SEALNET: Facial recognition software for ecological studies of harbor seals
Methods for long-term monitoring of coastal species such as harbor seals ( Phoca vitulina ), are often costly, time-consuming, and highly invasive, underscoring the need for improved techniques for data collection and analysis. Here, we propose the use of automated facial recognition technology for...
Main Authors: | , , , , , |
---|---|
Format: | Other/Unknown Material |
Language: | unknown |
Published: |
Zenodo
2022
|
Subjects: | |
Online Access: | https://doi.org/10.5281/zenodo.6415595 |
id |
ftzenodo:oai:zenodo.org:6415595 |
---|---|
record_format |
openpolar |
spelling |
ftzenodo:oai:zenodo.org:6415595 2024-09-15T18:30:22+00:00 SEALNET: Facial recognition software for ecological studies of harbor seals Birenbaum, Zach Do, Hieu Horstmyer, Lauren Orff, Hailey Ay, Ahmet Ingram, Krista 2022-04-05 https://doi.org/10.5281/zenodo.6415595 unknown Zenodo https://doi.org/10.5281/zenodo.6415594 https://doi.org/10.5281/zenodo.6415595 oai:zenodo.org:6415595 info:eu-repo/semantics/openAccess Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode facial recognition marine mammal Phoca vitulina info:eu-repo/semantics/other 2022 ftzenodo https://doi.org/10.5281/zenodo.641559510.5281/zenodo.6415594 2024-07-25T22:53:58Z Methods for long-term monitoring of coastal species such as harbor seals ( Phoca vitulina ), are often costly, time-consuming, and highly invasive, underscoring the need for improved techniques for data collection and analysis. Here, we propose the use of automated facial recognition technology for identification of individual seals and demonstrate its utility in ecological and population studies.We created a software package, SealNet, that automates photo identification of seals, using a graphical user interface (GUI) software to detect, align and chip seal faces from photographs and a deep convolutional neural network (CNN) suitable for small datasets (e.g., 100 seals with five photos per seal) to classify individual seals. We piloted the SealNet technology with a population of harbor seals located within Casco Bay on the coast of Maine, USA. Across two-years of sampling, 2019 and 2020, at seven haul-out sites in Middle Bay, we obtained a dataset optimized for development and testing of SealNet.We processed 1752 images representing 408 individual seals and achieved 88% Rank-1 and 96% Rank-5) accuracy in closed set seal identification. In identifying individual seals,SealNet software outperformed a similar face recognition method, PrimNet, developed for primates but retrained on seals. The ease and wealth of image data that can be processed using SealNet software contributes a vital tool for ecological and behavioral studies of marine mammals in the developing field of conservation technology. Other/Unknown Material Phoca vitulina Zenodo |
institution |
Open Polar |
collection |
Zenodo |
op_collection_id |
ftzenodo |
language |
unknown |
topic |
facial recognition marine mammal Phoca vitulina |
spellingShingle |
facial recognition marine mammal Phoca vitulina Birenbaum, Zach Do, Hieu Horstmyer, Lauren Orff, Hailey Ay, Ahmet Ingram, Krista SEALNET: Facial recognition software for ecological studies of harbor seals |
topic_facet |
facial recognition marine mammal Phoca vitulina |
description |
Methods for long-term monitoring of coastal species such as harbor seals ( Phoca vitulina ), are often costly, time-consuming, and highly invasive, underscoring the need for improved techniques for data collection and analysis. Here, we propose the use of automated facial recognition technology for identification of individual seals and demonstrate its utility in ecological and population studies.We created a software package, SealNet, that automates photo identification of seals, using a graphical user interface (GUI) software to detect, align and chip seal faces from photographs and a deep convolutional neural network (CNN) suitable for small datasets (e.g., 100 seals with five photos per seal) to classify individual seals. We piloted the SealNet technology with a population of harbor seals located within Casco Bay on the coast of Maine, USA. Across two-years of sampling, 2019 and 2020, at seven haul-out sites in Middle Bay, we obtained a dataset optimized for development and testing of SealNet.We processed 1752 images representing 408 individual seals and achieved 88% Rank-1 and 96% Rank-5) accuracy in closed set seal identification. In identifying individual seals,SealNet software outperformed a similar face recognition method, PrimNet, developed for primates but retrained on seals. The ease and wealth of image data that can be processed using SealNet software contributes a vital tool for ecological and behavioral studies of marine mammals in the developing field of conservation technology. |
format |
Other/Unknown Material |
author |
Birenbaum, Zach Do, Hieu Horstmyer, Lauren Orff, Hailey Ay, Ahmet Ingram, Krista |
author_facet |
Birenbaum, Zach Do, Hieu Horstmyer, Lauren Orff, Hailey Ay, Ahmet Ingram, Krista |
author_sort |
Birenbaum, Zach |
title |
SEALNET: Facial recognition software for ecological studies of harbor seals |
title_short |
SEALNET: Facial recognition software for ecological studies of harbor seals |
title_full |
SEALNET: Facial recognition software for ecological studies of harbor seals |
title_fullStr |
SEALNET: Facial recognition software for ecological studies of harbor seals |
title_full_unstemmed |
SEALNET: Facial recognition software for ecological studies of harbor seals |
title_sort |
sealnet: facial recognition software for ecological studies of harbor seals |
publisher |
Zenodo |
publishDate |
2022 |
url |
https://doi.org/10.5281/zenodo.6415595 |
genre |
Phoca vitulina |
genre_facet |
Phoca vitulina |
op_relation |
https://doi.org/10.5281/zenodo.6415594 https://doi.org/10.5281/zenodo.6415595 oai:zenodo.org:6415595 |
op_rights |
info:eu-repo/semantics/openAccess Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode |
op_doi |
https://doi.org/10.5281/zenodo.641559510.5281/zenodo.6415594 |
_version_ |
1810471831100981248 |