Localization and tracking of beluga whales in aerial video using deep learning

Aerial images are increasingly adopted and widely used in various research areas. In marine mammal studies, these imagery surveys serve multiple purposes: determining population size, mapping migration routes, and gaining behavioral insights. A single aerial scan using a drone yields a wealth of dat...

Full description

Saved in:
Bibliographic Details
Main Authors: Mostapha Alsaidi, Mohammed G. Al-Jassani, Chiron Bang, Gregory O’Corry-Crowe, Cortney Watt, Maha Ghazal, Hanqi Zhuang
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-11-01
Series:Frontiers in Marine Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fmars.2024.1445698/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846168023536238592
author Mostapha Alsaidi
Mohammed G. Al-Jassani
Chiron Bang
Gregory O’Corry-Crowe
Cortney Watt
Maha Ghazal
Hanqi Zhuang
author_facet Mostapha Alsaidi
Mohammed G. Al-Jassani
Chiron Bang
Gregory O’Corry-Crowe
Cortney Watt
Maha Ghazal
Hanqi Zhuang
author_sort Mostapha Alsaidi
collection DOAJ
description Aerial images are increasingly adopted and widely used in various research areas. In marine mammal studies, these imagery surveys serve multiple purposes: determining population size, mapping migration routes, and gaining behavioral insights. A single aerial scan using a drone yields a wealth of data, but processing it requires significant human effort. Our research demonstrates that deep learning models can significantly reduce human effort. They are not only able to detect marine mammals but also track their behavior using continuous aerial (video) footage. By distinguishing between different age classes, these algorithms can inform studies on population biology, ontogeny, and adult-calf relationships. To detect beluga whales from imagery footage, we trained the YOLOv7 model on a proprietary dataset of aerial footage of beluga whales. The deep learning model achieved impressive results with the following precision and recall scores: beluga adult = 92%—92%, beluga calf = 94%—89%. To track the detected beluga whales, we implemented the deep Simple Online and Realtime Tracking (SORT) algorithm. Unfortunately, the performance of the deep SORT algorithm was disappointing, with Multiple Object Tracking Accuracy (MOTA) scores ranging from 27% to 48%. An analysis revealed that the low tracking accuracy resulted from identity switching; that is, an identical beluga whale was given two IDs in two different frames. To overcome the problem of identity switching, a new post-processing algorithm was implemented, significantly improving MOTA to approximately 70%. The main contribution of this research is providing a system that accurately detects and tracks features of beluga whales, both adults and calves, from aerial footage. Additionally, this system can be customized to identify and analyze other marine mammal species by fine-tuning the model with annotated data.
format Article
id doaj-art-366895ac476e4292806700f82a969627
institution Kabale University
issn 2296-7745
language English
publishDate 2024-11-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Marine Science
spelling doaj-art-366895ac476e4292806700f82a9696272024-11-14T09:38:49ZengFrontiers Media S.A.Frontiers in Marine Science2296-77452024-11-011110.3389/fmars.2024.14456981445698Localization and tracking of beluga whales in aerial video using deep learningMostapha Alsaidi0Mohammed G. Al-Jassani1Chiron Bang2Gregory O’Corry-Crowe3Cortney Watt4Maha Ghazal5Hanqi Zhuang6Department of Electrical Engineering and Computer Science, Florida Atlantic University, Boca Raton, FL, United StatesDepartment of Electrical Engineering and Computer Science, Florida Atlantic University, Boca Raton, FL, United StatesDepartment of Electrical Engineering and Computer Science, Florida Atlantic University, Boca Raton, FL, United StatesHarbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, FL, United StatesFisheries and Oceans Canada, Freshwater Institute, Winnipeg, MB, CanadaFisheries and Oceans Canada, Freshwater Institute, Winnipeg, MB, CanadaDepartment of Electrical Engineering and Computer Science, Florida Atlantic University, Boca Raton, FL, United StatesAerial images are increasingly adopted and widely used in various research areas. In marine mammal studies, these imagery surveys serve multiple purposes: determining population size, mapping migration routes, and gaining behavioral insights. A single aerial scan using a drone yields a wealth of data, but processing it requires significant human effort. Our research demonstrates that deep learning models can significantly reduce human effort. They are not only able to detect marine mammals but also track their behavior using continuous aerial (video) footage. By distinguishing between different age classes, these algorithms can inform studies on population biology, ontogeny, and adult-calf relationships. To detect beluga whales from imagery footage, we trained the YOLOv7 model on a proprietary dataset of aerial footage of beluga whales. The deep learning model achieved impressive results with the following precision and recall scores: beluga adult = 92%—92%, beluga calf = 94%—89%. To track the detected beluga whales, we implemented the deep Simple Online and Realtime Tracking (SORT) algorithm. Unfortunately, the performance of the deep SORT algorithm was disappointing, with Multiple Object Tracking Accuracy (MOTA) scores ranging from 27% to 48%. An analysis revealed that the low tracking accuracy resulted from identity switching; that is, an identical beluga whale was given two IDs in two different frames. To overcome the problem of identity switching, a new post-processing algorithm was implemented, significantly improving MOTA to approximately 70%. The main contribution of this research is providing a system that accurately detects and tracks features of beluga whales, both adults and calves, from aerial footage. Additionally, this system can be customized to identify and analyze other marine mammal species by fine-tuning the model with annotated data.https://www.frontiersin.org/articles/10.3389/fmars.2024.1445698/fullmarine mammalsbeluga whalelocalizationtrackingdeep learningaerial video footage
spellingShingle Mostapha Alsaidi
Mohammed G. Al-Jassani
Chiron Bang
Gregory O’Corry-Crowe
Cortney Watt
Maha Ghazal
Hanqi Zhuang
Localization and tracking of beluga whales in aerial video using deep learning
Frontiers in Marine Science
marine mammals
beluga whale
localization
tracking
deep learning
aerial video footage
title Localization and tracking of beluga whales in aerial video using deep learning
title_full Localization and tracking of beluga whales in aerial video using deep learning
title_fullStr Localization and tracking of beluga whales in aerial video using deep learning
title_full_unstemmed Localization and tracking of beluga whales in aerial video using deep learning
title_short Localization and tracking of beluga whales in aerial video using deep learning
title_sort localization and tracking of beluga whales in aerial video using deep learning
topic marine mammals
beluga whale
localization
tracking
deep learning
aerial video footage
url https://www.frontiersin.org/articles/10.3389/fmars.2024.1445698/full
work_keys_str_mv AT mostaphaalsaidi localizationandtrackingofbelugawhalesinaerialvideousingdeeplearning
AT mohammedgaljassani localizationandtrackingofbelugawhalesinaerialvideousingdeeplearning
AT chironbang localizationandtrackingofbelugawhalesinaerialvideousingdeeplearning
AT gregoryocorrycrowe localizationandtrackingofbelugawhalesinaerialvideousingdeeplearning
AT cortneywatt localizationandtrackingofbelugawhalesinaerialvideousingdeeplearning
AT mahaghazal localizationandtrackingofbelugawhalesinaerialvideousingdeeplearning
AT hanqizhuang localizationandtrackingofbelugawhalesinaerialvideousingdeeplearning