EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations

Francesco Di Nardo iD, Rocco De Marco iD, Daniel Li Veli iD, Laura Screpanti iD, Benedetta Castagna iD, Alessandro Lucchetti iD, David Scaradozzi iD

Sensors · 2025

DOI: 10.3390/s25082499 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Francesco Di Nardo und 6 weiteren Autor:innen. Beteiligte Einrichtung: Dipartimento di Ingegneria dell'informazione, Università Politecnica delle Marche, 60131 Ancona, Italy. Hauptthemen laut MEDLINE: Vocalization, Animal, Bottle-Nosed Dolphin, Neural Networks, Computer, Dolphins. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Recovery and Resilience Plan (NRRP), Mission 4 Component 2 Investment 1.4 NextGenerationEU, LIFE Financial Instrument of the European Community, Life Delfi Project—Dolphin Experience: Lowering Fishing Interactions.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Monitoring dolphins in the open sea is essential for understanding their behavior and the impact of human activities on the marine ecosystems. Passive Acoustic Monitoring (PAM) is a non-invasive technique for tracking dolphins, providing continuous data. This study presents a novel approach for classifying dolphin vocalizations from a PAM acoustic recording using a convolutional neural network (CNN). Four types of common bottlenose dolphin (Tursiops truncatus) vocalizations were identified from underwater recordings: whistles, echolocation clicks, burst pulse sounds, and feeding buzzes. To enhance classification performances, edge-detection filters were applied to spectrograms, with the aim of removing unwanted noise components. A dataset of nearly 10,000 spectrograms was used to train and test the CNN through a 10-fold cross-validation procedure. The results showed that the CNN achieved an average accuracy of 95.2% and an F1-score of 87.8%. The class-specific results showed a high accuracy for whistles (97.9%), followed by echolocation clicks (94.5%), feeding buzzes (94.0%), and burst pulse sounds (92.3%). The highest F1-score was obtained for whistles, exceeding 95%, while the other three vocalization typologies maintained an F1-score above 80%. This method provides a promising step toward improving the passive acoustic monitoring of dolphins, contributing to both species conservation and the mitigation of conflicts with fisheries.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Vocalization, AnimalBottle-Nosed DolphinNeural Networks, ComputerDolphinsAnimalsAcousticsEcholocationSound Spectrography

Stichwörter der Autor:innen

convolutional neural networksdeep learningdolphinspassive acoustic monitoring

Autor:innen und Institutionen

  • Dipartimento di Ingegneria dell'informazione, Università Politecnica delle Marche, 60131 Ancona, Italy.
  • Institute of Biological Resources and Marine Biotechnology (IRBIM), National Research Council (CNR), 60125 Ancona, Italy.
  • ANcybernetics, Università Politecnica delle Marche, 60131 Ancona, Italy.
  • National Biodiversity Future Center, 90133 Palermo, Italy.

ORCID-Profile: Francesco Di Nardo, Rocco De Marco, Daniel Li Veli, Laura Screpanti, Benedetta Castagna, Alessandro Lucchetti, David Scaradozzi

Förderung

  • National Recovery and Resilience Plan (NRRP), Mission 4 Component 2 Investment 1.4 NextGenerationEU
  • LIFE Financial Instrument of the European Community
  • Life Delfi Project—Dolphin Experience: Lowering Fishing Interactions
  • National Recovery and Resilience Plan (NRRP), Mission 4 Component 2 Investment 1.4 (Call for tender No. 3138 of 16 December 2021, rectified by Decree n.3175 of…

Zitieren

APA

Di Nardo, F., De Marco, R., Li Veli, D., Screpanti, L., Castagna, B., Lucchetti, A., & Scaradozzi, D. (2025). Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations. Sensors. https://doi.org/10.3390/s25082499

Vancouver

Di Nardo F, De Marco R, Li Veli D, Screpanti L, Castagna B, Lucchetti A, et al. Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations. Sensors. 2025. doi:10.3390/s25082499

BibTeX
@article{DiNardo2025,
  author = {Di Nardo, Francesco and De Marco, Rocco and Li Veli, Daniel and Screpanti, Laura and Castagna, Benedetta and Lucchetti, Alessandro and Scaradozzi, David},
  title = {Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations},
  journal = {Sensors},
  year = {2025},
  doi = {10.3390/s25082499}
}
Kostenlos starten →