Characterising and detecting fin whale calls using deep learning at the Lofoten-Vesterålen Observatory, Norway
| Autor: | Alan J. Hunter, Gary Heald, Shaula Garibbo, Duncan P. Williams, Philippe Blondel, Ross Heyburn |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Soundscape
Acoustics and Ultrasonics Ambient noise level ambient noise Observatory Artificial Intelligence biology.animal SDG 14 - Life Below Water automatic signal identification Sound pressure acoustics biology Whale business.industry Deep learning deep learning Identification (information) underwater acoustics Artificial intelligence Underwater acoustics business Cartography Geology |
| Zdroj: | Garibbo, S, Blondel, P, Heald, G, Heyburn, R, Hunter, A J & Williams, D 2021, ' Characterising and detecting fin whale calls using deep learning at the Lofoten-Vesterålen Observatory, Norway ', Proceedings of Meetings on Acoustics, vol. 44, no. 1, 070021 . https://doi.org/10.1121/2.0001488, https://doi.org/10.1121/2.0001488 |
| DOI: | 10.1121/2.0001488 |
| Popis: | The application of deep learning to solving acoustic detection and identification challenges is a rapidly-evolving subfield of underwater acoustics. Automatic signal identification can be used for many applications, like enabling thecompilation of large datasets from many sources, which can be used to better constrain source-specific characteristics and trends. Earlier analyses (Garibbo et al., 2020) identified the different contributions of wind, weather, shippingand earthquakes. The long-term acoustic measurements regularly include calls from fin whales, whose presence and vocal activities in the area vary with seasons; their 20-Hz calls are sometimes mixed with other signals, likeearthquakes or shipping. We present here the application of deep learning to automatically identify these whale calls. Percentile analyses of the temporal variation of the frequency of calls, their Power Spectral Density (PSD), and SoundPressure Level (SPL) is carried out to determine their respective contributions to the overall soundscape and highlight relevant information about these whale populations. The deep learning approaches selected here can also beused for other types of animal vocalisations and for other short-term processes (e.g. passing ships, earthquakes of different types), assisting in their identification and in the statistical and temporal analyses of low-frequencysoundscapes. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|