Acoustic fall detection using Gaussian mixture models and GMM supervectors

Xiaodan Zhuang, Jing Huang, Gerasimos Potamianos, Mark Hasegawa-Johnson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a system that detects human falls in the home environment, distinguishing them from competing noise, by using only the audio signal from a single far-field microphone. The proposed system models each fall or noise segment by means of a Gaussian mixture model (GMM) supervector, whose Euclidean distance measures the pairwise difference between audio segments. A support vector machine built on a kernel between GMM supervectors is employed to classify audio segments into falls and various types of noise. Experiments on a dataset of human falls, collected as part of the Netcarity project, show that the method improves fall classification F-score to 67% from 59% of a baseline GMM classifier. The approach also effectively addresses the more difficult fall detection problem, where audio segment boundaries are unknown. Specifically, we employ it to reclassify confusable segments produced by a dynamic programming scheme based on traditional GMMs. Such post-processing improves a fall detection accuracy metric by 5% relative.

Original languageEnglish (US)
Title of host publication2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009
Pages69-72
Number of pages4
DOIs
StatePublished - Sep 23 2009
Event2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009 - Taipei, Taiwan, Province of China
Duration: Apr 19 2009Apr 24 2009

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)1520-6149

Other

Other2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009
CountryTaiwan, Province of China
CityTaipei
Period4/19/094/24/09

Fingerprint

Acoustics
Microphones
Dynamic programming
Support vector machines
Classifiers
Processing
Experiments

Keywords

  • Fall detection
  • GMM supervector
  • Gaussian mixture model
  • Support vector machine

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Zhuang, X., Huang, J., Potamianos, G., & Hasegawa-Johnson, M. (2009). Acoustic fall detection using Gaussian mixture models and GMM supervectors. In 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009 (pp. 69-72). [4959522] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2009.4959522

Acoustic fall detection using Gaussian mixture models and GMM supervectors. / Zhuang, Xiaodan; Huang, Jing; Potamianos, Gerasimos; Hasegawa-Johnson, Mark.

2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. p. 69-72 4959522 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhuang, X, Huang, J, Potamianos, G & Hasegawa-Johnson, M 2009, Acoustic fall detection using Gaussian mixture models and GMM supervectors. in 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009., 4959522, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, pp. 69-72, 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009, Taipei, Taiwan, Province of China, 4/19/09. https://doi.org/10.1109/ICASSP.2009.4959522
Zhuang X, Huang J, Potamianos G, Hasegawa-Johnson M. Acoustic fall detection using Gaussian mixture models and GMM supervectors. In 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. p. 69-72. 4959522. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2009.4959522
Zhuang, Xiaodan ; Huang, Jing ; Potamianos, Gerasimos ; Hasegawa-Johnson, Mark. / Acoustic fall detection using Gaussian mixture models and GMM supervectors. 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. pp. 69-72 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{417d062bd19f4025b402d0597fc014b2,
title = "Acoustic fall detection using Gaussian mixture models and GMM supervectors",
abstract = "We present a system that detects human falls in the home environment, distinguishing them from competing noise, by using only the audio signal from a single far-field microphone. The proposed system models each fall or noise segment by means of a Gaussian mixture model (GMM) supervector, whose Euclidean distance measures the pairwise difference between audio segments. A support vector machine built on a kernel between GMM supervectors is employed to classify audio segments into falls and various types of noise. Experiments on a dataset of human falls, collected as part of the Netcarity project, show that the method improves fall classification F-score to 67{\%} from 59{\%} of a baseline GMM classifier. The approach also effectively addresses the more difficult fall detection problem, where audio segment boundaries are unknown. Specifically, we employ it to reclassify confusable segments produced by a dynamic programming scheme based on traditional GMMs. Such post-processing improves a fall detection accuracy metric by 5{\%} relative.",
keywords = "Fall detection, GMM supervector, Gaussian mixture model, Support vector machine",
author = "Xiaodan Zhuang and Jing Huang and Gerasimos Potamianos and Mark Hasegawa-Johnson",
year = "2009",
month = "9",
day = "23",
doi = "10.1109/ICASSP.2009.4959522",
language = "English (US)",
isbn = "9781424423545",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
pages = "69--72",
booktitle = "2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009",

}

TY - GEN

T1 - Acoustic fall detection using Gaussian mixture models and GMM supervectors

AU - Zhuang, Xiaodan

AU - Huang, Jing

AU - Potamianos, Gerasimos

AU - Hasegawa-Johnson, Mark

PY - 2009/9/23

Y1 - 2009/9/23

N2 - We present a system that detects human falls in the home environment, distinguishing them from competing noise, by using only the audio signal from a single far-field microphone. The proposed system models each fall or noise segment by means of a Gaussian mixture model (GMM) supervector, whose Euclidean distance measures the pairwise difference between audio segments. A support vector machine built on a kernel between GMM supervectors is employed to classify audio segments into falls and various types of noise. Experiments on a dataset of human falls, collected as part of the Netcarity project, show that the method improves fall classification F-score to 67% from 59% of a baseline GMM classifier. The approach also effectively addresses the more difficult fall detection problem, where audio segment boundaries are unknown. Specifically, we employ it to reclassify confusable segments produced by a dynamic programming scheme based on traditional GMMs. Such post-processing improves a fall detection accuracy metric by 5% relative.

AB - We present a system that detects human falls in the home environment, distinguishing them from competing noise, by using only the audio signal from a single far-field microphone. The proposed system models each fall or noise segment by means of a Gaussian mixture model (GMM) supervector, whose Euclidean distance measures the pairwise difference between audio segments. A support vector machine built on a kernel between GMM supervectors is employed to classify audio segments into falls and various types of noise. Experiments on a dataset of human falls, collected as part of the Netcarity project, show that the method improves fall classification F-score to 67% from 59% of a baseline GMM classifier. The approach also effectively addresses the more difficult fall detection problem, where audio segment boundaries are unknown. Specifically, we employ it to reclassify confusable segments produced by a dynamic programming scheme based on traditional GMMs. Such post-processing improves a fall detection accuracy metric by 5% relative.

KW - Fall detection

KW - GMM supervector

KW - Gaussian mixture model

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=70349200911&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349200911&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2009.4959522

DO - 10.1109/ICASSP.2009.4959522

M3 - Conference contribution

AN - SCOPUS:70349200911

SN - 9781424423545

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 69

EP - 72

BT - 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009

ER -