Abstract

Accelerometer based activity recognition systems have typically focused on recognizing simple ambulatory activities of daily life, such as walking, sitting, standing, climbing stairs, etc. In this work, we developed and evaluated algorithms for detecting and recognizing short duration hand movements (lift to mouth, scoop, stir, pour, unscrew cap). These actions are a part of the larger and complex Instrumental Activities of Daily Life (IADL) making a drink and drinking. We collected data using small wireless tri-axial accelerometers worn simultaneously on different parts of the hand. Acceleration data for training was collected from 5 subjects, who also performed the two IADLs without being given specific instructions on how to complete them. Feature vectors (mean, variance, correlation, spectral entropy and spectral energy) were calculated and tested on three classifiers (AdaBoost, HMM, k-NN). AdaBoost showed the best performance, with an overall accuracy of 86% for detecting each of these hand actions. The results show that although some actions are recognized well with the generalized classifer trained on the subject-independent data, other actions require some amount of subject-specific training. We also observed an improvement in the performance of the system when multiple accelerometers placed on the right hand were used.

Original languageEnglish (US)
Pages (from-to)143-155
Number of pages13
JournalJournal of Ambient Intelligence and Smart Environments
Volume1
Issue number2
DOIs
StatePublished - 2009

Fingerprint

Accelerometers
Adaptive boosting
Stairs
Classifiers
Entropy

Keywords

  • Accelerometers
  • Adaboost
  • Classification
  • Hand movements

ASJC Scopus subject areas

  • Software

Cite this

Recognition of hand movements using wearable accelerometers. / Krishnan, Narayanan C.; Juillard, Colin; Colbry, Dirk; Panchanathan, Sethuraman.

In: Journal of Ambient Intelligence and Smart Environments, Vol. 1, No. 2, 2009, p. 143-155.

Research output: Contribution to journalArticle

Krishnan, Narayanan C. ; Juillard, Colin ; Colbry, Dirk ; Panchanathan, Sethuraman. / Recognition of hand movements using wearable accelerometers. In: Journal of Ambient Intelligence and Smart Environments. 2009 ; Vol. 1, No. 2. pp. 143-155.
@article{858ea49cb167403db6cdff637b31ca46,
title = "Recognition of hand movements using wearable accelerometers",
abstract = "Accelerometer based activity recognition systems have typically focused on recognizing simple ambulatory activities of daily life, such as walking, sitting, standing, climbing stairs, etc. In this work, we developed and evaluated algorithms for detecting and recognizing short duration hand movements (lift to mouth, scoop, stir, pour, unscrew cap). These actions are a part of the larger and complex Instrumental Activities of Daily Life (IADL) making a drink and drinking. We collected data using small wireless tri-axial accelerometers worn simultaneously on different parts of the hand. Acceleration data for training was collected from 5 subjects, who also performed the two IADLs without being given specific instructions on how to complete them. Feature vectors (mean, variance, correlation, spectral entropy and spectral energy) were calculated and tested on three classifiers (AdaBoost, HMM, k-NN). AdaBoost showed the best performance, with an overall accuracy of 86{\%} for detecting each of these hand actions. The results show that although some actions are recognized well with the generalized classifer trained on the subject-independent data, other actions require some amount of subject-specific training. We also observed an improvement in the performance of the system when multiple accelerometers placed on the right hand were used.",
keywords = "Accelerometers, Adaboost, Classification, Hand movements",
author = "Krishnan, {Narayanan C.} and Colin Juillard and Dirk Colbry and Sethuraman Panchanathan",
year = "2009",
doi = "10.3233/AIS-2009-0019",
language = "English (US)",
volume = "1",
pages = "143--155",
journal = "Journal of Ambient Intelligence and Smart Environments",
issn = "1876-1364",
publisher = "IOS Press",
number = "2",

}

TY - JOUR

T1 - Recognition of hand movements using wearable accelerometers

AU - Krishnan, Narayanan C.

AU - Juillard, Colin

AU - Colbry, Dirk

AU - Panchanathan, Sethuraman

PY - 2009

Y1 - 2009

N2 - Accelerometer based activity recognition systems have typically focused on recognizing simple ambulatory activities of daily life, such as walking, sitting, standing, climbing stairs, etc. In this work, we developed and evaluated algorithms for detecting and recognizing short duration hand movements (lift to mouth, scoop, stir, pour, unscrew cap). These actions are a part of the larger and complex Instrumental Activities of Daily Life (IADL) making a drink and drinking. We collected data using small wireless tri-axial accelerometers worn simultaneously on different parts of the hand. Acceleration data for training was collected from 5 subjects, who also performed the two IADLs without being given specific instructions on how to complete them. Feature vectors (mean, variance, correlation, spectral entropy and spectral energy) were calculated and tested on three classifiers (AdaBoost, HMM, k-NN). AdaBoost showed the best performance, with an overall accuracy of 86% for detecting each of these hand actions. The results show that although some actions are recognized well with the generalized classifer trained on the subject-independent data, other actions require some amount of subject-specific training. We also observed an improvement in the performance of the system when multiple accelerometers placed on the right hand were used.

AB - Accelerometer based activity recognition systems have typically focused on recognizing simple ambulatory activities of daily life, such as walking, sitting, standing, climbing stairs, etc. In this work, we developed and evaluated algorithms for detecting and recognizing short duration hand movements (lift to mouth, scoop, stir, pour, unscrew cap). These actions are a part of the larger and complex Instrumental Activities of Daily Life (IADL) making a drink and drinking. We collected data using small wireless tri-axial accelerometers worn simultaneously on different parts of the hand. Acceleration data for training was collected from 5 subjects, who also performed the two IADLs without being given specific instructions on how to complete them. Feature vectors (mean, variance, correlation, spectral entropy and spectral energy) were calculated and tested on three classifiers (AdaBoost, HMM, k-NN). AdaBoost showed the best performance, with an overall accuracy of 86% for detecting each of these hand actions. The results show that although some actions are recognized well with the generalized classifer trained on the subject-independent data, other actions require some amount of subject-specific training. We also observed an improvement in the performance of the system when multiple accelerometers placed on the right hand were used.

KW - Accelerometers

KW - Adaboost

KW - Classification

KW - Hand movements

UR - http://www.scopus.com/inward/record.url?scp=78651574474&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651574474&partnerID=8YFLogxK

U2 - 10.3233/AIS-2009-0019

DO - 10.3233/AIS-2009-0019

M3 - Article

AN - SCOPUS:78651574474

VL - 1

SP - 143

EP - 155

JO - Journal of Ambient Intelligence and Smart Environments

JF - Journal of Ambient Intelligence and Smart Environments

SN - 1876-1364

IS - 2

ER -