Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks
Classification of ships and icebergs in the Arctic in satellite images is an important problem. We study how to train deep neural networks for improving the discrimination of ships and icebergs in multispectral satellite images. We also analyze synthetic-aperture radar (SAR) images for comparison. T...
Published in: | Remote Sensing |
---|---|
Main Author: | |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2020
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs12152353 https://doaj.org/article/9088f12279c4428f90af6bc20ec9d418 |
id |
ftdoajarticles:oai:doaj.org/article:9088f12279c4428f90af6bc20ec9d418 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:9088f12279c4428f90af6bc20ec9d418 2023-05-15T14:57:53+02:00 Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks Henning Heiselberg 2020-07-01T00:00:00Z https://doi.org/10.3390/rs12152353 https://doaj.org/article/9088f12279c4428f90af6bc20ec9d418 EN eng MDPI AG https://www.mdpi.com/2072-4292/12/15/2353 https://doaj.org/toc/2072-4292 doi:10.3390/rs12152353 2072-4292 https://doaj.org/article/9088f12279c4428f90af6bc20ec9d418 Remote Sensing, Vol 12, Iss 2353, p 2353 (2020) Sentinel multispectral SAR ship iceberg convolutional neural networks Science Q article 2020 ftdoajarticles https://doi.org/10.3390/rs12152353 2022-12-31T07:32:45Z Classification of ships and icebergs in the Arctic in satellite images is an important problem. We study how to train deep neural networks for improving the discrimination of ships and icebergs in multispectral satellite images. We also analyze synthetic-aperture radar (SAR) images for comparison. The annotated datasets of ships and icebergs are collected from multispectral Sentinel-2 data and taken from the C-CORE dataset of Sentinel-1 SAR images. Convolutional Neural Networks with a range of hyperparameters are tested and optimized. Classification accuracies are considerably better for deep neural networks than for support vector machines. Deeper neural nets improve the accuracy per epoch but at the cost of longer processing time. Extending the datasets with semi-supervised data from Greenland improves the accuracy considerably whereas data augmentation by rotating and flipping the images has little effect. The resulting classification accuracies for ships and icebergs are 86% for the SAR data and 96% for the MSI data due to the better resolution and more multispectral bands. The size and quality of the datasets are essential for training the deep neural networks, and methods to improve them are discussed. The reduced false alarm rates and exploitation of multisensory data are important for Arctic search and rescue services. Article in Journal/Newspaper Arctic Greenland Iceberg* Directory of Open Access Journals: DOAJ Articles Arctic Greenland Remote Sensing 12 15 2353 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Sentinel multispectral SAR ship iceberg convolutional neural networks Science Q |
spellingShingle |
Sentinel multispectral SAR ship iceberg convolutional neural networks Science Q Henning Heiselberg Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks |
topic_facet |
Sentinel multispectral SAR ship iceberg convolutional neural networks Science Q |
description |
Classification of ships and icebergs in the Arctic in satellite images is an important problem. We study how to train deep neural networks for improving the discrimination of ships and icebergs in multispectral satellite images. We also analyze synthetic-aperture radar (SAR) images for comparison. The annotated datasets of ships and icebergs are collected from multispectral Sentinel-2 data and taken from the C-CORE dataset of Sentinel-1 SAR images. Convolutional Neural Networks with a range of hyperparameters are tested and optimized. Classification accuracies are considerably better for deep neural networks than for support vector machines. Deeper neural nets improve the accuracy per epoch but at the cost of longer processing time. Extending the datasets with semi-supervised data from Greenland improves the accuracy considerably whereas data augmentation by rotating and flipping the images has little effect. The resulting classification accuracies for ships and icebergs are 86% for the SAR data and 96% for the MSI data due to the better resolution and more multispectral bands. The size and quality of the datasets are essential for training the deep neural networks, and methods to improve them are discussed. The reduced false alarm rates and exploitation of multisensory data are important for Arctic search and rescue services. |
format |
Article in Journal/Newspaper |
author |
Henning Heiselberg |
author_facet |
Henning Heiselberg |
author_sort |
Henning Heiselberg |
title |
Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks |
title_short |
Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks |
title_full |
Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks |
title_fullStr |
Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks |
title_full_unstemmed |
Ship-Iceberg Classification in SAR and Multispectral Satellite Images with Neural Networks |
title_sort |
ship-iceberg classification in sar and multispectral satellite images with neural networks |
publisher |
MDPI AG |
publishDate |
2020 |
url |
https://doi.org/10.3390/rs12152353 https://doaj.org/article/9088f12279c4428f90af6bc20ec9d418 |
geographic |
Arctic Greenland |
geographic_facet |
Arctic Greenland |
genre |
Arctic Greenland Iceberg* |
genre_facet |
Arctic Greenland Iceberg* |
op_source |
Remote Sensing, Vol 12, Iss 2353, p 2353 (2020) |
op_relation |
https://www.mdpi.com/2072-4292/12/15/2353 https://doaj.org/toc/2072-4292 doi:10.3390/rs12152353 2072-4292 https://doaj.org/article/9088f12279c4428f90af6bc20ec9d418 |
op_doi |
https://doi.org/10.3390/rs12152353 |
container_title |
Remote Sensing |
container_volume |
12 |
container_issue |
15 |
container_start_page |
2353 |
_version_ |
1766329988341039104 |