Robust real-time detection of right whale upcalls using neural networks on the edge

Animals worldwide are facing ecological pressures from global climate change and increasing anthropogenic activities. To transition to a renewable energy future, extensive offshore wind development is planned globally. In the North Atlantic, future development sites overlap with the migratory range...

Full description

Saved in:
Bibliographic Details
Main Authors: Matthew D. Hyer, Austin T. Anderson, David A. Mann, T. Aran Mooney, Nadège Aoki, Frants H. Jensen
Format: Article
Language:English
Published: Elsevier 2025-11-01
Series:Ecological Informatics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1574954125001396
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849694901961228288
author Matthew D. Hyer
Austin T. Anderson
David A. Mann
T. Aran Mooney
Nadège Aoki
Frants H. Jensen
author_facet Matthew D. Hyer
Austin T. Anderson
David A. Mann
T. Aran Mooney
Nadège Aoki
Frants H. Jensen
author_sort Matthew D. Hyer
collection DOAJ
description Animals worldwide are facing ecological pressures from global climate change and increasing anthropogenic activities. To transition to a renewable energy future, extensive offshore wind development is planned globally. In the North Atlantic, future development sites overlap with the migratory range of critically endangered North Atlantic right whales (NARW) and will lead to increased risk of ship strikes, pile driving impacts, and other population risks. New methods to accurately detect cetaceans and provide real-time feedback for mitigation will be increasingly important to enact sustainable management actions to facilitate the recovery of the NARW. Recent developments in acoustic event detection made possible by deep learning have shown significantly improved detection performance across many different taxa, but such models tend to be too computationally expensive to run on existing wildlife monitoring platforms. Here, we use model compression techniques combined with an autonomous acoustic recording platform integrating an ESP32 microcontroller to bring real-time detection with deep learning to the edge. We test if edge-based inference using a compressed network running on a microprocessor entails significant performance loss and find that this loss is negligible. We leverage large, open-source datasets of noise from the NOAA SanctSound project for generating semi-synthetic training datasets that encourage model generalization to novel noise conditions. Our compressed model achieves improved performance across all tested recording sites in the Western North Atlantic Ocean, demonstrating that deep learning powered wildlife monitoring solutions can provide reliable real-time data for mitigation of human impacts and help ensure a sustainable green energy transition.
format Article
id doaj-art-0dd44bfb36fc49e8aeb16a3d18458c29
institution DOAJ
issn 1574-9541
language English
publishDate 2025-11-01
publisher Elsevier
record_format Article
series Ecological Informatics
spelling doaj-art-0dd44bfb36fc49e8aeb16a3d18458c292025-08-20T03:19:56ZengElsevierEcological Informatics1574-95412025-11-018910313010.1016/j.ecoinf.2025.103130Robust real-time detection of right whale upcalls using neural networks on the edgeMatthew D. Hyer0Austin T. Anderson1David A. Mann2T. Aran Mooney3Nadège Aoki4Frants H. Jensen5Department of Ecoscience, Aarhus University, Frederiksborgvej 399, 4000 Roskilde, Denmark; Corresponding author.Loggerhead Instruments, 6576 Palmer Park Circle, Sarasota, FL, 34238, USALoggerhead Instruments, 6576 Palmer Park Circle, Sarasota, FL, 34238, USABiology Department, Woods Hole Oceanographic Institution, 266 Woods Hole Road, Woods Hole, MA, 02543, USABiology Department, Woods Hole Oceanographic Institution, 266 Woods Hole Road, Woods Hole, MA, 02543, USA; Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, 77 Massachusetts Avenue, 55-101, Cambridge, MA, 02139, USADepartment of Ecoscience, Aarhus University, Frederiksborgvej 399, 4000 Roskilde, Denmark; Biology Department, Woods Hole Oceanographic Institution, 266 Woods Hole Road, Woods Hole, MA, 02543, USA; Department of Biology, Syracuse University, 107 College Pl, Syracuse, NY, 13244, USAAnimals worldwide are facing ecological pressures from global climate change and increasing anthropogenic activities. To transition to a renewable energy future, extensive offshore wind development is planned globally. In the North Atlantic, future development sites overlap with the migratory range of critically endangered North Atlantic right whales (NARW) and will lead to increased risk of ship strikes, pile driving impacts, and other population risks. New methods to accurately detect cetaceans and provide real-time feedback for mitigation will be increasingly important to enact sustainable management actions to facilitate the recovery of the NARW. Recent developments in acoustic event detection made possible by deep learning have shown significantly improved detection performance across many different taxa, but such models tend to be too computationally expensive to run on existing wildlife monitoring platforms. Here, we use model compression techniques combined with an autonomous acoustic recording platform integrating an ESP32 microcontroller to bring real-time detection with deep learning to the edge. We test if edge-based inference using a compressed network running on a microprocessor entails significant performance loss and find that this loss is negligible. We leverage large, open-source datasets of noise from the NOAA SanctSound project for generating semi-synthetic training datasets that encourage model generalization to novel noise conditions. Our compressed model achieves improved performance across all tested recording sites in the Western North Atlantic Ocean, demonstrating that deep learning powered wildlife monitoring solutions can provide reliable real-time data for mitigation of human impacts and help ensure a sustainable green energy transition.http://www.sciencedirect.com/science/article/pii/S1574954125001396Data augmentationDeep learningNorth Atlantic right whaleReal-time acoustic detectionSustainable marine developmentWildlife monitoring
spellingShingle Matthew D. Hyer
Austin T. Anderson
David A. Mann
T. Aran Mooney
Nadège Aoki
Frants H. Jensen
Robust real-time detection of right whale upcalls using neural networks on the edge
Ecological Informatics
Data augmentation
Deep learning
North Atlantic right whale
Real-time acoustic detection
Sustainable marine development
Wildlife monitoring
title Robust real-time detection of right whale upcalls using neural networks on the edge
title_full Robust real-time detection of right whale upcalls using neural networks on the edge
title_fullStr Robust real-time detection of right whale upcalls using neural networks on the edge
title_full_unstemmed Robust real-time detection of right whale upcalls using neural networks on the edge
title_short Robust real-time detection of right whale upcalls using neural networks on the edge
title_sort robust real time detection of right whale upcalls using neural networks on the edge
topic Data augmentation
Deep learning
North Atlantic right whale
Real-time acoustic detection
Sustainable marine development
Wildlife monitoring
url http://www.sciencedirect.com/science/article/pii/S1574954125001396
work_keys_str_mv AT matthewdhyer robustrealtimedetectionofrightwhaleupcallsusingneuralnetworksontheedge
AT austintanderson robustrealtimedetectionofrightwhaleupcallsusingneuralnetworksontheedge
AT davidamann robustrealtimedetectionofrightwhaleupcallsusingneuralnetworksontheedge
AT taranmooney robustrealtimedetectionofrightwhaleupcallsusingneuralnetworksontheedge
AT nadegeaoki robustrealtimedetectionofrightwhaleupcallsusingneuralnetworksontheedge
AT frantshjensen robustrealtimedetectionofrightwhaleupcallsusingneuralnetworksontheedge