Detail publikace

Efficient Feature Set Developed for Acoustic Gunshot Detection in Open Space

SIGMUND, M. HRABINA, M.

Originální název

Efficient Feature Set Developed for Acoustic Gunshot Detection in Open Space

Typ

článek v časopise ve Web of Science, Jimp

Jazyk

angličtina

Originální abstrakt

This paper presents an efficient approach to automatic gunshot detection based on a combination of two feature sets: adapted standard sound features and hand-crafted novel features. The standard features are mel-frequency cepstral coefficients adapted for gunshot recognition in terms of uniform gamma-tone filters linearly spaced over the whole frequency range from 0 kHz to 16 kHz. The novel features were derived in the time domain from individual significant points of the raw waveform after amplitude normalization. Experiments were performed using single and ensemble neural networks to verify the effectiveness of the novel features for supplementing the standard features. In binary classification, the developed approach achieved an accuracy of 95.02 % in gunshot detection.

Klíčová slova

Acoustic signal processing; gunshot detection; neural networks; parameter estimation

Autoři

SIGMUND, M.; HRABINA, M.

Vydáno

23. 8. 2021

Nakladatel

Kaunas University of Technology

Místo

Kaunas

ISSN

1392-1215

Periodikum

Elektronika Ir Elektrotechnika

Ročník

27

Číslo

4

Stát

Litevská republika

Strany od

62

Strany do

68

Strany počet

7

URL

https://eejournal.ktu.lt/index.php/elt/article/view/28877

Plný text v Digitální knihovně

http://hdl.handle.net/11012/203054

BibTex

@article{BUT173150,
  author="Milan {Sigmund} and Martin {Hrabina}",
  title="Efficient Feature Set Developed for Acoustic Gunshot Detection in Open Space",
  journal="Elektronika Ir Elektrotechnika",
  year="2021",
  volume="27",
  number="4",
  pages="62--68",
  doi="10.5755/j02.eie.28877",
  issn="1392-1215",
  url="https://eejournal.ktu.lt/index.php/elt/article/view/28877"
}