Species-specific audio detection: a comparison of three template-based detection algorithms using random forests
We developed a web-based cloud-hosted system that allow users to archive, listen, visualize, and annotate recordings. The system also provides tools to convert these annotations into datasets that can be used to train a computer to detect the presence or absence of a species. The algorithm used by t...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
PeerJ Inc.
2017-04-01
|
Series: | PeerJ Computer Science |
Subjects: | |
Online Access: | https://peerj.com/articles/cs-113.pdf |
id |
doaj-87fdf05fe5ac42b69986fa961e12aff3 |
---|---|
record_format |
Article |
spelling |
doaj-87fdf05fe5ac42b69986fa961e12aff32020-11-24T22:20:21ZengPeerJ Inc.PeerJ Computer Science2376-59922017-04-013e11310.7717/peerj-cs.113Species-specific audio detection: a comparison of three template-based detection algorithms using random forestsCarlos J. Corrada Bravo0Rafael Álvarez Berríos1T. Mitchell Aide2Department of Computer Science, University of Puerto Rico—Rio Piedras, San Juan, Puerto RicoSieve Analytics, Inc., San Juan, Puerto RicoSieve Analytics, Inc., San Juan, Puerto RicoWe developed a web-based cloud-hosted system that allow users to archive, listen, visualize, and annotate recordings. The system also provides tools to convert these annotations into datasets that can be used to train a computer to detect the presence or absence of a species. The algorithm used by the system was selected after comparing the accuracy and efficiency of three variants of a template-based detection. The algorithm computes a similarity vector by comparing a template of a species call with time increments across the spectrogram. Statistical features are extracted from this vector and used as input for a Random Forest classifier that predicts presence or absence of the species in the recording. The fastest algorithm variant had the highest average accuracy and specificity; therefore, it was implemented in the ARBIMON web-based system.https://peerj.com/articles/cs-113.pdfAcoustic monitoringMachine learningAnimal vocalizationsRecording visualizationRecording annotationGeneric species algorithm |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Carlos J. Corrada Bravo Rafael Álvarez Berríos T. Mitchell Aide |
spellingShingle |
Carlos J. Corrada Bravo Rafael Álvarez Berríos T. Mitchell Aide Species-specific audio detection: a comparison of three template-based detection algorithms using random forests PeerJ Computer Science Acoustic monitoring Machine learning Animal vocalizations Recording visualization Recording annotation Generic species algorithm |
author_facet |
Carlos J. Corrada Bravo Rafael Álvarez Berríos T. Mitchell Aide |
author_sort |
Carlos J. Corrada Bravo |
title |
Species-specific audio detection: a comparison of three template-based detection algorithms using random forests |
title_short |
Species-specific audio detection: a comparison of three template-based detection algorithms using random forests |
title_full |
Species-specific audio detection: a comparison of three template-based detection algorithms using random forests |
title_fullStr |
Species-specific audio detection: a comparison of three template-based detection algorithms using random forests |
title_full_unstemmed |
Species-specific audio detection: a comparison of three template-based detection algorithms using random forests |
title_sort |
species-specific audio detection: a comparison of three template-based detection algorithms using random forests |
publisher |
PeerJ Inc. |
series |
PeerJ Computer Science |
issn |
2376-5992 |
publishDate |
2017-04-01 |
description |
We developed a web-based cloud-hosted system that allow users to archive, listen, visualize, and annotate recordings. The system also provides tools to convert these annotations into datasets that can be used to train a computer to detect the presence or absence of a species. The algorithm used by the system was selected after comparing the accuracy and efficiency of three variants of a template-based detection. The algorithm computes a similarity vector by comparing a template of a species call with time increments across the spectrogram. Statistical features are extracted from this vector and used as input for a Random Forest classifier that predicts presence or absence of the species in the recording. The fastest algorithm variant had the highest average accuracy and specificity; therefore, it was implemented in the ARBIMON web-based system. |
topic |
Acoustic monitoring Machine learning Animal vocalizations Recording visualization Recording annotation Generic species algorithm |
url |
https://peerj.com/articles/cs-113.pdf |
work_keys_str_mv |
AT carlosjcorradabravo speciesspecificaudiodetectionacomparisonofthreetemplatebaseddetectionalgorithmsusingrandomforests AT rafaelalvarezberrios speciesspecificaudiodetectionacomparisonofthreetemplatebaseddetectionalgorithmsusingrandomforests AT tmitchellaide speciesspecificaudiodetectionacomparisonofthreetemplatebaseddetectionalgorithmsusingrandomforests |
_version_ |
1725775590437421056 |