KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products
International audience The SWOT (Surface Water Ocean Topography) mission will provide high-resolution and two-dimensional measurements of sea surface height (SSH). However, despite its unprecedented precision, SWOT’s Ka-band Radar Interferometer (KaRIn) still exhibits a substantial amount of random...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , , |
Other Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
HAL CCSD
2023
|
Subjects: | |
Online Access: | https://hal.science/hal-04501384 https://hal.science/hal-04501384/document https://hal.science/hal-04501384/file/remotesensing-15-02183.pdf https://doi.org/10.3390/rs15082183 |
id |
ftutoulouse3hal:oai:HAL:hal-04501384v1 |
---|---|
record_format |
openpolar |
spelling |
ftutoulouse3hal:oai:HAL:hal-04501384v1 2024-09-09T19:57:32+00:00 KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products Tréboutte, Anaëlle Carli, Elisa Ballarotta, Maxime Carpentier, Benjamin Faugère, Yannice Dibarboure, Gérald Laboratoire d'études en Géophysique et océanographie spatiales (LEGOS) Institut de Recherche pour le Développement (IRD)-Université Toulouse III - Paul Sabatier (UT3) Université de Toulouse (UT)-Université de Toulouse (UT)-Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire Midi-Pyrénées (OMP) Université de Toulouse (UT)-Université de Toulouse (UT)-Institut national des sciences de l'Univers (INSU - CNRS)-Centre National d'Études Spatiales Toulouse (CNES)-Centre National de la Recherche Scientifique (CNRS)-Météo-France-Institut de Recherche pour le Développement (IRD)-Institut national des sciences de l'Univers (INSU - CNRS)-Centre National d'Études Spatiales Toulouse (CNES)-Centre National de la Recherche Scientifique (CNRS)-Météo-France-Centre National de la Recherche Scientifique (CNRS) 2023-04-20 https://hal.science/hal-04501384 https://hal.science/hal-04501384/document https://hal.science/hal-04501384/file/remotesensing-15-02183.pdf https://doi.org/10.3390/rs15082183 en eng HAL CCSD MDPI info:eu-repo/semantics/altIdentifier/doi/10.3390/rs15082183 hal-04501384 https://hal.science/hal-04501384 https://hal.science/hal-04501384/document https://hal.science/hal-04501384/file/remotesensing-15-02183.pdf doi:10.3390/rs15082183 info:eu-repo/semantics/OpenAccess ISSN: 2072-4292 Remote Sensing https://hal.science/hal-04501384 Remote Sensing, 2023, 15 (8), pp.2183. ⟨10.3390/rs15082183⟩ [SPI]Engineering Sciences [physics] info:eu-repo/semantics/article Journal articles 2023 ftutoulouse3hal https://doi.org/10.3390/rs15082183 2024-06-25T00:02:29Z International audience The SWOT (Surface Water Ocean Topography) mission will provide high-resolution and two-dimensional measurements of sea surface height (SSH). However, despite its unprecedented precision, SWOT’s Ka-band Radar Interferometer (KaRIn) still exhibits a substantial amount of random noise. In turn, the random noise limits the ability of SWOT to capture the smallest scales of the ocean’s topography and its derivatives. In that context, this paper explores the feasibility, strengths and limits of a noise-reduction algorithm based on a convolutional neural network. The model is based on a U-Net architecture and is trained and tested with simulated data from the North Atlantic. Our results are compared to classical smoothing methods: a median filter, a Lanczos kernel smoother and the SWOT de-noising algorithm developed by Gomez-Navarro et al. Our U-Net model yields better results for all the evaluation metrics: 2 mm root mean square error, sub-millimetric bias, variance reduction by factor of 44 (16 dB) and an accurate power spectral density down to 10–20 km wavelengths. We also tested various scenarios to infer the robustness and the stability of the U-Net. The U-Net always exhibits good performance and can be further improved with retraining if necessary. This robustness in simulation is very encouraging: our findings show that the U-Net architecture is likely one of the best candidates to reduce the noise of flight data from KaRIn. Article in Journal/Newspaper North Atlantic Université Toulouse III - Paul Sabatier: HAL-UPS Gomez ENVELOPE(-58.795,-58.795,-62.196,-62.196) Navarro ENVELOPE(-62.167,-62.167,-64.650,-64.650) Remote Sensing 15 8 2183 |
institution |
Open Polar |
collection |
Université Toulouse III - Paul Sabatier: HAL-UPS |
op_collection_id |
ftutoulouse3hal |
language |
English |
topic |
[SPI]Engineering Sciences [physics] |
spellingShingle |
[SPI]Engineering Sciences [physics] Tréboutte, Anaëlle Carli, Elisa Ballarotta, Maxime Carpentier, Benjamin Faugère, Yannice Dibarboure, Gérald KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products |
topic_facet |
[SPI]Engineering Sciences [physics] |
description |
International audience The SWOT (Surface Water Ocean Topography) mission will provide high-resolution and two-dimensional measurements of sea surface height (SSH). However, despite its unprecedented precision, SWOT’s Ka-band Radar Interferometer (KaRIn) still exhibits a substantial amount of random noise. In turn, the random noise limits the ability of SWOT to capture the smallest scales of the ocean’s topography and its derivatives. In that context, this paper explores the feasibility, strengths and limits of a noise-reduction algorithm based on a convolutional neural network. The model is based on a U-Net architecture and is trained and tested with simulated data from the North Atlantic. Our results are compared to classical smoothing methods: a median filter, a Lanczos kernel smoother and the SWOT de-noising algorithm developed by Gomez-Navarro et al. Our U-Net model yields better results for all the evaluation metrics: 2 mm root mean square error, sub-millimetric bias, variance reduction by factor of 44 (16 dB) and an accurate power spectral density down to 10–20 km wavelengths. We also tested various scenarios to infer the robustness and the stability of the U-Net. The U-Net always exhibits good performance and can be further improved with retraining if necessary. This robustness in simulation is very encouraging: our findings show that the U-Net architecture is likely one of the best candidates to reduce the noise of flight data from KaRIn. |
author2 |
Laboratoire d'études en Géophysique et océanographie spatiales (LEGOS) Institut de Recherche pour le Développement (IRD)-Université Toulouse III - Paul Sabatier (UT3) Université de Toulouse (UT)-Université de Toulouse (UT)-Institut national des sciences de l'Univers (INSU - CNRS)-Observatoire Midi-Pyrénées (OMP) Université de Toulouse (UT)-Université de Toulouse (UT)-Institut national des sciences de l'Univers (INSU - CNRS)-Centre National d'Études Spatiales Toulouse (CNES)-Centre National de la Recherche Scientifique (CNRS)-Météo-France-Institut de Recherche pour le Développement (IRD)-Institut national des sciences de l'Univers (INSU - CNRS)-Centre National d'Études Spatiales Toulouse (CNES)-Centre National de la Recherche Scientifique (CNRS)-Météo-France-Centre National de la Recherche Scientifique (CNRS) |
format |
Article in Journal/Newspaper |
author |
Tréboutte, Anaëlle Carli, Elisa Ballarotta, Maxime Carpentier, Benjamin Faugère, Yannice Dibarboure, Gérald |
author_facet |
Tréboutte, Anaëlle Carli, Elisa Ballarotta, Maxime Carpentier, Benjamin Faugère, Yannice Dibarboure, Gérald |
author_sort |
Tréboutte, Anaëlle |
title |
KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products |
title_short |
KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products |
title_full |
KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products |
title_fullStr |
KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products |
title_full_unstemmed |
KaRIn Noise Reduction Using a Convolutional Neural Network for the SWOT Ocean Products |
title_sort |
karin noise reduction using a convolutional neural network for the swot ocean products |
publisher |
HAL CCSD |
publishDate |
2023 |
url |
https://hal.science/hal-04501384 https://hal.science/hal-04501384/document https://hal.science/hal-04501384/file/remotesensing-15-02183.pdf https://doi.org/10.3390/rs15082183 |
long_lat |
ENVELOPE(-58.795,-58.795,-62.196,-62.196) ENVELOPE(-62.167,-62.167,-64.650,-64.650) |
geographic |
Gomez Navarro |
geographic_facet |
Gomez Navarro |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_source |
ISSN: 2072-4292 Remote Sensing https://hal.science/hal-04501384 Remote Sensing, 2023, 15 (8), pp.2183. ⟨10.3390/rs15082183⟩ |
op_relation |
info:eu-repo/semantics/altIdentifier/doi/10.3390/rs15082183 hal-04501384 https://hal.science/hal-04501384 https://hal.science/hal-04501384/document https://hal.science/hal-04501384/file/remotesensing-15-02183.pdf doi:10.3390/rs15082183 |
op_rights |
info:eu-repo/semantics/OpenAccess |
op_doi |
https://doi.org/10.3390/rs15082183 |
container_title |
Remote Sensing |
container_volume |
15 |
container_issue |
8 |
container_start_page |
2183 |
_version_ |
1809928462529462272 |