Robust North Atlantic right whale detection using deep learning models for denoising

Vickers, William, Milner, Ben, Risch, Denise and Lee, Robert (2021) Robust North Atlantic right whale detection using deep learning models for denoising. Journal of the Acoustical Society of America, 149. ISSN 0001-4966

[img] PDF (JASA___ML_in_Acoustics_2021 (1)) - Accepted Version
Restricted to Repository staff only until 3 December 2021.

Download (8MB) | Request a copy


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 is applied to uncompensated classifiers, their 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 (DNCNN) and a denoising autoencoder (DAE), 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 SNRs from -10dB to +5dB to clean recordings to simulate noisy conditions. Experiments show that denoising gives substantial improvements to accuracy and 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.

Item Type: Article
Faculty \ School: Faculty of Science > School of Computing Sciences
Depositing User: LivePure Connector
Date Deposited: 20 May 2021 00:08
Last Modified: 05 Jun 2021 00:09

Actions (login required)

View Item View Item