op_relation |
info:eu-repo/semantics/altIdentifier/wos/WOS:000722170500011 firstpage:1 lastpage:15 numberofpages:15 journal:IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING http://hdl.handle.net/11584/321731 doi:10.1109/TGRS.2020.3049031 info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-85100464451 https://ieeexplore.ieee.org/document/9332239 |
op_rights |
info:eu-repo/semantics/closedAccess |
op_doi |
https://doi.org/10.1109/TGRS.2020.3049031 |
container_title |
IEEE Transactions on Geoscience and Remote Sensing |
container_volume |
59 |
container_issue |
12 |
container_start_page |
9887 |
op_container_end_page |
9901 |
_version_ |
1790607819684184064 |