Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data
Autor(a) principal: | |
---|---|
Data de Publicação: | 2023 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Anuário do Instituto de Geociências (Online) |
Texto Completo: | https://revistas.ufrj.br/index.php/aigeo/article/view/55296 |
Resumo: | Digital elevation models are responsible for providing altimetric information on a surface to be mapped. While global models of low and medium spatial resolution are available open source by several space agencies, the high- resolution ones, which are utilized in scales 1:25,000 and larger, are scarce and expensive. Here we address this limitation by the utilization of deep learning algorithms coupled with Single Image Super-Resolution techniques in digital elevation models to obtain better spatial quality versions from lower resolution inputs. The development of a GAN-based (Generative Adversarial Network-based) methodology enables the improvement of the initial spatial resolution of low-resolution images. In the geospatial data context, for example, these algorithms can be used with digital elevation models and satellite images. The methodological approach uses a dataset with digital elevation models SRTM (Shuttle Radar Topography Mission) (30 meters of spatial resolution) and ALOS PALSAR (12.5 meters of spatial resolution), created with the objective of allowing the study to be carried out, promoting the emergence of new research groups in the area as well as enabling the comparison between the results obtained. It has been found that by increasing the number of iterations the performance of the generated model was improved and the quality of the generated image increased. Furthermore, the visual analysis of the generated image against the high- and low-resolution ones showed a great similarity between the first two. |
id |
UFRJ-21_2aa923c3832a095b7fc9ef0b78cf296e |
---|---|
oai_identifier_str |
oai:ojs.pkp.sfu.ca:article/55296 |
network_acronym_str |
UFRJ-21 |
network_name_str |
Anuário do Instituto de Geociências (Online) |
repository_id_str |
|
spelling |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR DataDeep learningNeural networksDigital image processingDigital elevation models are responsible for providing altimetric information on a surface to be mapped. While global models of low and medium spatial resolution are available open source by several space agencies, the high- resolution ones, which are utilized in scales 1:25,000 and larger, are scarce and expensive. Here we address this limitation by the utilization of deep learning algorithms coupled with Single Image Super-Resolution techniques in digital elevation models to obtain better spatial quality versions from lower resolution inputs. The development of a GAN-based (Generative Adversarial Network-based) methodology enables the improvement of the initial spatial resolution of low-resolution images. In the geospatial data context, for example, these algorithms can be used with digital elevation models and satellite images. The methodological approach uses a dataset with digital elevation models SRTM (Shuttle Radar Topography Mission) (30 meters of spatial resolution) and ALOS PALSAR (12.5 meters of spatial resolution), created with the objective of allowing the study to be carried out, promoting the emergence of new research groups in the area as well as enabling the comparison between the results obtained. It has been found that by increasing the number of iterations the performance of the generated model was improved and the quality of the generated image increased. Furthermore, the visual analysis of the generated image against the high- and low-resolution ones showed a great similarity between the first two.Universidade Federal do Rio de Janeiro2023-07-27info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://revistas.ufrj.br/index.php/aigeo/article/view/5529610.11137/1982-3908_2023_46_55296Anuário do Instituto de Geociências; v. 46 (2023)Anuário do Instituto de Geociências; Vol. 46 (2023)1982-39080101-9759reponame:Anuário do Instituto de Geociências (Online)instname:Universidade Federal do Rio de Janeiro (UFRJ)instacron:UFRJenghttps://revistas.ufrj.br/index.php/aigeo/article/view/55296/pdfCopyright (c) 2023 Anuário do Instituto de Geociênciasinfo:eu-repo/semantics/openAccessMoreira, Leonardo AssumpçãoPoelking, Livia MoreiraGraça, Alan SalomãoAraki, Hideo2023-07-27T21:44:45Zoai:ojs.pkp.sfu.ca:article/55296Revistahttps://revistas.ufrj.br/index.php/aigeo/indexPUBhttps://revistas.ufrj.br/index.php/aigeo/oaianuario@igeo.ufrj.br||1982-39080101-9759opendoar:2023-07-27T21:44:45Anuário do Instituto de Geociências (Online) - Universidade Federal do Rio de Janeiro (UFRJ)false |
dc.title.none.fl_str_mv |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data |
title |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data |
spellingShingle |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data Moreira, Leonardo Assumpção Deep learning Neural networks Digital image processing |
title_short |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data |
title_full |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data |
title_fullStr |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data |
title_full_unstemmed |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data |
title_sort |
Use of Generative Adversarial Network Algorithm in Super-Resolution Images to Increase the Quality of Digital Elevation Models Based on ALOS PALSAR Data |
author |
Moreira, Leonardo Assumpção |
author_facet |
Moreira, Leonardo Assumpção Poelking, Livia Moreira Graça, Alan Salomão Araki, Hideo |
author_role |
author |
author2 |
Poelking, Livia Moreira Graça, Alan Salomão Araki, Hideo |
author2_role |
author author author |
dc.contributor.author.fl_str_mv |
Moreira, Leonardo Assumpção Poelking, Livia Moreira Graça, Alan Salomão Araki, Hideo |
dc.subject.por.fl_str_mv |
Deep learning Neural networks Digital image processing |
topic |
Deep learning Neural networks Digital image processing |
description |
Digital elevation models are responsible for providing altimetric information on a surface to be mapped. While global models of low and medium spatial resolution are available open source by several space agencies, the high- resolution ones, which are utilized in scales 1:25,000 and larger, are scarce and expensive. Here we address this limitation by the utilization of deep learning algorithms coupled with Single Image Super-Resolution techniques in digital elevation models to obtain better spatial quality versions from lower resolution inputs. The development of a GAN-based (Generative Adversarial Network-based) methodology enables the improvement of the initial spatial resolution of low-resolution images. In the geospatial data context, for example, these algorithms can be used with digital elevation models and satellite images. The methodological approach uses a dataset with digital elevation models SRTM (Shuttle Radar Topography Mission) (30 meters of spatial resolution) and ALOS PALSAR (12.5 meters of spatial resolution), created with the objective of allowing the study to be carried out, promoting the emergence of new research groups in the area as well as enabling the comparison between the results obtained. It has been found that by increasing the number of iterations the performance of the generated model was improved and the quality of the generated image increased. Furthermore, the visual analysis of the generated image against the high- and low-resolution ones showed a great similarity between the first two. |
publishDate |
2023 |
dc.date.none.fl_str_mv |
2023-07-27 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://revistas.ufrj.br/index.php/aigeo/article/view/55296 10.11137/1982-3908_2023_46_55296 |
url |
https://revistas.ufrj.br/index.php/aigeo/article/view/55296 |
identifier_str_mv |
10.11137/1982-3908_2023_46_55296 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
https://revistas.ufrj.br/index.php/aigeo/article/view/55296/pdf |
dc.rights.driver.fl_str_mv |
Copyright (c) 2023 Anuário do Instituto de Geociências info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Copyright (c) 2023 Anuário do Instituto de Geociências |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal do Rio de Janeiro |
publisher.none.fl_str_mv |
Universidade Federal do Rio de Janeiro |
dc.source.none.fl_str_mv |
Anuário do Instituto de Geociências; v. 46 (2023) Anuário do Instituto de Geociências; Vol. 46 (2023) 1982-3908 0101-9759 reponame:Anuário do Instituto de Geociências (Online) instname:Universidade Federal do Rio de Janeiro (UFRJ) instacron:UFRJ |
instname_str |
Universidade Federal do Rio de Janeiro (UFRJ) |
instacron_str |
UFRJ |
institution |
UFRJ |
reponame_str |
Anuário do Instituto de Geociências (Online) |
collection |
Anuário do Instituto de Geociências (Online) |
repository.name.fl_str_mv |
Anuário do Instituto de Geociências (Online) - Universidade Federal do Rio de Janeiro (UFRJ) |
repository.mail.fl_str_mv |
anuario@igeo.ufrj.br|| |
_version_ |
1797053535747047424 |