Automatic classification of humpback whale social calls
Acoustic methods are an established technique to monitor marine mammal populations and behavior, but developments in computer science can expand the current capabilities. A central aim of these methods is the automated detection and classification of marine mammal vocalizations. While many studies h...
Main Authors: | , , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
eScholarship, University of California
2017
|
Subjects: | |
Online Access: | https://escholarship.org/uc/item/4np9s56k |
id |
ftcdlib:oai:escholarship.org:ark:/13030/qt4np9s56k |
---|---|
record_format |
openpolar |
spelling |
ftcdlib:oai:escholarship.org:ark:/13030/qt4np9s56k 2023-08-20T04:07:05+02:00 Automatic classification of humpback whale social calls Tolkova, Irina Bauer, Lisa Wilby, Antonella Kastner, Ryan Seger, Kerri 3605 - 3605 2017-05-01 application/pdf https://escholarship.org/uc/item/4np9s56k unknown eScholarship, University of California qt4np9s56k https://escholarship.org/uc/item/4np9s56k public The Journal of the Acoustical Society of America, vol 141, iss 5 Life Below Water Acoustics article 2017 ftcdlib 2023-07-31T18:02:02Z Acoustic methods are an established technique to monitor marine mammal populations and behavior, but developments in computer science can expand the current capabilities. A central aim of these methods is the automated detection and classification of marine mammal vocalizations. While many studies have applied bioacoustic methods to cetacean calls, there has been limited success with humpback whale (Megaptera novaeangliae) social call classification, which has largely remained a manual task in the bioacoustics community. In this project, we automated this process by analyzing spectrograms of calls using PCA-based and connected-component-based methods, and derived features from relative power in the frequency bins of these spectrograms. These features were used to train and test a supervised Hidden Markov Model (HMM) algorithm to investigate classification feasibility. We varied the number of features used in this analysis by varying the sizes of frequency bins. Generally, we saw an increase in precision, recall, and accuracy for all three classified groups, across the individual data sets, as the number of features decreased. We will present the classification rates of our algorithm across multiple model parameters. Since this method is not specific to humpback whale vocalizations, we hope it will prove useful in other acoustic applications. Article in Journal/Newspaper Humpback Whale Megaptera novaeangliae University of California: eScholarship |
institution |
Open Polar |
collection |
University of California: eScholarship |
op_collection_id |
ftcdlib |
language |
unknown |
topic |
Life Below Water Acoustics |
spellingShingle |
Life Below Water Acoustics Tolkova, Irina Bauer, Lisa Wilby, Antonella Kastner, Ryan Seger, Kerri Automatic classification of humpback whale social calls |
topic_facet |
Life Below Water Acoustics |
description |
Acoustic methods are an established technique to monitor marine mammal populations and behavior, but developments in computer science can expand the current capabilities. A central aim of these methods is the automated detection and classification of marine mammal vocalizations. While many studies have applied bioacoustic methods to cetacean calls, there has been limited success with humpback whale (Megaptera novaeangliae) social call classification, which has largely remained a manual task in the bioacoustics community. In this project, we automated this process by analyzing spectrograms of calls using PCA-based and connected-component-based methods, and derived features from relative power in the frequency bins of these spectrograms. These features were used to train and test a supervised Hidden Markov Model (HMM) algorithm to investigate classification feasibility. We varied the number of features used in this analysis by varying the sizes of frequency bins. Generally, we saw an increase in precision, recall, and accuracy for all three classified groups, across the individual data sets, as the number of features decreased. We will present the classification rates of our algorithm across multiple model parameters. Since this method is not specific to humpback whale vocalizations, we hope it will prove useful in other acoustic applications. |
format |
Article in Journal/Newspaper |
author |
Tolkova, Irina Bauer, Lisa Wilby, Antonella Kastner, Ryan Seger, Kerri |
author_facet |
Tolkova, Irina Bauer, Lisa Wilby, Antonella Kastner, Ryan Seger, Kerri |
author_sort |
Tolkova, Irina |
title |
Automatic classification of humpback whale social calls |
title_short |
Automatic classification of humpback whale social calls |
title_full |
Automatic classification of humpback whale social calls |
title_fullStr |
Automatic classification of humpback whale social calls |
title_full_unstemmed |
Automatic classification of humpback whale social calls |
title_sort |
automatic classification of humpback whale social calls |
publisher |
eScholarship, University of California |
publishDate |
2017 |
url |
https://escholarship.org/uc/item/4np9s56k |
op_coverage |
3605 - 3605 |
genre |
Humpback Whale Megaptera novaeangliae |
genre_facet |
Humpback Whale Megaptera novaeangliae |
op_source |
The Journal of the Acoustical Society of America, vol 141, iss 5 |
op_relation |
qt4np9s56k https://escholarship.org/uc/item/4np9s56k |
op_rights |
public |
_version_ |
1774718526714544128 |