A smartphone based real-time daily activity monitoring system

Shumei Zhang, Paul McCullagh, Jing Zhang, Tiezhong Yu

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

A real-time activity monitoring system within anAndroid based smartphone is proposed and evaluated. Motionand motionless postures may be classified using principlesof kinematical theory, which underpins hierarchicalrule-based algorithms, based on accelerometer and orientationdata. Falls detection was implemented by analyzingwhether the postures classified as ‘lying’ or ‘sit-tilted’ postureare deemed normal or abnormal, based on the analysisof time, users’ current position and posture transition. Experimentalresults demonstrate that the approach can detectvarious types of falls efficiently (i.e., in real-time within asmart phone processor) and also correctly (95 % and 93 %true positives for falls ending with ‘lying’ and ‘sit-tilted’ respectively).The approach is reliable for different subjectsand different situations, since it is not only based on empiricalthresholds and subject-based training models, but inaddition it is underpinned by theory.
LanguageEnglish
Pages711-721
JournalCluster Computing
Volume17
Issue number3
DOIs
Publication statusPublished - 1 Sep 2014

Fingerprint

Smartphones
Accelerometers
Monitoring

Cite this

Zhang, Shumei ; McCullagh, Paul ; Zhang, Jing ; Yu, Tiezhong. / A smartphone based real-time daily activity monitoring system. In: Cluster Computing. 2014 ; Vol. 17, No. 3. pp. 711-721.
@article{814199ab29674b51bddc97100abd558a,
title = "A smartphone based real-time daily activity monitoring system",
abstract = "A real-time activity monitoring system within anAndroid based smartphone is proposed and evaluated. Motionand motionless postures may be classified using principlesof kinematical theory, which underpins hierarchicalrule-based algorithms, based on accelerometer and orientationdata. Falls detection was implemented by analyzingwhether the postures classified as ‘lying’ or ‘sit-tilted’ postureare deemed normal or abnormal, based on the analysisof time, users’ current position and posture transition. Experimentalresults demonstrate that the approach can detectvarious types of falls efficiently (i.e., in real-time within asmart phone processor) and also correctly (95 {\%} and 93 {\%}true positives for falls ending with ‘lying’ and ‘sit-tilted’ respectively).The approach is reliable for different subjectsand different situations, since it is not only based on empiricalthresholds and subject-based training models, but inaddition it is underpinned by theory.",
author = "Shumei Zhang and Paul McCullagh and Jing Zhang and Tiezhong Yu",
year = "2014",
month = "9",
day = "1",
doi = "10.1007/s10586-013-0335-y",
language = "English",
volume = "17",
pages = "711--721",
journal = "Cluster Computing",
issn = "1386-7857",
number = "3",

}

A smartphone based real-time daily activity monitoring system. / Zhang, Shumei; McCullagh, Paul; Zhang, Jing; Yu, Tiezhong.

In: Cluster Computing, Vol. 17, No. 3, 01.09.2014, p. 711-721.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A smartphone based real-time daily activity monitoring system

AU - Zhang, Shumei

AU - McCullagh, Paul

AU - Zhang, Jing

AU - Yu, Tiezhong

PY - 2014/9/1

Y1 - 2014/9/1

N2 - A real-time activity monitoring system within anAndroid based smartphone is proposed and evaluated. Motionand motionless postures may be classified using principlesof kinematical theory, which underpins hierarchicalrule-based algorithms, based on accelerometer and orientationdata. Falls detection was implemented by analyzingwhether the postures classified as ‘lying’ or ‘sit-tilted’ postureare deemed normal or abnormal, based on the analysisof time, users’ current position and posture transition. Experimentalresults demonstrate that the approach can detectvarious types of falls efficiently (i.e., in real-time within asmart phone processor) and also correctly (95 % and 93 %true positives for falls ending with ‘lying’ and ‘sit-tilted’ respectively).The approach is reliable for different subjectsand different situations, since it is not only based on empiricalthresholds and subject-based training models, but inaddition it is underpinned by theory.

AB - A real-time activity monitoring system within anAndroid based smartphone is proposed and evaluated. Motionand motionless postures may be classified using principlesof kinematical theory, which underpins hierarchicalrule-based algorithms, based on accelerometer and orientationdata. Falls detection was implemented by analyzingwhether the postures classified as ‘lying’ or ‘sit-tilted’ postureare deemed normal or abnormal, based on the analysisof time, users’ current position and posture transition. Experimentalresults demonstrate that the approach can detectvarious types of falls efficiently (i.e., in real-time within asmart phone processor) and also correctly (95 % and 93 %true positives for falls ending with ‘lying’ and ‘sit-tilted’ respectively).The approach is reliable for different subjectsand different situations, since it is not only based on empiricalthresholds and subject-based training models, but inaddition it is underpinned by theory.

U2 - 10.1007/s10586-013-0335-y

DO - 10.1007/s10586-013-0335-y

M3 - Article

VL - 17

SP - 711

EP - 721

JO - Cluster Computing

T2 - Cluster Computing

JF - Cluster Computing

SN - 1386-7857

IS - 3

ER -