EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automatic identification of the endangered hawksbill sea turtle behavior using deep learning and cross-species transfer learning

Lorène Jeantet iD, Kukhanya Zondo, Cyrielle Delvenne, Jordan Martin, Damien Chevallier iD, Emmanuel Dufourq iD

Journal of Experimental Biology · 2024

DOI: 10.1242/jeb.249232 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Experimental Biology, 2024. Verfasst von Lorène Jeantet und 5 weiteren Autor:innen. Beteiligte Einrichtung: African Institute for Mathematical Sciences, Muizenberg, Cape Town, 7945, South Africa. Hauptthemen laut MEDLINE: Turtles, Deep Learning, Endangered Species. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Carnegie Corporation of New York, Fonds Européen de Développement Régional, Collectivité Territoriale de Martinique.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The accelerometer, an onboard sensor, enables remote monitoring of animal posture and movement, allowing researchers to deduce behaviors. Despite the automated analysis capabilities provided by deep learning, data scarcity remains a challenge in ecology. We explored transfer learning to classify behaviors from acceleration data of critically endangered hawksbill sea turtles (Eretmochelys imbricata). Transfer learning reuses a model trained on one task from a large dataset to solve a related task. We applied this method using a model trained on green turtles (Chelonia mydas) and adapted it to identify hawksbill behaviors such as swimming, resting and feeding. We also compared this with a model trained on human activity data. The results showed an 8% and 4% F1-score improvement with transfer learning from green turtle and human datasets, respectively. Transfer learning allows researchers to adapt existing models to their study species, leveraging deep learning and expanding the use of accelerometers for wildlife monitoring.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

TurtlesDeep LearningEndangered SpeciesAnimalsBehavior, AnimalAccelerometry

Stichwörter der Autor:innen

AccelerometerAnimal behaviorBehavioral classificationBio-loggingMachine LearningNeural networks

Autor:innen und Institutionen

  • African Institute for Mathematical Sciences, Muizenberg, Cape Town, 7945, South Africa.
  • African Institute for Mathematical Sciences, Research and Innovation Centre, KN3 Kigali, Rwanda.
  • Stellenbosch University, 7602, Stellenbosch, South Africa.
  • Unité de Recherche BOREA, MNHN, CNRS 8067, SU, IRD 207, UCN, UA, Station de Recherche Marine de Martinique, Quartier Degras, Petite Anse, 97217 Les Anses d'Arlet, Martinique, France.

ORCID-Profile: Lorène Jeantet, Damien Chevallier, Emmanuel Dufourq

Förderung

  • Carnegie Corporation of New York
  • Fonds Européen de Développement Régional
  • Collectivité Territoriale de Martinique
  • Direction de l'Environnement, de l'Aménagement et du Logement
  • Office De l'Eau
  • Office Français de la Biodiversité
  • Stellenbosch University

Zitieren

APA

Jeantet, L., Zondo, K., Delvenne, C., Martin, J., Chevallier, D., & Dufourq, E. (2024). Automatic identification of the endangered hawksbill sea turtle behavior using deep learning and cross-species transfer learning. Journal of Experimental Biology. https://doi.org/10.1242/jeb.249232

Vancouver

Jeantet L, Zondo K, Delvenne C, Martin J, Chevallier D, Dufourq E. Automatic identification of the endangered hawksbill sea turtle behavior using deep learning and cross-species transfer learning. Journal of Experimental Biology. 2024. doi:10.1242/jeb.249232

BibTeX
@article{Jeantet2024,
  author = {Jeantet, Lorène and Zondo, Kukhanya and Delvenne, Cyrielle and Martin, Jordan and Chevallier, Damien and Dufourq, Emmanuel},
  title = {Automatic identification of the endangered hawksbill sea turtle behavior using deep learning and cross-species transfer learning},
  journal = {Journal of Experimental Biology},
  year = {2024},
  doi = {10.1242/jeb.249232}
}
Kostenlos starten →