Environmental sound classification based on feature collaboration

Byeong Jun Han; Eenjun Hwang

doi:10.1109/ICME.2009.5202553

Environmental sound classification based on feature collaboration

Byeong Jun Han, Eenjun Hwang

School of Electrical Engineering

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

25 Citations (Scopus)

Abstract

To date, common acoustic features such as MPEG-7 and Fourier/wavelet transform-based features have been frequently used for environmental sound classification. However, these transforms have difficulty dealing with specific properties of environmental sounds, due to their limited scopes. In this paper, we investigate three types of transforms as yet untried for this purpose, and show that they are more effective than traditional features. This result is mainly due to the fact that they have functionalities that were not easily treatable with traditional transforms. Experimental results show that the combination of these features with traditional features can achieve 86.09% of the maximum accuracy in environmental sound classification, compared to 74.35% of the maximum accuracy when confined to traditional features.

Original language	English
Title of host publication	Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009
Pages	542-545
Number of pages	4
DOIs	https://doi.org/10.1109/ICME.2009.5202553
Publication status	Published - 2009
Event	2009 IEEE International Conference on Multimedia and Expo, ICME 2009 - New York, NY, United States Duration: 2009 Jun 28 → 2009 Jul 3

Publication series

Name	Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009

Other

Other	2009 IEEE International Conference on Multimedia and Expo, ICME 2009
Country	United States
City	New York, NY
Period	09/6/28 → 09/7/3

Keywords

Discrete Hilbert transform
Discrete chirplet transform
Discrete curvelet transform
Environmental sound recognition
Feature extraction

ASJC Scopus subject areas

Computer Graphics and Computer-Aided Design
Computer Networks and Communications
Hardware and Architecture
Software

Access to Document

10.1109/ICME.2009.5202553

Fingerprint Dive into the research topics of 'Environmental sound classification based on feature collaboration'. Together they form a unique fingerprint.

Cite this

Environmental sound classification based on feature collaboration. / Han, Byeong Jun; Hwang, Eenjun.

Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009. 2009. p. 542-545 5202553 (Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009).

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

Han, BJ & Hwang, E 2009, Environmental sound classification based on feature collaboration. in Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009., 5202553, Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009, pp. 542-545, 2009 IEEE International Conference on Multimedia and Expo, ICME 2009, New York, NY, United States, 09/6/28. https://doi.org/10.1109/ICME.2009.5202553

@inproceedings{4d7b32e4b66e420d92ecb550e556ab20,

title = "Environmental sound classification based on feature collaboration",

abstract = "To date, common acoustic features such as MPEG-7 and Fourier/wavelet transform-based features have been frequently used for environmental sound classification. However, these transforms have difficulty dealing with specific properties of environmental sounds, due to their limited scopes. In this paper, we investigate three types of transforms as yet untried for this purpose, and show that they are more effective than traditional features. This result is mainly due to the fact that they have functionalities that were not easily treatable with traditional transforms. Experimental results show that the combination of these features with traditional features can achieve 86.09% of the maximum accuracy in environmental sound classification, compared to 74.35% of the maximum accuracy when confined to traditional features.",

keywords = "Discrete Hilbert transform, Discrete chirplet transform, Discrete curvelet transform, Environmental sound recognition, Feature extraction",

author = "Han, {Byeong Jun} and Eenjun Hwang",

year = "2009",

doi = "10.1109/ICME.2009.5202553",

language = "English",

isbn = "9781424442911",

series = "Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009",

pages = "542--545",

booktitle = "Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009",

note = "2009 IEEE International Conference on Multimedia and Expo, ICME 2009 ; Conference date: 28-06-2009 Through 03-07-2009",

}

TY - GEN

T1 - Environmental sound classification based on feature collaboration

AU - Han, Byeong Jun

AU - Hwang, Eenjun

PY - 2009

Y1 - 2009

N2 - To date, common acoustic features such as MPEG-7 and Fourier/wavelet transform-based features have been frequently used for environmental sound classification. However, these transforms have difficulty dealing with specific properties of environmental sounds, due to their limited scopes. In this paper, we investigate three types of transforms as yet untried for this purpose, and show that they are more effective than traditional features. This result is mainly due to the fact that they have functionalities that were not easily treatable with traditional transforms. Experimental results show that the combination of these features with traditional features can achieve 86.09% of the maximum accuracy in environmental sound classification, compared to 74.35% of the maximum accuracy when confined to traditional features.

AB - To date, common acoustic features such as MPEG-7 and Fourier/wavelet transform-based features have been frequently used for environmental sound classification. However, these transforms have difficulty dealing with specific properties of environmental sounds, due to their limited scopes. In this paper, we investigate three types of transforms as yet untried for this purpose, and show that they are more effective than traditional features. This result is mainly due to the fact that they have functionalities that were not easily treatable with traditional transforms. Experimental results show that the combination of these features with traditional features can achieve 86.09% of the maximum accuracy in environmental sound classification, compared to 74.35% of the maximum accuracy when confined to traditional features.

KW - Discrete Hilbert transform

KW - Discrete chirplet transform

KW - Discrete curvelet transform

KW - Environmental sound recognition

KW - Feature extraction

UR - http://www.scopus.com/inward/record.url?scp=70449559297&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449559297&partnerID=8YFLogxK

U2 - 10.1109/ICME.2009.5202553

DO - 10.1109/ICME.2009.5202553

M3 - Conference contribution

AN - SCOPUS:70449559297

SN - 9781424442911

T3 - Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009

SP - 542

EP - 545

BT - Proceedings - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009

T2 - 2009 IEEE International Conference on Multimedia and Expo, ICME 2009

Y2 - 28 June 2009 through 3 July 2009

ER -

Environmental sound classification based on feature collaboration

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Cite this