Human activity recognition from environmental background sounds for Wireless Sensor Networks

Yi Zhan, Shun Miura, Jun Nishimura, Tadahiro Kuroda

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

9 Citations (Scopus)

Abstract

Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and classification Dynamic Time Warping (DTW) algorithms are applied to recognizing the background sounds in the human daily activities. Applying these algorithms to fourteen typical daily activity sounds, average recognition accuracy of 92.5% can be achieved. In these algorithms, how two parameters (Le., Mel filters number and frame-to-frame overlap) affect system's calculation burden and accuracy is also investigated, By adjusting these two parameters to a suitable combination, the calculation burden can be reduced by 61.6% while maintaining the system's average accuracy rate at approximate 90%. This is promising for future integrating with other sensor(s) to fulfill daily activity recognition work by using power aware Wireless Sensor Networks (WSN) system.

Original languageEnglish
Title of host publication2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07
Pages307-312
Number of pages6
DOIs
Publication statusPublished - 2007
Event2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07 - London, United Kingdom
Duration: 2007 Apr 152007 Apr 17

Other

Other2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07
CountryUnited Kingdom
CityLondon
Period07/4/1507/4/17

Fingerprint

Wireless sensor networks
Acoustic waves
Feature extraction
Sensors

Keywords

  • Calculation burden
  • DTW
  • MFCC
  • Sound recognition
  • WSN

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Control and Systems Engineering

Cite this

Zhan, Y., Miura, S., Nishimura, J., & Kuroda, T. (2007). Human activity recognition from environmental background sounds for Wireless Sensor Networks. In 2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07 (pp. 307-312). [4239009] https://doi.org/10.1109/ICNSC.2007.372796

Human activity recognition from environmental background sounds for Wireless Sensor Networks. / Zhan, Yi; Miura, Shun; Nishimura, Jun; Kuroda, Tadahiro.

2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07. 2007. p. 307-312 4239009.

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

Zhan, Y, Miura, S, Nishimura, J & Kuroda, T 2007, Human activity recognition from environmental background sounds for Wireless Sensor Networks. in 2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07., 4239009, pp. 307-312, 2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07, London, United Kingdom, 07/4/15. https://doi.org/10.1109/ICNSC.2007.372796
Zhan Y, Miura S, Nishimura J, Kuroda T. Human activity recognition from environmental background sounds for Wireless Sensor Networks. In 2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07. 2007. p. 307-312. 4239009 https://doi.org/10.1109/ICNSC.2007.372796
Zhan, Yi ; Miura, Shun ; Nishimura, Jun ; Kuroda, Tadahiro. / Human activity recognition from environmental background sounds for Wireless Sensor Networks. 2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07. 2007. pp. 307-312
@inproceedings{9467d7cee0c747f3af040611b913466a,
title = "Human activity recognition from environmental background sounds for Wireless Sensor Networks",
abstract = "Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and classification Dynamic Time Warping (DTW) algorithms are applied to recognizing the background sounds in the human daily activities. Applying these algorithms to fourteen typical daily activity sounds, average recognition accuracy of 92.5{\%} can be achieved. In these algorithms, how two parameters (Le., Mel filters number and frame-to-frame overlap) affect system's calculation burden and accuracy is also investigated, By adjusting these two parameters to a suitable combination, the calculation burden can be reduced by 61.6{\%} while maintaining the system's average accuracy rate at approximate 90{\%}. This is promising for future integrating with other sensor(s) to fulfill daily activity recognition work by using power aware Wireless Sensor Networks (WSN) system.",
keywords = "Calculation burden, DTW, MFCC, Sound recognition, WSN",
author = "Yi Zhan and Shun Miura and Jun Nishimura and Tadahiro Kuroda",
year = "2007",
doi = "10.1109/ICNSC.2007.372796",
language = "English",
isbn = "1424410762",
pages = "307--312",
booktitle = "2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07",

}

TY - GEN

T1 - Human activity recognition from environmental background sounds for Wireless Sensor Networks

AU - Zhan, Yi

AU - Miura, Shun

AU - Nishimura, Jun

AU - Kuroda, Tadahiro

PY - 2007

Y1 - 2007

N2 - Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and classification Dynamic Time Warping (DTW) algorithms are applied to recognizing the background sounds in the human daily activities. Applying these algorithms to fourteen typical daily activity sounds, average recognition accuracy of 92.5% can be achieved. In these algorithms, how two parameters (Le., Mel filters number and frame-to-frame overlap) affect system's calculation burden and accuracy is also investigated, By adjusting these two parameters to a suitable combination, the calculation burden can be reduced by 61.6% while maintaining the system's average accuracy rate at approximate 90%. This is promising for future integrating with other sensor(s) to fulfill daily activity recognition work by using power aware Wireless Sensor Networks (WSN) system.

AB - Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and classification Dynamic Time Warping (DTW) algorithms are applied to recognizing the background sounds in the human daily activities. Applying these algorithms to fourteen typical daily activity sounds, average recognition accuracy of 92.5% can be achieved. In these algorithms, how two parameters (Le., Mel filters number and frame-to-frame overlap) affect system's calculation burden and accuracy is also investigated, By adjusting these two parameters to a suitable combination, the calculation burden can be reduced by 61.6% while maintaining the system's average accuracy rate at approximate 90%. This is promising for future integrating with other sensor(s) to fulfill daily activity recognition work by using power aware Wireless Sensor Networks (WSN) system.

KW - Calculation burden

KW - DTW

KW - MFCC

KW - Sound recognition

KW - WSN

UR - http://www.scopus.com/inward/record.url?scp=34748864703&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34748864703&partnerID=8YFLogxK

U2 - 10.1109/ICNSC.2007.372796

DO - 10.1109/ICNSC.2007.372796

M3 - Conference contribution

AN - SCOPUS:34748864703

SN - 1424410762

SN - 9781424410767

SP - 307

EP - 312

BT - 2007 IEEE International Conference on Networking, Sensing and Control, ICNSC'07

ER -