Discrimination of individual tigers (Panthera tigris) from long distance roars

An Ji, Michael T. Johnson, Edward J. Walsh, Joann McGee, Douglas L. Armstrong

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

This paper investigates the extent of tiger (Panthera tigris) vocal individuality through both qualitative and quantitative approaches using long distance roars from six individual tigers at Omaha's Henry Doorly Zoo in Omaha, NE. The framework for comparison across individuals includes statistical and discriminant function analysis across whole vocalization measures and statistical pattern classification using a hidden Markov model (HMM) with frame-based spectral features comprised of Greenwood frequency cepstral coefficients. Individual discrimination accuracy is evaluated as a function of spectral model complexity, represented by the number of mixtures in the underlying Gaussian mixture model (GMM), and temporal model complexity, represented by the number of sequential states in the HMM. Results indicate that the temporal pattern of the vocalization is the most significant factor in accurate discrimination. Overall baseline discrimination accuracy for this data set is about 70% using high level features without complex spectral or temporal models. Accuracy increases to about 80% when more complex spectral models (multiple mixture GMMs) are incorporated, and increases to a final accuracy of 90% when more detailed temporal models (10-state HMMs) are used. Classification accuracy is stable across a relatively wide range of configurations in terms of spectral and temporal model resolution.

Original languageEnglish (US)
Pages (from-to)1762-1769
Number of pages8
JournalJournal of the Acoustical Society of America
Volume133
Issue number3
DOIs
StatePublished - Mar 1 2013

Fingerprint

discrimination
Discrimination
Spectrality
Hidden Markov Model
coefficients
configurations
Vocalization

ASJC Scopus subject areas

  • Arts and Humanities (miscellaneous)
  • Acoustics and Ultrasonics

Cite this

Discrimination of individual tigers (Panthera tigris) from long distance roars. / Ji, An; Johnson, Michael T.; Walsh, Edward J.; McGee, Joann; Armstrong, Douglas L.

In: Journal of the Acoustical Society of America, Vol. 133, No. 3, 01.03.2013, p. 1762-1769.

Research output: Contribution to journalArticle

Ji, An ; Johnson, Michael T. ; Walsh, Edward J. ; McGee, Joann ; Armstrong, Douglas L. / Discrimination of individual tigers (Panthera tigris) from long distance roars. In: Journal of the Acoustical Society of America. 2013 ; Vol. 133, No. 3. pp. 1762-1769.
@article{38df04eaf1474376a67c7dd82ac8ecd9,
title = "Discrimination of individual tigers (Panthera tigris) from long distance roars",
abstract = "This paper investigates the extent of tiger (Panthera tigris) vocal individuality through both qualitative and quantitative approaches using long distance roars from six individual tigers at Omaha's Henry Doorly Zoo in Omaha, NE. The framework for comparison across individuals includes statistical and discriminant function analysis across whole vocalization measures and statistical pattern classification using a hidden Markov model (HMM) with frame-based spectral features comprised of Greenwood frequency cepstral coefficients. Individual discrimination accuracy is evaluated as a function of spectral model complexity, represented by the number of mixtures in the underlying Gaussian mixture model (GMM), and temporal model complexity, represented by the number of sequential states in the HMM. Results indicate that the temporal pattern of the vocalization is the most significant factor in accurate discrimination. Overall baseline discrimination accuracy for this data set is about 70{\%} using high level features without complex spectral or temporal models. Accuracy increases to about 80{\%} when more complex spectral models (multiple mixture GMMs) are incorporated, and increases to a final accuracy of 90{\%} when more detailed temporal models (10-state HMMs) are used. Classification accuracy is stable across a relatively wide range of configurations in terms of spectral and temporal model resolution.",
author = "An Ji and Johnson, {Michael T.} and Walsh, {Edward J.} and Joann McGee and Armstrong, {Douglas L.}",
year = "2013",
month = "3",
day = "1",
doi = "10.1121/1.4789936",
language = "English (US)",
volume = "133",
pages = "1762--1769",
journal = "Journal of the Acoustical Society of America",
issn = "0001-4966",
publisher = "Acoustical Society of America",
number = "3",

}

TY - JOUR

T1 - Discrimination of individual tigers (Panthera tigris) from long distance roars

AU - Ji, An

AU - Johnson, Michael T.

AU - Walsh, Edward J.

AU - McGee, Joann

AU - Armstrong, Douglas L.

PY - 2013/3/1

Y1 - 2013/3/1

N2 - This paper investigates the extent of tiger (Panthera tigris) vocal individuality through both qualitative and quantitative approaches using long distance roars from six individual tigers at Omaha's Henry Doorly Zoo in Omaha, NE. The framework for comparison across individuals includes statistical and discriminant function analysis across whole vocalization measures and statistical pattern classification using a hidden Markov model (HMM) with frame-based spectral features comprised of Greenwood frequency cepstral coefficients. Individual discrimination accuracy is evaluated as a function of spectral model complexity, represented by the number of mixtures in the underlying Gaussian mixture model (GMM), and temporal model complexity, represented by the number of sequential states in the HMM. Results indicate that the temporal pattern of the vocalization is the most significant factor in accurate discrimination. Overall baseline discrimination accuracy for this data set is about 70% using high level features without complex spectral or temporal models. Accuracy increases to about 80% when more complex spectral models (multiple mixture GMMs) are incorporated, and increases to a final accuracy of 90% when more detailed temporal models (10-state HMMs) are used. Classification accuracy is stable across a relatively wide range of configurations in terms of spectral and temporal model resolution.

AB - This paper investigates the extent of tiger (Panthera tigris) vocal individuality through both qualitative and quantitative approaches using long distance roars from six individual tigers at Omaha's Henry Doorly Zoo in Omaha, NE. The framework for comparison across individuals includes statistical and discriminant function analysis across whole vocalization measures and statistical pattern classification using a hidden Markov model (HMM) with frame-based spectral features comprised of Greenwood frequency cepstral coefficients. Individual discrimination accuracy is evaluated as a function of spectral model complexity, represented by the number of mixtures in the underlying Gaussian mixture model (GMM), and temporal model complexity, represented by the number of sequential states in the HMM. Results indicate that the temporal pattern of the vocalization is the most significant factor in accurate discrimination. Overall baseline discrimination accuracy for this data set is about 70% using high level features without complex spectral or temporal models. Accuracy increases to about 80% when more complex spectral models (multiple mixture GMMs) are incorporated, and increases to a final accuracy of 90% when more detailed temporal models (10-state HMMs) are used. Classification accuracy is stable across a relatively wide range of configurations in terms of spectral and temporal model resolution.

UR - http://www.scopus.com/inward/record.url?scp=84874911108&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874911108&partnerID=8YFLogxK

U2 - 10.1121/1.4789936

DO - 10.1121/1.4789936

M3 - Article

C2 - 23464045

AN - SCOPUS:84874911108

VL - 133

SP - 1762

EP - 1769

JO - Journal of the Acoustical Society of America

JF - Journal of the Acoustical Society of America

SN - 0001-4966

IS - 3

ER -