ANIMAL-SPOT enables animal-independent signal detection and classification using deep learning

Lade...
Vorschaubild
Dateien
Zu diesem Dokument gibt es keine Dateien.
Datum
2022
Autor:innen
Bergler, Christian
Tyndel, Stephen A.
Barnhill, Alexander
Ortiz, Sara T.
Kalan, Ammie K.
Cheng, Rachael Xi
Brinkløv, Signe
Osiecka, Anna N.
Klump, Barbara C.
et al.
Herausgeber:innen
Kontakt
ISSN der Zeitschrift
Electronic ISSN
ISBN
Bibliografische Daten
Verlag
Schriftenreihe
Auflagebezeichnung
ArXiv-ID
Internationale Patentnummer
Link zur Lizenz
Angaben zur Forschungsförderung
Projekt
Open Access-Veröffentlichung
Open Access Gold
Sammlungen
Core Facility der Universität Konstanz
Gesperrt bis
Titel in einer weiteren Sprache
Forschungsvorhaben
Organisationseinheiten
Zeitschriftenheft
Publikationstyp
Zeitschriftenartikel
Publikationsstatus
Published
Erschienen in
Scientific Reports. Springer. 2022, 12, 21966. eISSN 2045-2322. Available under: doi: 10.1038/s41598-022-26429-y
Zusammenfassung

Bioacoustic research spans a wide range of biological questions and applications, relying on identification of target species or smaller acoustic units, such as distinct call types. However, manually identifying the signal of interest is time-intensive, error-prone, and becomes unfeasible with large data volumes. Therefore, machine-driven algorithms are increasingly applied to various bioacoustic signal identification challenges. Nevertheless, biologists still have major difficulties trying to transfer existing animal- and/or scenario-related machine learning approaches to their specific animal datasets and scientific questions. This study presents an animal-independent, open-source deep learning framework, along with a detailed user guide. Three signal identification tasks, commonly encountered in bioacoustics research, were investigated: (1) target signal vs. background noise detection, (2) species classification, and (3) call type categorization. ANIMAL-SPOT successfully segmented human-annotated target signals in data volumes representing 10 distinct animal species and 1 additional genus, resulting in a mean test accuracy of 97.9%, together with an average area under the ROC curve (AUC) of 95.9%, when predicting on unseen recordings. Moreover, an average segmentation accuracy and F1-score of 95.4% was achieved on the publicly available BirdVox-Full-Night data corpus. In addition, multi-class species and call type classification resulted in 96.6% and 92.7% accuracy on unseen test data, as well as 95.2% and 88.4% regarding previous animal-specific machine-based detection excerpts. Furthermore, an Unweighted Average Recall (UAR) of 89.3% outperformed the multi-species classification baseline system of the ComParE 2021 Primate Sub-Challenge. Besides animal independence, ANIMAL-SPOT does not rely on expert knowledge or special computing resources, thereby making deep-learning-based bioacoustic signal identification accessible to a broad audience.

Zusammenfassung in einer weiteren Sprache
Fachgebiet (DDC)
570 Biowissenschaften, Biologie
Schlagwörter
Konferenz
Rezension
undefined / . - undefined, undefined
Zitieren
ISO 690BERGLER, Christian, Simeon Q. SMEELE, Stephen A. TYNDEL, Alexander BARNHILL, Sara T. ORTIZ, Ammie K. KALAN, Rachael Xi CHENG, Signe BRINKLØV, Anna N. OSIECKA, Barbara C. KLUMP, 2022. ANIMAL-SPOT enables animal-independent signal detection and classification using deep learning. In: Scientific Reports. Springer. 2022, 12, 21966. eISSN 2045-2322. Available under: doi: 10.1038/s41598-022-26429-y
BibTex
@article{Bergler2022ANIMA-68971,
  year={2022},
  doi={10.1038/s41598-022-26429-y},
  title={ANIMAL-SPOT enables animal-independent signal detection and classification using deep learning},
  volume={12},
  journal={Scientific Reports},
  author={Bergler, Christian and Smeele, Simeon Q. and Tyndel, Stephen A. and Barnhill, Alexander and Ortiz, Sara T. and Kalan, Ammie K. and Cheng, Rachael Xi and Brinkløv, Signe and Osiecka, Anna N. and Klump, Barbara C.},
  note={Article Number: 21966}
}
RDF
<rdf:RDF
    xmlns:dcterms="http://purl.org/dc/terms/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:bibo="http://purl.org/ontology/bibo/"
    xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:void="http://rdfs.org/ns/void#"
    xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > 
  <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/68971">
    <dc:contributor>Tyndel, Stephen A.</dc:contributor>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/28"/>
    <dc:contributor>Brinkløv, Signe</dc:contributor>
    <dc:language>eng</dc:language>
    <dc:creator>Brinkløv, Signe</dc:creator>
    <dc:creator>Cheng, Rachael Xi</dc:creator>
    <dcterms:title>ANIMAL-SPOT enables animal-independent signal detection and classification using deep learning</dcterms:title>
    <dc:contributor>Bergler, Christian</dc:contributor>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dc:creator>Klump, Barbara C.</dc:creator>
    <dc:contributor>Cheng, Rachael Xi</dc:contributor>
    <dc:creator>Barnhill, Alexander</dc:creator>
    <dc:contributor>Ortiz, Sara T.</dc:contributor>
    <dc:creator>Tyndel, Stephen A.</dc:creator>
    <dc:contributor>Kalan, Ammie K.</dc:contributor>
    <dc:contributor>Osiecka, Anna N.</dc:contributor>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/68971"/>
    <dc:rights>Attribution 4.0 International</dc:rights>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2024-01-09T08:24:02Z</dc:date>
    <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/68971/1/Bergler_2-8evz4j1afnu70.pdf"/>
    <dc:creator>Bergler, Christian</dc:creator>
    <dcterms:issued>2022</dcterms:issued>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2024-01-09T08:24:02Z</dcterms:available>
    <dc:contributor>Klump, Barbara C.</dc:contributor>
    <dc:creator>Osiecka, Anna N.</dc:creator>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/28"/>
    <dc:contributor>Barnhill, Alexander</dc:contributor>
    <dc:creator>Kalan, Ammie K.</dc:creator>
    <dc:creator>Ortiz, Sara T.</dc:creator>
    <dc:creator>Smeele, Simeon Q.</dc:creator>
    <dcterms:abstract>Bioacoustic research spans a wide range of biological questions and applications, relying on identification of target species or smaller acoustic units, such as distinct call types. However, manually identifying the signal of interest is time-intensive, error-prone, and becomes unfeasible with large data volumes. Therefore, machine-driven algorithms are increasingly applied to various bioacoustic signal identification challenges. Nevertheless, biologists still have major difficulties trying to transfer existing animal- and/or scenario-related machine learning approaches to their specific animal datasets and scientific questions. This study presents an animal-independent, open-source deep learning framework, along with a detailed user guide. Three signal identification tasks, commonly encountered in bioacoustics research, were investigated: (1) target signal vs. background noise detection, (2) species classification, and (3) call type categorization. ANIMAL-SPOT successfully segmented human-annotated target signals in data volumes representing 10 distinct animal species and 1 additional genus, resulting in a mean test accuracy of 97.9%, together with an average area under the ROC curve (AUC) of 95.9%, when predicting on unseen recordings. Moreover, an average segmentation accuracy and F1-score of 95.4% was achieved on the publicly available BirdVox-Full-Night data corpus. In addition, multi-class species and call type classification resulted in 96.6% and 92.7% accuracy on unseen test data, as well as 95.2% and 88.4% regarding previous animal-specific machine-based detection excerpts. Furthermore, an Unweighted Average Recall (UAR) of 89.3% outperformed the multi-species classification baseline system of the ComParE 2021 Primate Sub-Challenge. Besides animal independence, ANIMAL-SPOT does not rely on expert knowledge or special computing resources, thereby making deep-learning-based bioacoustic signal identification accessible to a broad audience.</dcterms:abstract>
    <dc:contributor>Smeele, Simeon Q.</dc:contributor>
    <dcterms:rights rdf:resource="http://creativecommons.org/licenses/by/4.0/"/>
    <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/68971/1/Bergler_2-8evz4j1afnu70.pdf"/>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
  </rdf:Description>
</rdf:RDF>
Interner Vermerk
xmlui.Submission.submit.DescribeStep.inputForms.label.kops_note_fromSubmitter
Kontakt
URL der Originalveröffentl.
Prüfdatum der URL
Prüfungsdatum der Dissertation
Finanzierungsart
Kommentar zur Publikation
Allianzlizenz
Corresponding Authors der Uni Konstanz vorhanden
Internationale Co-Autor:innen
Universitätsbibliographie
Begutachtet
Ja
Diese Publikation teilen