Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to s...
Published in: | IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
IEEE
2021
|
Subjects: | |
Online Access: | https://hdl.handle.net/10037/21057 https://doi.org/10.1109/IGARSS39084.2020.9324192 |
id |
ftunivtroemsoe:oai:munin.uit.no:10037/21057 |
---|---|
record_format |
openpolar |
spelling |
ftunivtroemsoe:oai:munin.uit.no:10037/21057 2023-05-15T14:27:10+02:00 Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data Blix, Katalin Espeseth, Martine Eltoft, Torbjørn 2021-02-17 https://hdl.handle.net/10037/21057 https://doi.org/10.1109/IGARSS39084.2020.9324192 eng eng IEEE IEEE International Geoscience and Remote Sensing Symposium proceedings info:eu-repo/grantAgreement/RCN/SFI/237906/Norway/Centre for Integrated Remote Sensing and Forecasting for Arctic Operations/CIRFA/ Blix K, Espeseth M, Eltoft T. Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data. IEEE International Geoscience and Remote Sensing Symposium proceedings. 2020 FRIDAID 1807873 doi:10.1109/IGARSS39084.2020.9324192 2153-6996 2153-7003 https://hdl.handle.net/10037/21057 embargoedAccess Copyright 2021 The Author(s) VDP::Technology: 500::Information and communication technology: 550::Geographical information systems: 555 VDP::Teknologi: 500::Informasjons- og kommunikasjonsteknologi: 550::Geografiske informasjonssystemer: 555 Journal article Tidsskriftartikkel Peer reviewed acceptedVersion 2021 ftunivtroemsoe https://doi.org/10.1109/IGARSS39084.2020.9324192 2021-06-25T17:58:07Z © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. This paper addresses the problem of up-scaling full polarimetric (quad-pol) parameters from small quad-pol synthetic aperture radar (SAR) scenes to large dual-pol scenes, using a sophisticated Machine Learning (ML) method, namely the Gaussian Process Regression (GPR). The approach is to let the GPR model learn the relationships between the dual-pol input data and the quad-pol parameters on a quad-pol scene, and then extrapolate the relationships to the whole dual-pol scene. We demonstrate the procedure on two pairs of quadpol Radarsat-2 (RS2) and dual-pol ScanSAR Sentinel-1 (S1) scenes, acquired less than 20 minutes apart. The results are visualised as pixel-wise parametric maps, supported by three quantitative regression performance measures. In addition, we show certainty level maps for the estimated parameters. Our results indicate the potential of using the ML GPR model to upscale quad-pol scenes to large dual-pol images. Article in Journal/Newspaper Arctic University of Tromsø: Munin Open Research Archive IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium 738 741 |
institution |
Open Polar |
collection |
University of Tromsø: Munin Open Research Archive |
op_collection_id |
ftunivtroemsoe |
language |
English |
topic |
VDP::Technology: 500::Information and communication technology: 550::Geographical information systems: 555 VDP::Teknologi: 500::Informasjons- og kommunikasjonsteknologi: 550::Geografiske informasjonssystemer: 555 |
spellingShingle |
VDP::Technology: 500::Information and communication technology: 550::Geographical information systems: 555 VDP::Teknologi: 500::Informasjons- og kommunikasjonsteknologi: 550::Geografiske informasjonssystemer: 555 Blix, Katalin Espeseth, Martine Eltoft, Torbjørn Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data |
topic_facet |
VDP::Technology: 500::Information and communication technology: 550::Geographical information systems: 555 VDP::Teknologi: 500::Informasjons- og kommunikasjonsteknologi: 550::Geografiske informasjonssystemer: 555 |
description |
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. This paper addresses the problem of up-scaling full polarimetric (quad-pol) parameters from small quad-pol synthetic aperture radar (SAR) scenes to large dual-pol scenes, using a sophisticated Machine Learning (ML) method, namely the Gaussian Process Regression (GPR). The approach is to let the GPR model learn the relationships between the dual-pol input data and the quad-pol parameters on a quad-pol scene, and then extrapolate the relationships to the whole dual-pol scene. We demonstrate the procedure on two pairs of quadpol Radarsat-2 (RS2) and dual-pol ScanSAR Sentinel-1 (S1) scenes, acquired less than 20 minutes apart. The results are visualised as pixel-wise parametric maps, supported by three quantitative regression performance measures. In addition, we show certainty level maps for the estimated parameters. Our results indicate the potential of using the ML GPR model to upscale quad-pol scenes to large dual-pol images. |
format |
Article in Journal/Newspaper |
author |
Blix, Katalin Espeseth, Martine Eltoft, Torbjørn |
author_facet |
Blix, Katalin Espeseth, Martine Eltoft, Torbjørn |
author_sort |
Blix, Katalin |
title |
Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data |
title_short |
Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data |
title_full |
Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data |
title_fullStr |
Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data |
title_full_unstemmed |
Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data |
title_sort |
comparison of machine learning methods for predicting quad-polarimetric parameters from dual-polarimetric sar data |
publisher |
IEEE |
publishDate |
2021 |
url |
https://hdl.handle.net/10037/21057 https://doi.org/10.1109/IGARSS39084.2020.9324192 |
genre |
Arctic |
genre_facet |
Arctic |
op_relation |
IEEE International Geoscience and Remote Sensing Symposium proceedings info:eu-repo/grantAgreement/RCN/SFI/237906/Norway/Centre for Integrated Remote Sensing and Forecasting for Arctic Operations/CIRFA/ Blix K, Espeseth M, Eltoft T. Comparison of Machine Learning Methods for Predicting Quad-Polarimetric Parameters from Dual-Polarimetric SAR Data. IEEE International Geoscience and Remote Sensing Symposium proceedings. 2020 FRIDAID 1807873 doi:10.1109/IGARSS39084.2020.9324192 2153-6996 2153-7003 https://hdl.handle.net/10037/21057 |
op_rights |
embargoedAccess Copyright 2021 The Author(s) |
op_doi |
https://doi.org/10.1109/IGARSS39084.2020.9324192 |
container_title |
IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium |
container_start_page |
738 |
op_container_end_page |
741 |
_version_ |
1766300775881900032 |