Robust North Atlantic right whale detection using deep learning models for denoising
This paper proposes a robust system for detecting North Atlantic right whales by using deep learning methods to denoise noisy recordings. Passive acoustic recordings of right whale vocalisations are subject to noise contamination from many sources, such as shipping and offshore activities. When such...
Published in: | The Journal of the Acoustical Society of America |
---|---|
Main Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2021
|
Subjects: | |
Online Access: | https://pure.uhi.ac.uk/en/publications/d9888db2-7312-4ffb-b7bd-9cd6d802c28a https://doi.org/10.1121/10.0005128 https://pureadmin.uhi.ac.uk/ws/files/16041078/JASA_ML_in_Acoustics_2021_R1_002_.pdf https://asa.scitation.org/doi/10.1121/10.0005128 |
id |
ftuhipublicatio:oai:pure.atira.dk:publications/d9888db2-7312-4ffb-b7bd-9cd6d802c28a |
---|---|
record_format |
openpolar |
spelling |
ftuhipublicatio:oai:pure.atira.dk:publications/d9888db2-7312-4ffb-b7bd-9cd6d802c28a 2024-06-23T07:55:00+00:00 Robust North Atlantic right whale detection using deep learning models for denoising Vickers, William Milner, Ben Risch, Denise Lee, Robert 2021-06-03 application/pdf https://pure.uhi.ac.uk/en/publications/d9888db2-7312-4ffb-b7bd-9cd6d802c28a https://doi.org/10.1121/10.0005128 https://pureadmin.uhi.ac.uk/ws/files/16041078/JASA_ML_in_Acoustics_2021_R1_002_.pdf https://asa.scitation.org/doi/10.1121/10.0005128 eng eng https://pure.uhi.ac.uk/en/publications/d9888db2-7312-4ffb-b7bd-9cd6d802c28a info:eu-repo/semantics/openAccess Vickers , W , Milner , B , Risch , D & Lee , R 2021 , ' Robust North Atlantic right whale detection using deep learning models for denoising ' , The Journal of the Acoustical Society of America , vol. 149 , no. 6 , pp. 3797-3812 . https://doi.org/10.1121/10.0005128 article 2021 ftuhipublicatio https://doi.org/10.1121/10.0005128 2024-05-27T23:56:27Z This paper proposes a robust system for detecting North Atlantic right whales by using deep learning methods to denoise noisy recordings. Passive acoustic recordings of right whale vocalisations are subject to noise contamination from many sources, such as shipping and offshore activities. When such data are applied to uncompensated classifiers, accuracy falls substantially. To build robustness into the detection process, two separate approaches that have proved successful for image denoising are considered. Specifically, a denoising convolutional neural network and a denoising autoencoder, each of which is applied to spectrogram representations of the noisy audio signal, are developed. Performance is improved further by matching the classifier training to include the vestigial signal that remains in clean estimates after the denoising process. Evaluations are performed first by adding white, tanker, trawler, and shot noises at signal-to-noise ratios from −10 to +5 dB to clean recordings to simulate noisy conditions. Experiments show that denoising gives substantial improvements to accuracy, particularly when using the vestigial-trained classifier. A final test applies the proposed methods to previously unseen noisy right whale recordings and finds that denoising is able to improve performance over the baseline clean-trained model in this new noise environment Article in Journal/Newspaper North Atlantic North Atlantic right whale University of the Highlands and Islands: Research Database of UHI The Journal of the Acoustical Society of America 149 6 3797 3812 |
institution |
Open Polar |
collection |
University of the Highlands and Islands: Research Database of UHI |
op_collection_id |
ftuhipublicatio |
language |
English |
description |
This paper proposes a robust system for detecting North Atlantic right whales by using deep learning methods to denoise noisy recordings. Passive acoustic recordings of right whale vocalisations are subject to noise contamination from many sources, such as shipping and offshore activities. When such data are applied to uncompensated classifiers, accuracy falls substantially. To build robustness into the detection process, two separate approaches that have proved successful for image denoising are considered. Specifically, a denoising convolutional neural network and a denoising autoencoder, each of which is applied to spectrogram representations of the noisy audio signal, are developed. Performance is improved further by matching the classifier training to include the vestigial signal that remains in clean estimates after the denoising process. Evaluations are performed first by adding white, tanker, trawler, and shot noises at signal-to-noise ratios from −10 to +5 dB to clean recordings to simulate noisy conditions. Experiments show that denoising gives substantial improvements to accuracy, particularly when using the vestigial-trained classifier. A final test applies the proposed methods to previously unseen noisy right whale recordings and finds that denoising is able to improve performance over the baseline clean-trained model in this new noise environment |
format |
Article in Journal/Newspaper |
author |
Vickers, William Milner, Ben Risch, Denise Lee, Robert |
spellingShingle |
Vickers, William Milner, Ben Risch, Denise Lee, Robert Robust North Atlantic right whale detection using deep learning models for denoising |
author_facet |
Vickers, William Milner, Ben Risch, Denise Lee, Robert |
author_sort |
Vickers, William |
title |
Robust North Atlantic right whale detection using deep learning models for denoising |
title_short |
Robust North Atlantic right whale detection using deep learning models for denoising |
title_full |
Robust North Atlantic right whale detection using deep learning models for denoising |
title_fullStr |
Robust North Atlantic right whale detection using deep learning models for denoising |
title_full_unstemmed |
Robust North Atlantic right whale detection using deep learning models for denoising |
title_sort |
robust north atlantic right whale detection using deep learning models for denoising |
publishDate |
2021 |
url |
https://pure.uhi.ac.uk/en/publications/d9888db2-7312-4ffb-b7bd-9cd6d802c28a https://doi.org/10.1121/10.0005128 https://pureadmin.uhi.ac.uk/ws/files/16041078/JASA_ML_in_Acoustics_2021_R1_002_.pdf https://asa.scitation.org/doi/10.1121/10.0005128 |
genre |
North Atlantic North Atlantic right whale |
genre_facet |
North Atlantic North Atlantic right whale |
op_source |
Vickers , W , Milner , B , Risch , D & Lee , R 2021 , ' Robust North Atlantic right whale detection using deep learning models for denoising ' , The Journal of the Acoustical Society of America , vol. 149 , no. 6 , pp. 3797-3812 . https://doi.org/10.1121/10.0005128 |
op_relation |
https://pure.uhi.ac.uk/en/publications/d9888db2-7312-4ffb-b7bd-9cd6d802c28a |
op_rights |
info:eu-repo/semantics/openAccess |
op_doi |
https://doi.org/10.1121/10.0005128 |
container_title |
The Journal of the Acoustical Society of America |
container_volume |
149 |
container_issue |
6 |
container_start_page |
3797 |
op_container_end_page |
3812 |
_version_ |
1802647384268734464 |