Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations

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 clas...

Full description

Saved in:
Bibliographic Details
Main Authors: Francesco Di Nardo, Rocco De Marco, Daniel Li Veli, Laura Screpanti, Benedetta Castagna, Alessandro Lucchetti, David Scaradozzi
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/8/2499
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849713495676813312
author Francesco Di Nardo
Rocco De Marco
Daniel Li Veli
Laura Screpanti
Benedetta Castagna
Alessandro Lucchetti
David Scaradozzi
author_facet Francesco Di Nardo
Rocco De Marco
Daniel Li Veli
Laura Screpanti
Benedetta Castagna
Alessandro Lucchetti
David Scaradozzi
author_sort Francesco Di Nardo
collection DOAJ
description 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 (<i>Tursiops truncatus</i>) 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.
format Article
id doaj-art-e32b03881315406eb4bd23cf7d6f0178
institution DOAJ
issn 1424-8220
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-e32b03881315406eb4bd23cf7d6f01782025-08-20T03:13:57ZengMDPI AGSensors1424-82202025-04-01258249910.3390/s25082499Multiclass CNN Approach for Automatic Classification of Dolphin VocalizationsFrancesco Di Nardo0Rocco De Marco1Daniel Li Veli2Laura Screpanti3Benedetta Castagna4Alessandro Lucchetti5David Scaradozzi6Dipartimento di Ingegneria dell’informazione, Università Politecnica delle Marche, 60131 Ancona, ItalyInstitute of Biological Resources and Marine Biotechnology (IRBIM), National Research Council (CNR), 60125 Ancona, ItalyInstitute of Biological Resources and Marine Biotechnology (IRBIM), National Research Council (CNR), 60125 Ancona, ItalyDipartimento di Ingegneria dell’informazione, Università Politecnica delle Marche, 60131 Ancona, ItalyDipartimento di Ingegneria dell’informazione, Università Politecnica delle Marche, 60131 Ancona, ItalyInstitute of Biological Resources and Marine Biotechnology (IRBIM), National Research Council (CNR), 60125 Ancona, ItalyDipartimento di Ingegneria dell’informazione, Università Politecnica delle Marche, 60131 Ancona, ItalyMonitoring 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 (<i>Tursiops truncatus</i>) 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.https://www.mdpi.com/1424-8220/25/8/2499convolutional neural networksdeep learningdolphinspassive acoustic monitoring
spellingShingle Francesco Di Nardo
Rocco De Marco
Daniel Li Veli
Laura Screpanti
Benedetta Castagna
Alessandro Lucchetti
David Scaradozzi
Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations
Sensors
convolutional neural networks
deep learning
dolphins
passive acoustic monitoring
title Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations
title_full Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations
title_fullStr Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations
title_full_unstemmed Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations
title_short Multiclass CNN Approach for Automatic Classification of Dolphin Vocalizations
title_sort multiclass cnn approach for automatic classification of dolphin vocalizations
topic convolutional neural networks
deep learning
dolphins
passive acoustic monitoring
url https://www.mdpi.com/1424-8220/25/8/2499
work_keys_str_mv AT francescodinardo multiclasscnnapproachforautomaticclassificationofdolphinvocalizations
AT roccodemarco multiclasscnnapproachforautomaticclassificationofdolphinvocalizations
AT danielliveli multiclasscnnapproachforautomaticclassificationofdolphinvocalizations
AT laurascrepanti multiclasscnnapproachforautomaticclassificationofdolphinvocalizations
AT benedettacastagna multiclasscnnapproachforautomaticclassificationofdolphinvocalizations
AT alessandrolucchetti multiclasscnnapproachforautomaticclassificationofdolphinvocalizations
AT davidscaradozzi multiclasscnnapproachforautomaticclassificationofdolphinvocalizations