Sea Ice Detection from GNSS-R Data Based on Residual Network
Sea ice is an important component of the polar circle and influences atmospheric change. Global navigation satellite system reflectometry (GNSS-R) not only realizes time-continuous and wide-area sea ice detection, but also greatly reduces the cost of sea ice remote sensing research, which has been a...
Main Authors: | , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2023
|
Subjects: | |
Online Access: | https://depositonce.tu-berlin.de/handle/11303/20159 https://doi.org/10.14279/depositonce-18957 |
id |
ftdepositonce:oai:depositonce.tu-berlin.de:11303/20159 |
---|---|
record_format |
openpolar |
spelling |
ftdepositonce:oai:depositonce.tu-berlin.de:11303/20159 2023-11-12T04:25:50+01:00 Sea Ice Detection from GNSS-R Data Based on Residual Network Hu, Yuan Hua, Xifan Liu, Wei Wickert, Jens 2023-09-12 application/pdf https://depositonce.tu-berlin.de/handle/11303/20159 https://doi.org/10.14279/depositonce-18957 en eng https://depositonce.tu-berlin.de/handle/11303/20159 https://doi.org/10.14279/depositonce-18957 2072-4292 https://creativecommons.org/licenses/by/4.0/ 500 Naturwissenschaften und Mathematik::550 Geowissenschaften Geologie::550 Geowissenschaften delay-Doppler maps global navigation satellite system-reflectometry convolutional neural networks sea ice detection DDMs GNSS-R CNNs Article publishedVersion 2023 ftdepositonce https://doi.org/10.14279/depositonce-18957 2023-10-30T17:19:30Z Sea ice is an important component of the polar circle and influences atmospheric change. Global navigation satellite system reflectometry (GNSS-R) not only realizes time-continuous and wide-area sea ice detection, but also greatly reduces the cost of sea ice remote sensing research, which has been a hot topic in recent years. To tackle the challenges of noise interference and the reduced accuracy of sea ice detection during the melting period, this paper proposes a sea ice detection method based on a residual neural network (ResNet). ResNet addresses the issue of vanishing gradients in deep neural networks and introduces residual connections, which allows the network to reuse learned features from previous layers. Delay-Doppler maps (DDMs) collected from TechDemoSat-1 (TDS-1) are used as input, and National Oceanic and Atmospheric Administration (NOAA) surface-type data above 60°N are selected as the true values. Based on ResNet, the sea ice detection achieved an accuracy of 98.61%, demonstrating high robustness to noise and strong stability during the sea ice melting period (June to September). In comparison to other sea ice detection algorithms, it stands out with its advantages of high accuracy, stability, and insensitivity to noise. Article in Journal/Newspaper Sea ice TU Berlin: Deposit Once |
institution |
Open Polar |
collection |
TU Berlin: Deposit Once |
op_collection_id |
ftdepositonce |
language |
English |
topic |
500 Naturwissenschaften und Mathematik::550 Geowissenschaften Geologie::550 Geowissenschaften delay-Doppler maps global navigation satellite system-reflectometry convolutional neural networks sea ice detection DDMs GNSS-R CNNs |
spellingShingle |
500 Naturwissenschaften und Mathematik::550 Geowissenschaften Geologie::550 Geowissenschaften delay-Doppler maps global navigation satellite system-reflectometry convolutional neural networks sea ice detection DDMs GNSS-R CNNs Hu, Yuan Hua, Xifan Liu, Wei Wickert, Jens Sea Ice Detection from GNSS-R Data Based on Residual Network |
topic_facet |
500 Naturwissenschaften und Mathematik::550 Geowissenschaften Geologie::550 Geowissenschaften delay-Doppler maps global navigation satellite system-reflectometry convolutional neural networks sea ice detection DDMs GNSS-R CNNs |
description |
Sea ice is an important component of the polar circle and influences atmospheric change. Global navigation satellite system reflectometry (GNSS-R) not only realizes time-continuous and wide-area sea ice detection, but also greatly reduces the cost of sea ice remote sensing research, which has been a hot topic in recent years. To tackle the challenges of noise interference and the reduced accuracy of sea ice detection during the melting period, this paper proposes a sea ice detection method based on a residual neural network (ResNet). ResNet addresses the issue of vanishing gradients in deep neural networks and introduces residual connections, which allows the network to reuse learned features from previous layers. Delay-Doppler maps (DDMs) collected from TechDemoSat-1 (TDS-1) are used as input, and National Oceanic and Atmospheric Administration (NOAA) surface-type data above 60°N are selected as the true values. Based on ResNet, the sea ice detection achieved an accuracy of 98.61%, demonstrating high robustness to noise and strong stability during the sea ice melting period (June to September). In comparison to other sea ice detection algorithms, it stands out with its advantages of high accuracy, stability, and insensitivity to noise. |
format |
Article in Journal/Newspaper |
author |
Hu, Yuan Hua, Xifan Liu, Wei Wickert, Jens |
author_facet |
Hu, Yuan Hua, Xifan Liu, Wei Wickert, Jens |
author_sort |
Hu, Yuan |
title |
Sea Ice Detection from GNSS-R Data Based on Residual Network |
title_short |
Sea Ice Detection from GNSS-R Data Based on Residual Network |
title_full |
Sea Ice Detection from GNSS-R Data Based on Residual Network |
title_fullStr |
Sea Ice Detection from GNSS-R Data Based on Residual Network |
title_full_unstemmed |
Sea Ice Detection from GNSS-R Data Based on Residual Network |
title_sort |
sea ice detection from gnss-r data based on residual network |
publishDate |
2023 |
url |
https://depositonce.tu-berlin.de/handle/11303/20159 https://doi.org/10.14279/depositonce-18957 |
genre |
Sea ice |
genre_facet |
Sea ice |
op_relation |
https://depositonce.tu-berlin.de/handle/11303/20159 https://doi.org/10.14279/depositonce-18957 2072-4292 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.14279/depositonce-18957 |
_version_ |
1782340001289207808 |