5G AI-IoT System for Bird Species Monitoring and Song Classification
Sensors · 2024
DOI: 10.3390/s24113687 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2024. Verfasst von Jaume Segura-Garcia und 5 weiteren Autor:innen. Beteiligte Einrichtung: Escola Tecnica Superior d'Enginyeria, Universitat de Valencia, 46100 Burjassot, Spain. Hauptthemen laut MEDLINE: Birds, Vocalization, Animal, Neural Networks, Computer. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministerio de Universidades de España, Ministerio de Ciencia e Innovación/Agencia Española de Investigación (MCIN/AEI), Spanish Ministry of Universities.
Abstract (Originalsprache)
Identification of different species of animals has become an important issue in biology and ecology. Ornithology has made alliances with other disciplines in order to establish a set of methods that play an important role in the birds’ protection and the evaluation of the environmental quality of different ecosystems. In this case, the use of machine learning and deep learning techniques has produced big progress in birdsong identification. To make an approach from AI-IoT, we have used different approaches based on image feature comparison (through CNNs trained with Imagenet weights, such as EfficientNet or MobileNet) using the feature spectrogram for the birdsong, but also the use of the deep CNN (DCNN) has shown good performance for birdsong classification for reduction of the model size. A 5G IoT-based system for raw audio gathering has been developed, and different CNNs have been tested for bird identification from audio recordings. This comparison shows that Imagenet-weighted CNN shows a relatively high performance for most species, achieving 75% accuracy. However, this network contains a large number of parameters, leading to a less energy efficient inference. We have designed two DCNNs to reduce the amount of parameters, to keep the accuracy at a certain level, and to allow their integration into a small board computer (SBC) or a microcontroller unit (MCU).
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Escola Tecnica Superior d'Enginyeria, Universitat de Valencia, 46100 Burjassot, Spain.
- School of Computing, Engineering & Physical Sciences, University of the West of Scotland, Paisley PA1 2BE, Scotland, UK.
ORCID-Profile: Jaume Segura-Garcia, Sean Sturley, Miguel Arevalillo-Herraez, Jose M Alcaraz-Calero, Santiago Felici-Castell, Enrique A Navarro-Camba
Förderung
- Ministerio de Universidades de España
- Ministerio de Ciencia e Innovación/Agencia Española de Investigación (MCIN/AEI)
- Spanish Ministry of Universities
- Spanish Ministry of Science and Innovation/Spanish Research Agency (MCIN/AEI)
- MCIN/AEI/ 10.13039/501100011033
- European Union “NextGenerationEU”/PRTR
Zitieren
APA
Segura-Garcia, J., Sturley, S., Arevalillo-Herraez, M., Alcaraz-Calero, J. M., Felici-Castell, S., & Navarro-Camba, E. A. (2024). 5G AI-IoT System for Bird Species Monitoring and Song Classification. Sensors. https://doi.org/10.3390/s24113687
Vancouver
Segura-Garcia J, Sturley S, Arevalillo-Herraez M, Alcaraz-Calero JM, Felici-Castell S, Navarro-Camba EA. 5G AI-IoT System for Bird Species Monitoring and Song Classification. Sensors. 2024. doi:10.3390/s24113687
BibTeX
@article{SeguraGarcia2024,
author = {Segura-Garcia, Jaume and Sturley, Sean and Arevalillo-Herraez, Miguel and Alcaraz-Calero, Jose M and Felici-Castell, Santiago and Navarro-Camba, Enrique A},
title = {5G AI-IoT System for Bird Species Monitoring and Song Classification},
journal = {Sensors},
year = {2024},
doi = {10.3390/s24113687}
}
EUVIMED