EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

InsectSound1000 An insect sound dataset for deep learning based acoustic insect recognition

Jelto Branding iD, Dieter von Hörsten, Elias Böckmann, Jens Karl Wegener, Eberhard Hartung

Scientific Data · 2024

DOI: 10.1038/s41597-024-03301-4 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Scientific Data, 2024. Verfasst von Jelto Branding und 4 weiteren Autor:innen. Beteiligte Einrichtung: Julius Kühn Institute (JKI), Institute for Application Techniques in Plant Protection, Braunschweig, 38104, Germany. jelto.branding@julius-kuehn.de. Hauptthemen laut MEDLINE: Deep Learning, Acoustics, Sound. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Federal Ministry of Food and Agriculture | Bundesanstalt für Landwirtschaft und Ernährung (Federal Office of Agriculture and Food).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

InsectSound1000 is a dataset comprising more than 169000 labelled sound samples of 12 insects. The insect sound level spans from very loud ( Bombus terrestris ) to inaudible to human ears ( Aphidoletes aphidimyza ). The samples were extracted from more than 1000 h of recordings made in an anechoic box with a four-channel low-noise measurement microphone array. Each sample is a four-channel wave-file of 2500 kHz length, at 16 kHz sample rate and 32 bit resolution. Acoustic insect recognition holds great potential to form the basis of a digital insect sensor. Such sensors are desperately needed to automate pest monitoring and ecological monitoring. With its significant size and high-quality recordings, InsectSound1000 can be used to train data-hungry deep learning models. Used to pretrain models, it can also be leveraged to enable the development of acoustic insect recognition systems on different hardware or for different insects. Further, the methodology employed to create the dataset is presented in detail to allow for the extension of the published dataset.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningAcousticsSoundAnimalsInsecta

Autor:innen und Institutionen

  • Julius Kühn Institute (JKI), Institute for Application Techniques in Plant Protection, Braunschweig, 38104, Germany. jelto.branding@julius-kuehn.de.
  • Julius Kühn Institute (JKI), Institute for Application Techniques in Plant Protection, Braunschweig, 38104, Germany.
  • Julius Kühn Institute (JKI), Institute for Plant Protection in Horticulture and Urban Green, Braunschweig, 38104, Germany.
  • Christian-Albrechts-Universität zu Kiel, Institute of Agricultural Process Engineering, Kiel, 24118, Germany.

ORCID-Profile: Jelto Branding

Förderung

  • Federal Ministry of Food and Agriculture | Bundesanstalt für Landwirtschaft und Ernährung (Federal Office of Agriculture and Food)

Zitieren

APA

Branding, J., von Hörsten, D., Böckmann, E., Wegener, J. K., & Hartung, E. (2024). InsectSound1000 An insect sound dataset for deep learning based acoustic insect recognition. Scientific Data. https://doi.org/10.1038/s41597-024-03301-4

Vancouver

Branding J, von Hörsten D, Böckmann E, Wegener JK, Hartung E. InsectSound1000 An insect sound dataset for deep learning based acoustic insect recognition. Scientific Data. 2024. doi:10.1038/s41597-024-03301-4

BibTeX
@article{Branding2024,
  author = {Branding, Jelto and von Hörsten, Dieter and Böckmann, Elias and Wegener, Jens Karl and Hartung, Eberhard},
  title = {InsectSound1000 An insect sound dataset for deep learning based acoustic insect recognition},
  journal = {Scientific Data},
  year = {2024},
  doi = {10.1038/s41597-024-03301-4}
}
Kostenlos starten →