Local Binary Pattern with Random Forest for Acoustic Scene Classification

Shamsiah Abidin; Xianjun Xia; Robert Togneri; Ferdous Sohel

doi:10.1109/ICME.2018.8486578

Local Binary Pattern with Random Forest for Acoustic Scene Classification

Shamsiah Abidin, Xianjun Xia, Robert Togneri, Ferdous Sohel

Research output: Chapter in Book/Conference paper › Conference paper › peer-review

8 Citations (Scopus)

Abstract

This paper presents an approach for acoustic scene classification using the local binary pattern (LBP) and random forest (RF). The audio signal is converted to a Constant-Q transform (CQT) representation and LBP is used to extract the features from this time-frequency representation. The CQT representations are divided into a number of sub-bands to obtain more localized features relevant to the spectral information. We then use random forest to select the most important features for each band of extracted LBP features. For further performance enhancement, we use feature level fusion of LBP and HOG features. The proposed system has achieved an accuracy of 85% on the DCASE 2016 dataset.

Original language	English
Title of host publication	2018 IEEE International Conference on Multimedia and Expo, ICME 2018
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Volume	2018-July
ISBN (Electronic)	9781538617373
DOIs	https://doi.org/10.1109/ICME.2018.8486578
Publication status	Published - 8 Oct 2018
Event	2018 IEEE International Conference on Multimedia and Expo, ICME 2018 - San Diego, United States Duration: 23 Jul 2018 → 27 Jul 2018

Conference

Conference	2018 IEEE International Conference on Multimedia and Expo, ICME 2018
Country/Territory	United States
City	San Diego
Period	23/07/18 → 27/07/18

Access to Document

10.1109/ICME.2018.8486578

Cite this

@inproceedings{afc2c4b1627d454e844755928cf1dc07,

title = "Local Binary Pattern with Random Forest for Acoustic Scene Classification",

abstract = "This paper presents an approach for acoustic scene classification using the local binary pattern (LBP) and random forest (RF). The audio signal is converted to a Constant-Q transform (CQT) representation and LBP is used to extract the features from this time-frequency representation. The CQT representations are divided into a number of sub-bands to obtain more localized features relevant to the spectral information. We then use random forest to select the most important features for each band of extracted LBP features. For further performance enhancement, we use feature level fusion of LBP and HOG features. The proposed system has achieved an accuracy of 85% on the DCASE 2016 dataset.",

keywords = "acoustic scene, feature extraction, fusion, local binary patterns, random forest, time-frequency analysis",

author = "Shamsiah Abidin and Xianjun Xia and Robert Togneri and Ferdous Sohel",

year = "2018",

month = oct,

day = "8",

doi = "10.1109/ICME.2018.8486578",

language = "English",

volume = "2018-July",

booktitle = "2018 IEEE International Conference on Multimedia and Expo, ICME 2018",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2018 IEEE International Conference on Multimedia and Expo, ICME 2018 ; Conference date: 23-07-2018 Through 27-07-2018",

}

Abidin, S, Xia, X, Togneri, R & Sohel, F 2018, Local Binary Pattern with Random Forest for Acoustic Scene Classification. in 2018 IEEE International Conference on Multimedia and Expo, ICME 2018. vol. 2018-July, 8486578, IEEE, Institute of Electrical and Electronics Engineers, 2018 IEEE International Conference on Multimedia and Expo, ICME 2018, San Diego, United States, 23/07/18. https://doi.org/10.1109/ICME.2018.8486578

Local Binary Pattern with Random Forest for Acoustic Scene Classification. / Abidin, Shamsiah; Xia, Xianjun; Togneri, Robert et al.
2018 IEEE International Conference on Multimedia and Expo, ICME 2018. Vol. 2018-July IEEE, Institute of Electrical and Electronics Engineers, 2018. 8486578.

Research output: Chapter in Book/Conference paper › Conference paper › peer-review

TY - GEN

T1 - Local Binary Pattern with Random Forest for Acoustic Scene Classification

AU - Abidin, Shamsiah

AU - Xia, Xianjun

AU - Togneri, Robert

AU - Sohel, Ferdous

PY - 2018/10/8

Y1 - 2018/10/8

N2 - This paper presents an approach for acoustic scene classification using the local binary pattern (LBP) and random forest (RF). The audio signal is converted to a Constant-Q transform (CQT) representation and LBP is used to extract the features from this time-frequency representation. The CQT representations are divided into a number of sub-bands to obtain more localized features relevant to the spectral information. We then use random forest to select the most important features for each band of extracted LBP features. For further performance enhancement, we use feature level fusion of LBP and HOG features. The proposed system has achieved an accuracy of 85% on the DCASE 2016 dataset.

AB - This paper presents an approach for acoustic scene classification using the local binary pattern (LBP) and random forest (RF). The audio signal is converted to a Constant-Q transform (CQT) representation and LBP is used to extract the features from this time-frequency representation. The CQT representations are divided into a number of sub-bands to obtain more localized features relevant to the spectral information. We then use random forest to select the most important features for each band of extracted LBP features. For further performance enhancement, we use feature level fusion of LBP and HOG features. The proposed system has achieved an accuracy of 85% on the DCASE 2016 dataset.

KW - acoustic scene

KW - feature extraction

KW - fusion

KW - local binary patterns

KW - random forest

KW - time-frequency analysis

UR - http://www.scopus.com/inward/record.url?scp=85061448460&partnerID=8YFLogxK

U2 - 10.1109/ICME.2018.8486578

DO - 10.1109/ICME.2018.8486578

M3 - Conference paper

AN - SCOPUS:85061448460

VL - 2018-July

BT - 2018 IEEE International Conference on Multimedia and Expo, ICME 2018

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - 2018 IEEE International Conference on Multimedia and Expo, ICME 2018

Y2 - 23 July 2018 through 27 July 2018

ER -

Local Binary Pattern with Random Forest for Acoustic Scene Classification

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this