Detection of abnormal sound using multi-stage GMM for surveillance microphone

Akinori Ito; Akihito Aiba; Masashi Ito; Shozo Makino

doi:10.1109/IAS.2009.160

Detection of abnormal sound using multi-stage GMM for surveillance microphone

Akinori Ito, Akihito Aiba, Masashi Ito, Shozo Makino

ENG - Communications Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

11 Citations (Scopus)

Abstract

We developed a system that detects abnormal sound from sound signal observed by a surveillance microphone. Our system learns the "normal sound" from observation of the microphone, and then detects sounds never observed before as "abnormal sounds." To this end, we developed a technique that uses multiple GMMs for modeling different levels of sound events efficiently. We also consider how to determine thresholds of GMM switching and event detection. As a result, we obtained almost same detection performance using the percentile method to the manually optimized GMMs. Besides, we exploited the segment-based feature, which gave the best result among all methods.

Original language	English
Title of host publication	5th International Conference on Information Assurance and Security, IAS 2009
Pages	733-736
Number of pages	4
DOIs	https://doi.org/10.1109/IAS.2009.160
Publication status	Published - 2009 Dec 1
Event	5th International Conference on Information Assurance and Security, IAS 2009 - Xian, China Duration: 2009 Aug 18 → 2009 Sep 20

Publication series

Name	5th International Conference on Information Assurance and Security, IAS 2009
Volume	1

Other

Other	5th International Conference on Information Assurance and Security, IAS 2009
Country	China
City	Xian
Period	09/8/18 → 09/9/20

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications
Hardware and Architecture
Software

Access to Document

10.1109/IAS.2009.160

Fingerprint Dive into the research topics of 'Detection of abnormal sound using multi-stage GMM for surveillance microphone'. Together they form a unique fingerprint.

Cite this

Detection of abnormal sound using multi-stage GMM for surveillance microphone. / Ito, Akinori; Aiba, Akihito; Ito, Masashi; Makino, Shozo.

5th International Conference on Information Assurance and Security, IAS 2009. 2009. p. 733-736 5283337 (5th International Conference on Information Assurance and Security, IAS 2009; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Ito, A, Aiba, A, Ito, M & Makino, S 2009, Detection of abnormal sound using multi-stage GMM for surveillance microphone. in 5th International Conference on Information Assurance and Security, IAS 2009., 5283337, 5th International Conference on Information Assurance and Security, IAS 2009, vol. 1, pp. 733-736, 5th International Conference on Information Assurance and Security, IAS 2009, Xian, China, 09/8/18. https://doi.org/10.1109/IAS.2009.160

@inproceedings{63aa1afc52d24ea69bc232daa5ab32c5,

title = "Detection of abnormal sound using multi-stage GMM for surveillance microphone",

abstract = "We developed a system that detects abnormal sound from sound signal observed by a surveillance microphone. Our system learns the {"}normal sound{"} from observation of the microphone, and then detects sounds never observed before as {"}abnormal sounds.{"} To this end, we developed a technique that uses multiple GMMs for modeling different levels of sound events efficiently. We also consider how to determine thresholds of GMM switching and event detection. As a result, we obtained almost same detection performance using the percentile method to the manually optimized GMMs. Besides, we exploited the segment-based feature, which gave the best result among all methods.",

author = "Akinori Ito and Akihito Aiba and Masashi Ito and Shozo Makino",

year = "2009",

month = dec,

day = "1",

doi = "10.1109/IAS.2009.160",

language = "English",

isbn = "9780769537443",

series = "5th International Conference on Information Assurance and Security, IAS 2009",

pages = "733--736",

booktitle = "5th International Conference on Information Assurance and Security, IAS 2009",

note = "5th International Conference on Information Assurance and Security, IAS 2009 ; Conference date: 18-08-2009 Through 20-09-2009",

}

TY - GEN

T1 - Detection of abnormal sound using multi-stage GMM for surveillance microphone

AU - Ito, Akinori

AU - Aiba, Akihito

AU - Ito, Masashi

AU - Makino, Shozo

PY - 2009/12/1

Y1 - 2009/12/1

N2 - We developed a system that detects abnormal sound from sound signal observed by a surveillance microphone. Our system learns the "normal sound" from observation of the microphone, and then detects sounds never observed before as "abnormal sounds." To this end, we developed a technique that uses multiple GMMs for modeling different levels of sound events efficiently. We also consider how to determine thresholds of GMM switching and event detection. As a result, we obtained almost same detection performance using the percentile method to the manually optimized GMMs. Besides, we exploited the segment-based feature, which gave the best result among all methods.

AB - We developed a system that detects abnormal sound from sound signal observed by a surveillance microphone. Our system learns the "normal sound" from observation of the microphone, and then detects sounds never observed before as "abnormal sounds." To this end, we developed a technique that uses multiple GMMs for modeling different levels of sound events efficiently. We also consider how to determine thresholds of GMM switching and event detection. As a result, we obtained almost same detection performance using the percentile method to the manually optimized GMMs. Besides, we exploited the segment-based feature, which gave the best result among all methods.

UR - http://www.scopus.com/inward/record.url?scp=74049143834&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74049143834&partnerID=8YFLogxK

U2 - 10.1109/IAS.2009.160

DO - 10.1109/IAS.2009.160

M3 - Conference contribution

AN - SCOPUS:74049143834

SN - 9780769537443

T3 - 5th International Conference on Information Assurance and Security, IAS 2009

SP - 733

EP - 736

BT - 5th International Conference on Information Assurance and Security, IAS 2009

T2 - 5th International Conference on Information Assurance and Security, IAS 2009

Y2 - 18 August 2009 through 20 September 2009

ER -