Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery
Rapid global warming is catalyzing widespread permafrost degradation in the Arctic, leading to destructive land-surface subsidence that destabilizes and deforms the ground. Consequently, human-built infrastructure constructed upon permafrost is currently at major risk of structural failure. Risk ass...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2022
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs14112719 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/14/11/2719/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/14/11/2719/ 2023-08-20T04:03:46+02:00 Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery Elias Manos Chandi Witharana Mahendra Rajitha Udawalpola Amit Hasan Anna K. Liljedahl agris 2022-06-06 application/pdf https://doi.org/10.3390/rs14112719 EN eng Multidisciplinary Digital Publishing Institute Remote Sensing Image Processing https://dx.doi.org/10.3390/rs14112719 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 14; Issue 11; Pages: 2719 deep learning artificial intelligence semantic segmentation U-Net VHSR imagery building extraction infrastructure Arctic permafrost Text 2022 ftmdpi https://doi.org/10.3390/rs14112719 2023-08-01T05:17:21Z Rapid global warming is catalyzing widespread permafrost degradation in the Arctic, leading to destructive land-surface subsidence that destabilizes and deforms the ground. Consequently, human-built infrastructure constructed upon permafrost is currently at major risk of structural failure. Risk assessment frameworks that attempt to study this issue assume that precise information on the location and extent of infrastructure is known. However, complete, high-quality, uniform geospatial datasets of built infrastructure that are readily available for such scientific studies are lacking. While imagery-enabled mapping can fill this knowledge gap, the small size of individual structures and vast geographical extent of the Arctic necessitate large volumes of very high spatial resolution remote sensing imagery. Transforming this ‘big’ imagery data into ‘science-ready’ information demands highly automated image analysis pipelines driven by advanced computer vision algorithms. Despite this, previous fine resolution studies have been limited to manual digitization of features on locally confined scales. Therefore, this exploratory study serves as the first investigation into fully automated analysis of sub-meter spatial resolution satellite imagery for automated detection of Arctic built infrastructure. We tasked the U-Net, a deep learning-based semantic segmentation model, with classifying different infrastructure types (residential, commercial, public, and industrial buildings, as well as roads) from commercial satellite imagery of Utqiagvik and Prudhoe Bay, Alaska. We also conducted a systematic experiment to understand how image augmentation can impact model performance when labeled training data is limited. When optimal augmentation methods were applied, the U-Net achieved an average F1 score of 0.83. Overall, our experimental findings show that the U-Net-based workflow is a promising method for automated Arctic built infrastructure detection that, combined with existing optimized workflows, such as MAPLE, could be ... Text Arctic Global warming permafrost Prudhoe Bay Alaska MDPI Open Access Publishing Arctic Remote Sensing 14 11 2719 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
deep learning artificial intelligence semantic segmentation U-Net VHSR imagery building extraction infrastructure Arctic permafrost |
spellingShingle |
deep learning artificial intelligence semantic segmentation U-Net VHSR imagery building extraction infrastructure Arctic permafrost Elias Manos Chandi Witharana Mahendra Rajitha Udawalpola Amit Hasan Anna K. Liljedahl Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery |
topic_facet |
deep learning artificial intelligence semantic segmentation U-Net VHSR imagery building extraction infrastructure Arctic permafrost |
description |
Rapid global warming is catalyzing widespread permafrost degradation in the Arctic, leading to destructive land-surface subsidence that destabilizes and deforms the ground. Consequently, human-built infrastructure constructed upon permafrost is currently at major risk of structural failure. Risk assessment frameworks that attempt to study this issue assume that precise information on the location and extent of infrastructure is known. However, complete, high-quality, uniform geospatial datasets of built infrastructure that are readily available for such scientific studies are lacking. While imagery-enabled mapping can fill this knowledge gap, the small size of individual structures and vast geographical extent of the Arctic necessitate large volumes of very high spatial resolution remote sensing imagery. Transforming this ‘big’ imagery data into ‘science-ready’ information demands highly automated image analysis pipelines driven by advanced computer vision algorithms. Despite this, previous fine resolution studies have been limited to manual digitization of features on locally confined scales. Therefore, this exploratory study serves as the first investigation into fully automated analysis of sub-meter spatial resolution satellite imagery for automated detection of Arctic built infrastructure. We tasked the U-Net, a deep learning-based semantic segmentation model, with classifying different infrastructure types (residential, commercial, public, and industrial buildings, as well as roads) from commercial satellite imagery of Utqiagvik and Prudhoe Bay, Alaska. We also conducted a systematic experiment to understand how image augmentation can impact model performance when labeled training data is limited. When optimal augmentation methods were applied, the U-Net achieved an average F1 score of 0.83. Overall, our experimental findings show that the U-Net-based workflow is a promising method for automated Arctic built infrastructure detection that, combined with existing optimized workflows, such as MAPLE, could be ... |
format |
Text |
author |
Elias Manos Chandi Witharana Mahendra Rajitha Udawalpola Amit Hasan Anna K. Liljedahl |
author_facet |
Elias Manos Chandi Witharana Mahendra Rajitha Udawalpola Amit Hasan Anna K. Liljedahl |
author_sort |
Elias Manos |
title |
Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery |
title_short |
Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery |
title_full |
Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery |
title_fullStr |
Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery |
title_full_unstemmed |
Convolutional Neural Networks for Automated Built Infrastructure Detection in the Arctic Using Sub-Meter Spatial Resolution Satellite Imagery |
title_sort |
convolutional neural networks for automated built infrastructure detection in the arctic using sub-meter spatial resolution satellite imagery |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2022 |
url |
https://doi.org/10.3390/rs14112719 |
op_coverage |
agris |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic Global warming permafrost Prudhoe Bay Alaska |
genre_facet |
Arctic Global warming permafrost Prudhoe Bay Alaska |
op_source |
Remote Sensing; Volume 14; Issue 11; Pages: 2719 |
op_relation |
Remote Sensing Image Processing https://dx.doi.org/10.3390/rs14112719 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs14112719 |
container_title |
Remote Sensing |
container_volume |
14 |
container_issue |
11 |
container_start_page |
2719 |
_version_ |
1774714204620587008 |