Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information
Operational sea ice maps are usually generated manually using dual-polarization (DP) synthetic aperture radar (SAR) satellite imagery, but there is strong interest in automating this process. Recently launched satellites offer compact polarimetry (CP) imagery that provides more comprehensive polarim...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2023
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs15123199 https://doaj.org/article/0b7ab7a74c9e4ef496f8215a8485e449 |
id |
ftdoajarticles:oai:doaj.org/article:0b7ab7a74c9e4ef496f8215a8485e449 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:0b7ab7a74c9e4ef496f8215a8485e449 2023-07-23T04:21:40+02:00 Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information Saeid Taleghanidoozdoozan Linlin Xu David A. Clausi 2023-06-01T00:00:00Z https://doi.org/10.3390/rs15123199 https://doaj.org/article/0b7ab7a74c9e4ef496f8215a8485e449 EN eng MDPI AG https://www.mdpi.com/2072-4292/15/12/3199 https://doaj.org/toc/2072-4292 doi:10.3390/rs15123199 2072-4292 https://doaj.org/article/0b7ab7a74c9e4ef496f8215a8485e449 Remote Sensing, Vol 15, Iss 3199, p 3199 (2023) RADARSAT Constellation Mission (RCM) synthetic aperture radar (SAR) compact polarimetry ice types contextual information feature learning Science Q article 2023 ftdoajarticles https://doi.org/10.3390/rs15123199 2023-07-02T00:37:09Z Operational sea ice maps are usually generated manually using dual-polarization (DP) synthetic aperture radar (SAR) satellite imagery, but there is strong interest in automating this process. Recently launched satellites offer compact polarimetry (CP) imagery that provides more comprehensive polarimetric information compared to DP, which compels the use of CP for automated classification of SAR sea ice imagery. Existing sea ice scene classification algorithms using CP imagery rely on handcrafted features, while neural networks offer the potential of features that are more discriminating. We have developed a new and effective sea ice classification algorithm that leverages the nature of CP data. First, a residual-based convolutional neural network (ResCNN) is implemented to classify each pixel. In parallel, an unsupervised segmentation is performed to generate regions based on CP statistical properties. Regions are assigned a single class label by majority voting using the ResCNN output. For testing, quad-polarimetric (QP) SAR sea ice scenes from the RADARSAT Constellation Mission (RCM) are used, and QP, DP, CP, and reconstructed QP modes are compared for classification accuracy, while also comparing them to other classification approaches. Using CP achieves an overall accuracy of 96.86%, which is comparable to QP (97.16%), and higher than reconstructed QP and DP data by about 2% and 10%, respectively. The implemented algorithm using CP imagery provides an improved option for automated sea ice mapping. Article in Journal/Newspaper Sea ice Directory of Open Access Journals: DOAJ Articles Remote Sensing 15 12 3199 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
RADARSAT Constellation Mission (RCM) synthetic aperture radar (SAR) compact polarimetry ice types contextual information feature learning Science Q |
spellingShingle |
RADARSAT Constellation Mission (RCM) synthetic aperture radar (SAR) compact polarimetry ice types contextual information feature learning Science Q Saeid Taleghanidoozdoozan Linlin Xu David A. Clausi Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information |
topic_facet |
RADARSAT Constellation Mission (RCM) synthetic aperture radar (SAR) compact polarimetry ice types contextual information feature learning Science Q |
description |
Operational sea ice maps are usually generated manually using dual-polarization (DP) synthetic aperture radar (SAR) satellite imagery, but there is strong interest in automating this process. Recently launched satellites offer compact polarimetry (CP) imagery that provides more comprehensive polarimetric information compared to DP, which compels the use of CP for automated classification of SAR sea ice imagery. Existing sea ice scene classification algorithms using CP imagery rely on handcrafted features, while neural networks offer the potential of features that are more discriminating. We have developed a new and effective sea ice classification algorithm that leverages the nature of CP data. First, a residual-based convolutional neural network (ResCNN) is implemented to classify each pixel. In parallel, an unsupervised segmentation is performed to generate regions based on CP statistical properties. Regions are assigned a single class label by majority voting using the ResCNN output. For testing, quad-polarimetric (QP) SAR sea ice scenes from the RADARSAT Constellation Mission (RCM) are used, and QP, DP, CP, and reconstructed QP modes are compared for classification accuracy, while also comparing them to other classification approaches. Using CP achieves an overall accuracy of 96.86%, which is comparable to QP (97.16%), and higher than reconstructed QP and DP data by about 2% and 10%, respectively. The implemented algorithm using CP imagery provides an improved option for automated sea ice mapping. |
format |
Article in Journal/Newspaper |
author |
Saeid Taleghanidoozdoozan Linlin Xu David A. Clausi |
author_facet |
Saeid Taleghanidoozdoozan Linlin Xu David A. Clausi |
author_sort |
Saeid Taleghanidoozdoozan |
title |
Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information |
title_short |
Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information |
title_full |
Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information |
title_fullStr |
Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information |
title_full_unstemmed |
Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information |
title_sort |
region-based sea ice mapping using compact polarimetric synthetic aperture radar imagery with learned features and contextual information |
publisher |
MDPI AG |
publishDate |
2023 |
url |
https://doi.org/10.3390/rs15123199 https://doaj.org/article/0b7ab7a74c9e4ef496f8215a8485e449 |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
Remote Sensing, Vol 15, Iss 3199, p 3199 (2023) |
op_relation |
https://www.mdpi.com/2072-4292/15/12/3199 https://doaj.org/toc/2072-4292 doi:10.3390/rs15123199 2072-4292 https://doaj.org/article/0b7ab7a74c9e4ef496f8215a8485e449 |
op_doi |
https://doi.org/10.3390/rs15123199 |
container_title |
Remote Sensing |
container_volume |
15 |
container_issue |
12 |
container_start_page |
3199 |
_version_ |
1772187669183332352 |