Human activity recognition from environmental background sounds for Wireless Sensor Networks

Yi Zhan, Jun Nishimura, Tadahiro Kuroda

Research output: Contribution to journalArticle

Abstract

Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and Vector Quantization (VQ) classification Linde-Buzo-Gray algorithm (LBG) algorithms are applied for recognizing the background sounds in the human daily activities. Applying these algorithms to twenty typical daily activity sounds, average recognition accuracy of 93.8% can be achieved. In these algorithms, how three parameters (i.e., Mel filters number, frame-to-frame overlap and LBG codebook cluster number) affect system's calculation burden and accuracy is also investigated. By adjusting these three parameters to an optimized combination, the multiplication and addition calculation burden can be reduced by 87.0% and 87.1% individually while maintaining the system's average accuracy rate at 92.5%. This is promising for future integration with other sensor (s) to fulfill daily activity recognition by using power aware Wireless Sensor Networks (WSN) systems.

Original languageEnglish
Pages (from-to)565-572
Number of pages8
JournalIEEJ Transactions on Electronics, Information and Systems
Volume130
Issue number4
Publication statusPublished - 2010

Fingerprint

Wireless sensor networks
Acoustic waves
Vector quantization
Feature extraction
Sensors

Keywords

  • LBQ calculation burden (Cost)
  • MFCC
  • Sound recognition
  • VQ
  • WSN

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

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

In: IEEJ Transactions on Electronics, Information and Systems, Vol. 130, No. 4, 2010, p. 565-572.

Research output: Contribution to journalArticle

@article{7776b5c2abd549aea7805073053de828,
title = "Human activity recognition from environmental background sounds for Wireless Sensor Networks",
abstract = "Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and Vector Quantization (VQ) classification Linde-Buzo-Gray algorithm (LBG) algorithms are applied for recognizing the background sounds in the human daily activities. Applying these algorithms to twenty typical daily activity sounds, average recognition accuracy of 93.8{\%} can be achieved. In these algorithms, how three parameters (i.e., Mel filters number, frame-to-frame overlap and LBG codebook cluster number) affect system's calculation burden and accuracy is also investigated. By adjusting these three parameters to an optimized combination, the multiplication and addition calculation burden can be reduced by 87.0{\%} and 87.1{\%} individually while maintaining the system's average accuracy rate at 92.5{\%}. This is promising for future integration with other sensor (s) to fulfill daily activity recognition by using power aware Wireless Sensor Networks (WSN) systems.",
keywords = "LBQ calculation burden (Cost), MFCC, Sound recognition, VQ, WSN",
author = "Yi Zhan and Jun Nishimura and Tadahiro Kuroda",
year = "2010",
language = "English",
volume = "130",
pages = "565--572",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "4",

}

TY - JOUR

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

AU - Zhan, Yi

AU - Nishimura, Jun

AU - Kuroda, Tadahiro

PY - 2010

Y1 - 2010

N2 - Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and Vector Quantization (VQ) classification Linde-Buzo-Gray algorithm (LBG) algorithms are applied for recognizing the background sounds in the human daily activities. Applying these algorithms to twenty typical daily activity sounds, average recognition accuracy of 93.8% can be achieved. In these algorithms, how three parameters (i.e., Mel filters number, frame-to-frame overlap and LBG codebook cluster number) affect system's calculation burden and accuracy is also investigated. By adjusting these three parameters to an optimized combination, the multiplication and addition calculation burden can be reduced by 87.0% and 87.1% individually while maintaining the system's average accuracy rate at 92.5%. This is promising for future integration with other sensor (s) to fulfill daily activity recognition by using power aware Wireless Sensor Networks (WSN) systems.

AB - Sound feature extraction Mel Frequency Cepstral Coefficients (MFCC) and Vector Quantization (VQ) classification Linde-Buzo-Gray algorithm (LBG) algorithms are applied for recognizing the background sounds in the human daily activities. Applying these algorithms to twenty typical daily activity sounds, average recognition accuracy of 93.8% can be achieved. In these algorithms, how three parameters (i.e., Mel filters number, frame-to-frame overlap and LBG codebook cluster number) affect system's calculation burden and accuracy is also investigated. By adjusting these three parameters to an optimized combination, the multiplication and addition calculation burden can be reduced by 87.0% and 87.1% individually while maintaining the system's average accuracy rate at 92.5%. This is promising for future integration with other sensor (s) to fulfill daily activity recognition by using power aware Wireless Sensor Networks (WSN) systems.

KW - LBQ calculation burden (Cost)

KW - MFCC

KW - Sound recognition

KW - VQ

KW - WSN

UR - http://www.scopus.com/inward/record.url?scp=77953547677&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953547677&partnerID=8YFLogxK

M3 - Article

VL - 130

SP - 565

EP - 572

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 4

ER -