Using deep learning to count albatrosses from space
In this paper we test the use of a deep learning approach to automatically count Wandering Albatrosses in Very High Resolution (VHR) satellite imagery. We use a dataset of manually labelled imagery provided by the British Antarctic Survey to train and develop our methods. We employ a U-Net architect...
Published in: | IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium |
---|---|
Main Authors: | , , , |
Format: | Book Part |
Language: | unknown |
Published: |
IEEE
2019
|
Subjects: | |
Online Access: | http://nora.nerc.ac.uk/id/eprint/526568/ https://doi.org/10.1109/IGARSS.2019.8898079 |
id |
ftnerc:oai:nora.nerc.ac.uk:526568 |
---|---|
record_format |
openpolar |
spelling |
ftnerc:oai:nora.nerc.ac.uk:526568 2023-05-15T13:41:44+02:00 Using deep learning to count albatrosses from space Bowler, Ellen Fretwell, Peter T. French, Geoffrey Mackiewicz, Michal 2019-11 http://nora.nerc.ac.uk/id/eprint/526568/ https://doi.org/10.1109/IGARSS.2019.8898079 unknown IEEE Bowler, Ellen; Fretwell, Peter T. orcid:0000-0002-1988-5844 French, Geoffrey; Mackiewicz, Michal. 2019 Using deep learning to count albatrosses from space. In: IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 10099-10102. Publication - Book Section NonPeerReviewed 2019 ftnerc https://doi.org/10.1109/IGARSS.2019.8898079 2023-02-04T19:50:02Z In this paper we test the use of a deep learning approach to automatically count Wandering Albatrosses in Very High Resolution (VHR) satellite imagery. We use a dataset of manually labelled imagery provided by the British Antarctic Survey to train and develop our methods. We employ a U-Net architecture, designed for image segmentation, to simultaneously classify and localise potential albatrosses. We aid training with the use of the Focal Loss criterion, to deal with extreme class imbalance in the dataset. Initial results achieve peak precision and recall values of approximately 80%. Finally we assess the model's performance in relation to inter-observer variation, by comparing errors against an image labelled by multiple observers. We conclude model accuracy falls within the range of human counters. We hope that the methods will streamline the analysis of VHR satellite images, enabling more frequent monitoring of a species which is of high conservation concern. Book Part Antarc* Antarctic British Antarctic Survey Natural Environment Research Council: NERC Open Research Archive Antarctic IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium 10099 10102 |
institution |
Open Polar |
collection |
Natural Environment Research Council: NERC Open Research Archive |
op_collection_id |
ftnerc |
language |
unknown |
description |
In this paper we test the use of a deep learning approach to automatically count Wandering Albatrosses in Very High Resolution (VHR) satellite imagery. We use a dataset of manually labelled imagery provided by the British Antarctic Survey to train and develop our methods. We employ a U-Net architecture, designed for image segmentation, to simultaneously classify and localise potential albatrosses. We aid training with the use of the Focal Loss criterion, to deal with extreme class imbalance in the dataset. Initial results achieve peak precision and recall values of approximately 80%. Finally we assess the model's performance in relation to inter-observer variation, by comparing errors against an image labelled by multiple observers. We conclude model accuracy falls within the range of human counters. We hope that the methods will streamline the analysis of VHR satellite images, enabling more frequent monitoring of a species which is of high conservation concern. |
format |
Book Part |
author |
Bowler, Ellen Fretwell, Peter T. French, Geoffrey Mackiewicz, Michal |
spellingShingle |
Bowler, Ellen Fretwell, Peter T. French, Geoffrey Mackiewicz, Michal Using deep learning to count albatrosses from space |
author_facet |
Bowler, Ellen Fretwell, Peter T. French, Geoffrey Mackiewicz, Michal |
author_sort |
Bowler, Ellen |
title |
Using deep learning to count albatrosses from space |
title_short |
Using deep learning to count albatrosses from space |
title_full |
Using deep learning to count albatrosses from space |
title_fullStr |
Using deep learning to count albatrosses from space |
title_full_unstemmed |
Using deep learning to count albatrosses from space |
title_sort |
using deep learning to count albatrosses from space |
publisher |
IEEE |
publishDate |
2019 |
url |
http://nora.nerc.ac.uk/id/eprint/526568/ https://doi.org/10.1109/IGARSS.2019.8898079 |
geographic |
Antarctic |
geographic_facet |
Antarctic |
genre |
Antarc* Antarctic British Antarctic Survey |
genre_facet |
Antarc* Antarctic British Antarctic Survey |
op_relation |
Bowler, Ellen; Fretwell, Peter T. orcid:0000-0002-1988-5844 French, Geoffrey; Mackiewicz, Michal. 2019 Using deep learning to count albatrosses from space. In: IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 10099-10102. |
op_doi |
https://doi.org/10.1109/IGARSS.2019.8898079 |
container_title |
IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium |
container_start_page |
10099 |
op_container_end_page |
10102 |
_version_ |
1766156491391238144 |