How to Get the Most Out of U-Net for Glacier Calving Front Segmentation
The melting of ice sheets and glaciers is one of the main contributors to global sea-level rise. Hence, continuous monitoring of glacier changes and in particular the mapping of positional changes of their calving front is of significant importance. This delineation process, in general, has been car...
Published in: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
---|---|
Main Authors: | , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://opus4.kobv.de/opus4-fau/frontdoor/index/index/docId/23446 https://nbn-resolving.org/urn:nbn:de:bvb:29-opus4-234461 https://doi.org/10.1109/JSTARS.2022.3148033 https://opus4.kobv.de/opus4-fau/files/23446/How_to_Get_the_Most_Out_of_U-Net.pdf |
id |
ftuniverlangen:oai:ub.uni-erlangen.de-opus:23446 |
---|---|
record_format |
openpolar |
spelling |
ftuniverlangen:oai:ub.uni-erlangen.de-opus:23446 2023-07-23T04:15:48+02:00 How to Get the Most Out of U-Net for Glacier Calving Front Segmentation Periyasamy, Maniraman Davari, Amirabbas Seehaus, Thorsten Braun, Matthias Maier, Andreas Christlein, Vincent 2022-02-01 application/pdf https://opus4.kobv.de/opus4-fau/frontdoor/index/index/docId/23446 https://nbn-resolving.org/urn:nbn:de:bvb:29-opus4-234461 https://doi.org/10.1109/JSTARS.2022.3148033 https://opus4.kobv.de/opus4-fau/files/23446/How_to_Get_the_Most_Out_of_U-Net.pdf eng eng https://opus4.kobv.de/opus4-fau/frontdoor/index/index/docId/23446 urn:nbn:de:bvb:29-opus4-234461 https://nbn-resolving.org/urn:nbn:de:bvb:29-opus4-234461 https://doi.org/10.1109/JSTARS.2022.3148033 https://opus4.kobv.de/opus4-fau/files/23446/How_to_Get_the_Most_Out_of_U-Net.pdf https://creativecommons.org/licenses/by/4.0/deed.de info:eu-repo/semantics/openAccess ddc:000 article doc-type:article 2022 ftuniverlangen https://doi.org/10.1109/JSTARS.2022.3148033 2023-07-02T22:42:44Z The melting of ice sheets and glaciers is one of the main contributors to global sea-level rise. Hence, continuous monitoring of glacier changes and in particular the mapping of positional changes of their calving front is of significant importance. This delineation process, in general, has been carried out manually, which is time-consuming and not feasible for the abundance of available data within the past decade. Automatic delineation of the glacier fronts in synthetic aperture radar (SAR) images can be performed using deep learning-based U-Net models. This article aims to study and survey the components of a U-Net model and optimize the model to get the most out of U-Net for glacier (calving front) segmentation. We trained the U-Net to segment the SAR images of Sjogren-Inlet and Dinsmoore–Bombardier–Edgworth glacier systems on the Antarctica Peninsula region taken by ERS-1/2, Envisat, RadarSAT-1, ALOS, TerraSAR-X, and TanDEM-X missions. The U-Net model was optimized in six aspects. The first two aspects, namely data preprocessing and data augmentation, enhanced the representation of information in the image. The remaining four aspects optimized the feature extraction of U-Net by finding the best-suited loss function, bottleneck, normalization technique, and dropouts for the glacier segmentation task. The optimized U-Net model achieves a dice coefficient score of 0.9378 with a 20% improvement over the baseline U-Net model, which achieved a score of 0.7377. This segmentation result is further postprocessed to delineate the calving front. The optimized U-Net model shows 23% improvement in the glacier front delineation compared to the baseline model. Article in Journal/Newspaper Antarc* Antarctica OPUS FAU - Online publication system of Friedrich-Alexander-Universität Erlangen-Nürnberg Sjogren ENVELOPE(-58.867,-58.867,-64.233,-64.233) IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 15 1712 1723 |
institution |
Open Polar |
collection |
OPUS FAU - Online publication system of Friedrich-Alexander-Universität Erlangen-Nürnberg |
op_collection_id |
ftuniverlangen |
language |
English |
topic |
ddc:000 |
spellingShingle |
ddc:000 Periyasamy, Maniraman Davari, Amirabbas Seehaus, Thorsten Braun, Matthias Maier, Andreas Christlein, Vincent How to Get the Most Out of U-Net for Glacier Calving Front Segmentation |
topic_facet |
ddc:000 |
description |
The melting of ice sheets and glaciers is one of the main contributors to global sea-level rise. Hence, continuous monitoring of glacier changes and in particular the mapping of positional changes of their calving front is of significant importance. This delineation process, in general, has been carried out manually, which is time-consuming and not feasible for the abundance of available data within the past decade. Automatic delineation of the glacier fronts in synthetic aperture radar (SAR) images can be performed using deep learning-based U-Net models. This article aims to study and survey the components of a U-Net model and optimize the model to get the most out of U-Net for glacier (calving front) segmentation. We trained the U-Net to segment the SAR images of Sjogren-Inlet and Dinsmoore–Bombardier–Edgworth glacier systems on the Antarctica Peninsula region taken by ERS-1/2, Envisat, RadarSAT-1, ALOS, TerraSAR-X, and TanDEM-X missions. The U-Net model was optimized in six aspects. The first two aspects, namely data preprocessing and data augmentation, enhanced the representation of information in the image. The remaining four aspects optimized the feature extraction of U-Net by finding the best-suited loss function, bottleneck, normalization technique, and dropouts for the glacier segmentation task. The optimized U-Net model achieves a dice coefficient score of 0.9378 with a 20% improvement over the baseline U-Net model, which achieved a score of 0.7377. This segmentation result is further postprocessed to delineate the calving front. The optimized U-Net model shows 23% improvement in the glacier front delineation compared to the baseline model. |
format |
Article in Journal/Newspaper |
author |
Periyasamy, Maniraman Davari, Amirabbas Seehaus, Thorsten Braun, Matthias Maier, Andreas Christlein, Vincent |
author_facet |
Periyasamy, Maniraman Davari, Amirabbas Seehaus, Thorsten Braun, Matthias Maier, Andreas Christlein, Vincent |
author_sort |
Periyasamy, Maniraman |
title |
How to Get the Most Out of U-Net for Glacier Calving Front Segmentation |
title_short |
How to Get the Most Out of U-Net for Glacier Calving Front Segmentation |
title_full |
How to Get the Most Out of U-Net for Glacier Calving Front Segmentation |
title_fullStr |
How to Get the Most Out of U-Net for Glacier Calving Front Segmentation |
title_full_unstemmed |
How to Get the Most Out of U-Net for Glacier Calving Front Segmentation |
title_sort |
how to get the most out of u-net for glacier calving front segmentation |
publishDate |
2022 |
url |
https://opus4.kobv.de/opus4-fau/frontdoor/index/index/docId/23446 https://nbn-resolving.org/urn:nbn:de:bvb:29-opus4-234461 https://doi.org/10.1109/JSTARS.2022.3148033 https://opus4.kobv.de/opus4-fau/files/23446/How_to_Get_the_Most_Out_of_U-Net.pdf |
long_lat |
ENVELOPE(-58.867,-58.867,-64.233,-64.233) |
geographic |
Sjogren |
geographic_facet |
Sjogren |
genre |
Antarc* Antarctica |
genre_facet |
Antarc* Antarctica |
op_relation |
https://opus4.kobv.de/opus4-fau/frontdoor/index/index/docId/23446 urn:nbn:de:bvb:29-opus4-234461 https://nbn-resolving.org/urn:nbn:de:bvb:29-opus4-234461 https://doi.org/10.1109/JSTARS.2022.3148033 https://opus4.kobv.de/opus4-fau/files/23446/How_to_Get_the_Most_Out_of_U-Net.pdf |
op_rights |
https://creativecommons.org/licenses/by/4.0/deed.de info:eu-repo/semantics/openAccess |
op_doi |
https://doi.org/10.1109/JSTARS.2022.3148033 |
container_title |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
container_volume |
15 |
container_start_page |
1712 |
op_container_end_page |
1723 |
_version_ |
1772176825939656704 |