Hierarchical approach for abnormal acoustic event classification in an elevator

Kwangyoun Kim, Hanseok Ko

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

13 Citations (Scopus)

Abstract

In this paper, we propose a hierarchical method to detect and classify abnormal acoustic events occurring in an elevator environment. The Gaussian Mixture Model (GMM) based event classifier essentially employs two types of acoustic features; Mel Frequency Cepstral Coefficient (MFCC) and Timbre. We explore the effectiveness of various combinations of the two features in terms of classification performance. In addition, we design a hierarchical approach for realizing acoustic event classification and compare it with a single-level approach. It can be verified from an experiment, that the classification performance is improved when the proposed hierarchical approach is applied. In particular, for detection of abnormal situations, we employ a maximum likelihood estimation approach for acoustic event recognition at the 1st step, and then on the 2nd step we determine the abnormal contexts by using the ratio of abnormal events to cumulative events during a certain period. For performance evaluation, we employ a database collected in an actual elevator under several scenarios. By experimental results, our proposed method demonstrates 91% correct detection rate and 2.5% error detection rate for abnormal context.

Original languageEnglish
Title of host publication2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011
Pages89-94
Number of pages6
DOIs
Publication statusPublished - 2011 Oct 17
Event2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011 - Klagenfurt, Austria
Duration: 2011 Aug 302011 Sep 2

Other

Other2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011
CountryAustria
CityKlagenfurt
Period11/8/3011/9/2

Fingerprint

Elevators
Acoustics
Maximum likelihood estimation
Error detection
Classifiers
Experiments

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Signal Processing

Cite this

Kim, K., & Ko, H. (2011). Hierarchical approach for abnormal acoustic event classification in an elevator. In 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011 (pp. 89-94). [6027300] https://doi.org/10.1109/AVSS.2011.6027300

Hierarchical approach for abnormal acoustic event classification in an elevator. / Kim, Kwangyoun; Ko, Hanseok.

2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011. 2011. p. 89-94 6027300.

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

Kim, K & Ko, H 2011, Hierarchical approach for abnormal acoustic event classification in an elevator. in 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011., 6027300, pp. 89-94, 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011, Klagenfurt, Austria, 11/8/30. https://doi.org/10.1109/AVSS.2011.6027300
Kim K, Ko H. Hierarchical approach for abnormal acoustic event classification in an elevator. In 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011. 2011. p. 89-94. 6027300 https://doi.org/10.1109/AVSS.2011.6027300
Kim, Kwangyoun ; Ko, Hanseok. / Hierarchical approach for abnormal acoustic event classification in an elevator. 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011. 2011. pp. 89-94
@inproceedings{74a22bc603a44595b4984904e580b159,
title = "Hierarchical approach for abnormal acoustic event classification in an elevator",
abstract = "In this paper, we propose a hierarchical method to detect and classify abnormal acoustic events occurring in an elevator environment. The Gaussian Mixture Model (GMM) based event classifier essentially employs two types of acoustic features; Mel Frequency Cepstral Coefficient (MFCC) and Timbre. We explore the effectiveness of various combinations of the two features in terms of classification performance. In addition, we design a hierarchical approach for realizing acoustic event classification and compare it with a single-level approach. It can be verified from an experiment, that the classification performance is improved when the proposed hierarchical approach is applied. In particular, for detection of abnormal situations, we employ a maximum likelihood estimation approach for acoustic event recognition at the 1st step, and then on the 2nd step we determine the abnormal contexts by using the ratio of abnormal events to cumulative events during a certain period. For performance evaluation, we employ a database collected in an actual elevator under several scenarios. By experimental results, our proposed method demonstrates 91{\%} correct detection rate and 2.5{\%} error detection rate for abnormal context.",
author = "Kwangyoun Kim and Hanseok Ko",
year = "2011",
month = "10",
day = "17",
doi = "10.1109/AVSS.2011.6027300",
language = "English",
isbn = "9781457708459",
pages = "89--94",
booktitle = "2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011",

}

TY - GEN

T1 - Hierarchical approach for abnormal acoustic event classification in an elevator

AU - Kim, Kwangyoun

AU - Ko, Hanseok

PY - 2011/10/17

Y1 - 2011/10/17

N2 - In this paper, we propose a hierarchical method to detect and classify abnormal acoustic events occurring in an elevator environment. The Gaussian Mixture Model (GMM) based event classifier essentially employs two types of acoustic features; Mel Frequency Cepstral Coefficient (MFCC) and Timbre. We explore the effectiveness of various combinations of the two features in terms of classification performance. In addition, we design a hierarchical approach for realizing acoustic event classification and compare it with a single-level approach. It can be verified from an experiment, that the classification performance is improved when the proposed hierarchical approach is applied. In particular, for detection of abnormal situations, we employ a maximum likelihood estimation approach for acoustic event recognition at the 1st step, and then on the 2nd step we determine the abnormal contexts by using the ratio of abnormal events to cumulative events during a certain period. For performance evaluation, we employ a database collected in an actual elevator under several scenarios. By experimental results, our proposed method demonstrates 91% correct detection rate and 2.5% error detection rate for abnormal context.

AB - In this paper, we propose a hierarchical method to detect and classify abnormal acoustic events occurring in an elevator environment. The Gaussian Mixture Model (GMM) based event classifier essentially employs two types of acoustic features; Mel Frequency Cepstral Coefficient (MFCC) and Timbre. We explore the effectiveness of various combinations of the two features in terms of classification performance. In addition, we design a hierarchical approach for realizing acoustic event classification and compare it with a single-level approach. It can be verified from an experiment, that the classification performance is improved when the proposed hierarchical approach is applied. In particular, for detection of abnormal situations, we employ a maximum likelihood estimation approach for acoustic event recognition at the 1st step, and then on the 2nd step we determine the abnormal contexts by using the ratio of abnormal events to cumulative events during a certain period. For performance evaluation, we employ a database collected in an actual elevator under several scenarios. By experimental results, our proposed method demonstrates 91% correct detection rate and 2.5% error detection rate for abnormal context.

UR - http://www.scopus.com/inward/record.url?scp=80053935158&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053935158&partnerID=8YFLogxK

U2 - 10.1109/AVSS.2011.6027300

DO - 10.1109/AVSS.2011.6027300

M3 - Conference contribution

AN - SCOPUS:80053935158

SN - 9781457708459

SP - 89

EP - 94

BT - 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2011

ER -