Layered hidden Markov models to recognize activity with built-in sensors on Android smartphone

Young Seol Lee, Sung-Bae Cho

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Recently, activity recognition using built-in sensors in a mobile phone becomes an important issue. It can help us to provide context-aware services: health care, suitable content recommendation for a user’s activity, and user adaptive interface. This paper proposes a layered hidden Markov model (HMM) to recognize both short-term activity and long-term activity in real time. The first layer of HMMs detects short, primitive activities with acceleration, magnetic field, and orientation data, while the second layer exploits the inference of the previous layer and other sensor values to recognize goal-oriented activities of longer time period. Experimental results demonstrate the superior performance of the proposed method over the alternatives in classifying long-term activities as well as short-term activities. The performance improvement is up to 10 % in the experiments, depending on the models compared.

Original languageEnglish
Pages (from-to)1181-1193
Number of pages13
JournalPattern Analysis and Applications
Volume19
Issue number4
DOIs
Publication statusPublished - 2016 Nov 1

Fingerprint

Smartphones
Hidden Markov models
Sensors
Mobile phones
Health care
Magnetic fields
Experiments
Android (operating system)

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

@article{ce0aeb9ec3424c328a7b64ce88313380,
title = "Layered hidden Markov models to recognize activity with built-in sensors on Android smartphone",
abstract = "Recently, activity recognition using built-in sensors in a mobile phone becomes an important issue. It can help us to provide context-aware services: health care, suitable content recommendation for a user’s activity, and user adaptive interface. This paper proposes a layered hidden Markov model (HMM) to recognize both short-term activity and long-term activity in real time. The first layer of HMMs detects short, primitive activities with acceleration, magnetic field, and orientation data, while the second layer exploits the inference of the previous layer and other sensor values to recognize goal-oriented activities of longer time period. Experimental results demonstrate the superior performance of the proposed method over the alternatives in classifying long-term activities as well as short-term activities. The performance improvement is up to 10 {\%} in the experiments, depending on the models compared.",
author = "Lee, {Young Seol} and Sung-Bae Cho",
year = "2016",
month = "11",
day = "1",
doi = "10.1007/s10044-016-0549-8",
language = "English",
volume = "19",
pages = "1181--1193",
journal = "Pattern Analysis and Applications",
issn = "1433-7541",
publisher = "Springer London",
number = "4",

}

Layered hidden Markov models to recognize activity with built-in sensors on Android smartphone. / Lee, Young Seol; Cho, Sung-Bae.

In: Pattern Analysis and Applications, Vol. 19, No. 4, 01.11.2016, p. 1181-1193.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Layered hidden Markov models to recognize activity with built-in sensors on Android smartphone

AU - Lee, Young Seol

AU - Cho, Sung-Bae

PY - 2016/11/1

Y1 - 2016/11/1

N2 - Recently, activity recognition using built-in sensors in a mobile phone becomes an important issue. It can help us to provide context-aware services: health care, suitable content recommendation for a user’s activity, and user adaptive interface. This paper proposes a layered hidden Markov model (HMM) to recognize both short-term activity and long-term activity in real time. The first layer of HMMs detects short, primitive activities with acceleration, magnetic field, and orientation data, while the second layer exploits the inference of the previous layer and other sensor values to recognize goal-oriented activities of longer time period. Experimental results demonstrate the superior performance of the proposed method over the alternatives in classifying long-term activities as well as short-term activities. The performance improvement is up to 10 % in the experiments, depending on the models compared.

AB - Recently, activity recognition using built-in sensors in a mobile phone becomes an important issue. It can help us to provide context-aware services: health care, suitable content recommendation for a user’s activity, and user adaptive interface. This paper proposes a layered hidden Markov model (HMM) to recognize both short-term activity and long-term activity in real time. The first layer of HMMs detects short, primitive activities with acceleration, magnetic field, and orientation data, while the second layer exploits the inference of the previous layer and other sensor values to recognize goal-oriented activities of longer time period. Experimental results demonstrate the superior performance of the proposed method over the alternatives in classifying long-term activities as well as short-term activities. The performance improvement is up to 10 % in the experiments, depending on the models compared.

UR - http://www.scopus.com/inward/record.url?scp=84964344286&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964344286&partnerID=8YFLogxK

U2 - 10.1007/s10044-016-0549-8

DO - 10.1007/s10044-016-0549-8

M3 - Article

AN - SCOPUS:84964344286

VL - 19

SP - 1181

EP - 1193

JO - Pattern Analysis and Applications

JF - Pattern Analysis and Applications

SN - 1433-7541

IS - 4

ER -