Physics-aware feature learning of SAR images with deep neural networks: A case study
This paper proposes a novel unsupervised learning method to learn discriminative physics-aware features of Synthetic Aperture Radar images with deep neural networks. We conduct a case study of sea-ice classification using Sentinel-1Dual-polarized SAR data and the corresponding scattering mechanisms...
Published in: | 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS |
---|---|
Main Authors: | , , |
Format: | Conference Object |
Language: | unknown |
Published: |
2021
|
Subjects: | |
Online Access: | https://elib.dlr.de/142805/ https://igarss2021.com/view_paper.php?PaperNum=2617 |
id |
ftdlr:oai:elib.dlr.de:142805 |
---|---|
record_format |
openpolar |
spelling |
ftdlr:oai:elib.dlr.de:142805 2024-05-19T07:48:20+00:00 Physics-aware feature learning of SAR images with deep neural networks: A case study Huang, Zhongling Dumitru, Corneliu Octavian Ren, Jun 2021-07 https://elib.dlr.de/142805/ https://igarss2021.com/view_paper.php?PaperNum=2617 unknown Huang, Zhongling und Dumitru, Corneliu Octavian und Ren, Jun (2021) Physics-aware feature learning of SAR images with deep neural networks: A case study. In: International Geoscience and Remote Sensing Symposium (IGARSS), Seiten 1-4. IGARSS 2021, 2021-07-12 - 2021-07-16, Brussels, Belgium. doi:10.1109/IGARSS47720.2021.9554842 <https://doi.org/10.1109/IGARSS47720.2021.9554842>. EO Data Science Konferenzbeitrag PeerReviewed 2021 ftdlr https://doi.org/10.1109/IGARSS47720.2021.9554842 2024-04-25T00:56:38Z This paper proposes a novel unsupervised learning method to learn discriminative physics-aware features of Synthetic Aperture Radar images with deep neural networks. We conduct a case study of sea-ice classification using Sentinel-1Dual-polarized SAR data and the corresponding scattering mechanisms derived from H/αWishart classification. The scattering mechanisms are encoded as a combination of topics for each SAR image as physics attributes, which guide the deep convolutional neural network to learn physics-aware features automatically. A novel objective function is designed to demonstrate how to conduct the physics-guided learning processing. The experiments show the proposed method can learn discriminative features from SAR images without labeled data, which can achieve a comparable classification result with supervised CNN learning. Conference Object Sea ice German Aerospace Center: elib - DLR electronic library 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS 1264 1267 |
institution |
Open Polar |
collection |
German Aerospace Center: elib - DLR electronic library |
op_collection_id |
ftdlr |
language |
unknown |
topic |
EO Data Science |
spellingShingle |
EO Data Science Huang, Zhongling Dumitru, Corneliu Octavian Ren, Jun Physics-aware feature learning of SAR images with deep neural networks: A case study |
topic_facet |
EO Data Science |
description |
This paper proposes a novel unsupervised learning method to learn discriminative physics-aware features of Synthetic Aperture Radar images with deep neural networks. We conduct a case study of sea-ice classification using Sentinel-1Dual-polarized SAR data and the corresponding scattering mechanisms derived from H/αWishart classification. The scattering mechanisms are encoded as a combination of topics for each SAR image as physics attributes, which guide the deep convolutional neural network to learn physics-aware features automatically. A novel objective function is designed to demonstrate how to conduct the physics-guided learning processing. The experiments show the proposed method can learn discriminative features from SAR images without labeled data, which can achieve a comparable classification result with supervised CNN learning. |
format |
Conference Object |
author |
Huang, Zhongling Dumitru, Corneliu Octavian Ren, Jun |
author_facet |
Huang, Zhongling Dumitru, Corneliu Octavian Ren, Jun |
author_sort |
Huang, Zhongling |
title |
Physics-aware feature learning of SAR images with deep neural networks: A case study |
title_short |
Physics-aware feature learning of SAR images with deep neural networks: A case study |
title_full |
Physics-aware feature learning of SAR images with deep neural networks: A case study |
title_fullStr |
Physics-aware feature learning of SAR images with deep neural networks: A case study |
title_full_unstemmed |
Physics-aware feature learning of SAR images with deep neural networks: A case study |
title_sort |
physics-aware feature learning of sar images with deep neural networks: a case study |
publishDate |
2021 |
url |
https://elib.dlr.de/142805/ https://igarss2021.com/view_paper.php?PaperNum=2617 |
genre |
Sea ice |
genre_facet |
Sea ice |
op_relation |
Huang, Zhongling und Dumitru, Corneliu Octavian und Ren, Jun (2021) Physics-aware feature learning of SAR images with deep neural networks: A case study. In: International Geoscience and Remote Sensing Symposium (IGARSS), Seiten 1-4. IGARSS 2021, 2021-07-12 - 2021-07-16, Brussels, Belgium. doi:10.1109/IGARSS47720.2021.9554842 <https://doi.org/10.1109/IGARSS47720.2021.9554842>. |
op_doi |
https://doi.org/10.1109/IGARSS47720.2021.9554842 |
container_title |
2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS |
container_start_page |
1264 |
op_container_end_page |
1267 |
_version_ |
1799466546017337344 |