A convolutional neural network architecture designed for the automated survey of seabird colonies
Abstract Satellite imagery is now well established as a method of finding and estimating the abundance of Antarctic penguin colonies. However, the delineation and classification of penguin colonies in sub‐meter satellite imagery has required the use of expert observers and is highly labor intensive,...
Published in: | Remote Sensing in Ecology and Conservation |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wiley
2022
|
Subjects: | |
Online Access: | https://doi.org/10.1002/rse2.240 https://doaj.org/article/4e873627ddcc41098f0fcfe536585e9b |
id |
ftdoajarticles:oai:doaj.org/article:4e873627ddcc41098f0fcfe536585e9b |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:4e873627ddcc41098f0fcfe536585e9b 2023-05-15T13:37:39+02:00 A convolutional neural network architecture designed for the automated survey of seabird colonies Hieu Le Dimitris Samaras Heather J. Lynch 2022-04-01T00:00:00Z https://doi.org/10.1002/rse2.240 https://doaj.org/article/4e873627ddcc41098f0fcfe536585e9b EN eng Wiley https://doi.org/10.1002/rse2.240 https://doaj.org/toc/2056-3485 2056-3485 doi:10.1002/rse2.240 https://doaj.org/article/4e873627ddcc41098f0fcfe536585e9b Remote Sensing in Ecology and Conservation, Vol 8, Iss 2, Pp 251-262 (2022) Adélie penguin convolutional neural network high‐resolution satellite imagery prior information Technology T Ecology QH540-549.5 article 2022 ftdoajarticles https://doi.org/10.1002/rse2.240 2022-12-31T00:28:16Z Abstract Satellite imagery is now well established as a method of finding and estimating the abundance of Antarctic penguin colonies. However, the delineation and classification of penguin colonies in sub‐meter satellite imagery has required the use of expert observers and is highly labor intensive, precluding regular censuses at the pan‐Antarctic scale. Here we present the first automated pipeline for the segmentation and classification of seabird colonies in high‐resolution satellite imagery. Our method leverages site‐fidelity by using images from previous years to improve classification performance but is robust to georegistration artifacts imposed by misalignment between sensors or terrain correction. We use a segmentation network with an additional branch that extracts the useful information from the prior mask of the input image. This prior branch provides the main model information on the location and size of guano in a prior annotation yet automatically learns to compensate for potential misalignment between the prior mask and the input image being classified. Our approach outperforms the previous approach by 44%, improving the average Intersection‐over‐Union segmentation score from 0.34 to 0.50. While penguin guano remains a challenging target for segmentation due to its indistinct and highly variable appearance, the inclusion of prior information represents a key step toward automated image annotation for population monitoring. Moreover, this method can be adapted for other ecological applications where the dynamics of landscape change are slow relative to the repeat frequency of available imagery and prior information may be available to aid with image annotation. Article in Journal/Newspaper Antarc* Antarctic Directory of Open Access Journals: DOAJ Articles Antarctic Guano ENVELOPE(141.604,141.604,-66.775,-66.775) Remote Sensing in Ecology and Conservation 8 2 251 262 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Adélie penguin convolutional neural network high‐resolution satellite imagery prior information Technology T Ecology QH540-549.5 |
spellingShingle |
Adélie penguin convolutional neural network high‐resolution satellite imagery prior information Technology T Ecology QH540-549.5 Hieu Le Dimitris Samaras Heather J. Lynch A convolutional neural network architecture designed for the automated survey of seabird colonies |
topic_facet |
Adélie penguin convolutional neural network high‐resolution satellite imagery prior information Technology T Ecology QH540-549.5 |
description |
Abstract Satellite imagery is now well established as a method of finding and estimating the abundance of Antarctic penguin colonies. However, the delineation and classification of penguin colonies in sub‐meter satellite imagery has required the use of expert observers and is highly labor intensive, precluding regular censuses at the pan‐Antarctic scale. Here we present the first automated pipeline for the segmentation and classification of seabird colonies in high‐resolution satellite imagery. Our method leverages site‐fidelity by using images from previous years to improve classification performance but is robust to georegistration artifacts imposed by misalignment between sensors or terrain correction. We use a segmentation network with an additional branch that extracts the useful information from the prior mask of the input image. This prior branch provides the main model information on the location and size of guano in a prior annotation yet automatically learns to compensate for potential misalignment between the prior mask and the input image being classified. Our approach outperforms the previous approach by 44%, improving the average Intersection‐over‐Union segmentation score from 0.34 to 0.50. While penguin guano remains a challenging target for segmentation due to its indistinct and highly variable appearance, the inclusion of prior information represents a key step toward automated image annotation for population monitoring. Moreover, this method can be adapted for other ecological applications where the dynamics of landscape change are slow relative to the repeat frequency of available imagery and prior information may be available to aid with image annotation. |
format |
Article in Journal/Newspaper |
author |
Hieu Le Dimitris Samaras Heather J. Lynch |
author_facet |
Hieu Le Dimitris Samaras Heather J. Lynch |
author_sort |
Hieu Le |
title |
A convolutional neural network architecture designed for the automated survey of seabird colonies |
title_short |
A convolutional neural network architecture designed for the automated survey of seabird colonies |
title_full |
A convolutional neural network architecture designed for the automated survey of seabird colonies |
title_fullStr |
A convolutional neural network architecture designed for the automated survey of seabird colonies |
title_full_unstemmed |
A convolutional neural network architecture designed for the automated survey of seabird colonies |
title_sort |
convolutional neural network architecture designed for the automated survey of seabird colonies |
publisher |
Wiley |
publishDate |
2022 |
url |
https://doi.org/10.1002/rse2.240 https://doaj.org/article/4e873627ddcc41098f0fcfe536585e9b |
long_lat |
ENVELOPE(141.604,141.604,-66.775,-66.775) |
geographic |
Antarctic Guano |
geographic_facet |
Antarctic Guano |
genre |
Antarc* Antarctic |
genre_facet |
Antarc* Antarctic |
op_source |
Remote Sensing in Ecology and Conservation, Vol 8, Iss 2, Pp 251-262 (2022) |
op_relation |
https://doi.org/10.1002/rse2.240 https://doaj.org/toc/2056-3485 2056-3485 doi:10.1002/rse2.240 https://doaj.org/article/4e873627ddcc41098f0fcfe536585e9b |
op_doi |
https://doi.org/10.1002/rse2.240 |
container_title |
Remote Sensing in Ecology and Conservation |
container_volume |
8 |
container_issue |
2 |
container_start_page |
251 |
op_container_end_page |
262 |
_version_ |
1766095572909948928 |