Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network
Application of a neural network to ERS-SAR images to retrieve pressure ridge spatial frequencies is presented. For an independent dataset, the rmserror between the retrieved and the true ridge frequency as determined by means of laser profiling was about 5 ridges per kilometre, or 30%. The network i...
Published in: | International Journal of Remote Sensing |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Taylor & Francis
1999
|
Subjects: | |
Online Access: | https://oceanrep.geomar.de/id/eprint/2561/ https://oceanrep.geomar.de/id/eprint/2561/1/Retrieval%20of%20Antarctic%20sea%20ice%20.pdf https://doi.org/10.1080/014311699211642 |
id |
ftoceanrep:oai:oceanrep.geomar.de:2561 |
---|---|
record_format |
openpolar |
spelling |
ftoceanrep:oai:oceanrep.geomar.de:2561 2023-05-15T13:43:14+02:00 Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network Haas, C. Liu, Quanhua Martin, Thomas 1999 text https://oceanrep.geomar.de/id/eprint/2561/ https://oceanrep.geomar.de/id/eprint/2561/1/Retrieval%20of%20Antarctic%20sea%20ice%20.pdf https://doi.org/10.1080/014311699211642 en eng Taylor & Francis https://oceanrep.geomar.de/id/eprint/2561/1/Retrieval%20of%20Antarctic%20sea%20ice%20.pdf Haas, C., Liu, Q. and Martin, T. (1999) Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network. International Journal of Remote Sensing, 20 (15&16). pp. 3111-3121. DOI 10.1080/014311699211642 <https://doi.org/10.1080/014311699211642>. doi:10.1080/014311699211642 info:eu-repo/semantics/restrictedAccess Article PeerReviewed 1999 ftoceanrep https://doi.org/10.1080/014311699211642 2023-04-07T14:46:00Z Application of a neural network to ERS-SAR images to retrieve pressure ridge spatial frequencies is presented. For an independent dataset, the rmserror between the retrieved and the true ridge frequency as determined by means of laser profiling was about 5 ridges per kilometre, or 30%. The network is trained with results from in situ laser profiling of ridge distributions and coincident SAR backscatter properties. The study focuses on summer data from the Bellingshausen, Amundsen and Weddell Seas in Antarctica, which were gathered in February 1994 and 1997. Pressure ridge frequencies varied from 3 to 30 ridges per kilometre between different regions, thus providing a wide range of training and test data for the algorithm development. From ERS-SAR images covering the area of the laser flights with a time difference of a few days at maximum, histograms of the backscatter coefficient sigma0 were extracted. Statistical parameters (e.g. mean, standard deviation, tail-to-mean ratio) were calculated from these distributions and compared with the results of the laser flights. Generally, the mean backscatter increases with a growing ridge frequency, and the signal range becomes narrower. However, these correlations are only poor, and improved results are obtained when the statistical parameters are combined to train the neural network. Article in Journal/Newspaper Antarc* Antarctic Antarctica Sea ice OceanRep (GEOMAR Helmholtz Centre für Ocean Research Kiel) Antarctic Weddell International Journal of Remote Sensing 20 15-16 3111 3123 |
institution |
Open Polar |
collection |
OceanRep (GEOMAR Helmholtz Centre für Ocean Research Kiel) |
op_collection_id |
ftoceanrep |
language |
English |
description |
Application of a neural network to ERS-SAR images to retrieve pressure ridge spatial frequencies is presented. For an independent dataset, the rmserror between the retrieved and the true ridge frequency as determined by means of laser profiling was about 5 ridges per kilometre, or 30%. The network is trained with results from in situ laser profiling of ridge distributions and coincident SAR backscatter properties. The study focuses on summer data from the Bellingshausen, Amundsen and Weddell Seas in Antarctica, which were gathered in February 1994 and 1997. Pressure ridge frequencies varied from 3 to 30 ridges per kilometre between different regions, thus providing a wide range of training and test data for the algorithm development. From ERS-SAR images covering the area of the laser flights with a time difference of a few days at maximum, histograms of the backscatter coefficient sigma0 were extracted. Statistical parameters (e.g. mean, standard deviation, tail-to-mean ratio) were calculated from these distributions and compared with the results of the laser flights. Generally, the mean backscatter increases with a growing ridge frequency, and the signal range becomes narrower. However, these correlations are only poor, and improved results are obtained when the statistical parameters are combined to train the neural network. |
format |
Article in Journal/Newspaper |
author |
Haas, C. Liu, Quanhua Martin, Thomas |
spellingShingle |
Haas, C. Liu, Quanhua Martin, Thomas Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network |
author_facet |
Haas, C. Liu, Quanhua Martin, Thomas |
author_sort |
Haas, C. |
title |
Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network |
title_short |
Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network |
title_full |
Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network |
title_fullStr |
Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network |
title_full_unstemmed |
Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network |
title_sort |
retrieval of antarctic sea-ice pressure ridge frequencies from ers sar imagery by means of in situ laser profiling and usage of a neural network |
publisher |
Taylor & Francis |
publishDate |
1999 |
url |
https://oceanrep.geomar.de/id/eprint/2561/ https://oceanrep.geomar.de/id/eprint/2561/1/Retrieval%20of%20Antarctic%20sea%20ice%20.pdf https://doi.org/10.1080/014311699211642 |
geographic |
Antarctic Weddell |
geographic_facet |
Antarctic Weddell |
genre |
Antarc* Antarctic Antarctica Sea ice |
genre_facet |
Antarc* Antarctic Antarctica Sea ice |
op_relation |
https://oceanrep.geomar.de/id/eprint/2561/1/Retrieval%20of%20Antarctic%20sea%20ice%20.pdf Haas, C., Liu, Q. and Martin, T. (1999) Retrieval of Antarctic sea-ice pressure ridge frequencies from ERS SAR imagery by means of in situ laser profiling and usage of a neural network. International Journal of Remote Sensing, 20 (15&16). pp. 3111-3121. DOI 10.1080/014311699211642 <https://doi.org/10.1080/014311699211642>. doi:10.1080/014311699211642 |
op_rights |
info:eu-repo/semantics/restrictedAccess |
op_doi |
https://doi.org/10.1080/014311699211642 |
container_title |
International Journal of Remote Sensing |
container_volume |
20 |
container_issue |
15-16 |
container_start_page |
3111 |
op_container_end_page |
3123 |
_version_ |
1766186127657533440 |