Single-accelerometer-based daily physical activity classification

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

136 Citations (Scopus)

Abstract

In this study, a single tri-axial accelerometer placed on the waist was used to record the acceleration data for human physical activity classification. The data collection involved 24 subjects performing daily real-life activities in a naturalistic environment without researchers' intervention. For the purpose of assessing customers' daily energy expenditure, walking, running, cycling, driving, and sports were chosen as target activities for classification. This study compared a Bayesian classification with that of a Decision Tree based approach. A Bayes classifier has the advantage to be more extensible, requiring little effort in classifier retraining and software update upon further expansion or modification of the target activities. Principal components analysis was applied to remove the correlation among features and to reduce the feature vector dimension. Experiments using leave-one-subject-out and 10-fold cross validation protocols revealed a classification accuracy of ~80%, which was comparable with that obtained by a Decision Tree classifier.
LanguageEnglish
Title of host publicationProceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages6107-6110
ISBN (Print)978-1-4244-3296-7
DOIs
StatePublished - 2009

Fingerprint

Accelerometers
Classifiers
Decision trees
Sports
Principal component analysis
Experiments

Cite this

Long, X., Yin, B., & Aarts, R. M. (2009). Single-accelerometer-based daily physical activity classification. In Proceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota (pp. 6107-6110). Piscataway: Institute of Electrical and Electronics Engineers. DOI: 10.1109/IEMBS.2009.5334925
Long, Xi ; Yin, B. ; Aarts, R.M./ Single-accelerometer-based daily physical activity classification. Proceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota. Piscataway : Institute of Electrical and Electronics Engineers, 2009. pp. 6107-6110
@inproceedings{0caa0afa1c934014a53e76e407563afd,
title = "Single-accelerometer-based daily physical activity classification",
abstract = "In this study, a single tri-axial accelerometer placed on the waist was used to record the acceleration data for human physical activity classification. The data collection involved 24 subjects performing daily real-life activities in a naturalistic environment without researchers' intervention. For the purpose of assessing customers' daily energy expenditure, walking, running, cycling, driving, and sports were chosen as target activities for classification. This study compared a Bayesian classification with that of a Decision Tree based approach. A Bayes classifier has the advantage to be more extensible, requiring little effort in classifier retraining and software update upon further expansion or modification of the target activities. Principal components analysis was applied to remove the correlation among features and to reduce the feature vector dimension. Experiments using leave-one-subject-out and 10-fold cross validation protocols revealed a classification accuracy of ~80{\%}, which was comparable with that obtained by a Decision Tree classifier.",
author = "Xi Long and B. Yin and R.M. Aarts",
year = "2009",
doi = "10.1109/IEMBS.2009.5334925",
language = "English",
isbn = "978-1-4244-3296-7",
pages = "6107--6110",
booktitle = "Proceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Long, X, Yin, B & Aarts, RM 2009, Single-accelerometer-based daily physical activity classification. in Proceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota. Institute of Electrical and Electronics Engineers, Piscataway, pp. 6107-6110. DOI: 10.1109/IEMBS.2009.5334925

Single-accelerometer-based daily physical activity classification. / Long, Xi; Yin, B.; Aarts, R.M.

Proceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota. Piscataway : Institute of Electrical and Electronics Engineers, 2009. p. 6107-6110.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Single-accelerometer-based daily physical activity classification

AU - Long,Xi

AU - Yin,B.

AU - Aarts,R.M.

PY - 2009

Y1 - 2009

N2 - In this study, a single tri-axial accelerometer placed on the waist was used to record the acceleration data for human physical activity classification. The data collection involved 24 subjects performing daily real-life activities in a naturalistic environment without researchers' intervention. For the purpose of assessing customers' daily energy expenditure, walking, running, cycling, driving, and sports were chosen as target activities for classification. This study compared a Bayesian classification with that of a Decision Tree based approach. A Bayes classifier has the advantage to be more extensible, requiring little effort in classifier retraining and software update upon further expansion or modification of the target activities. Principal components analysis was applied to remove the correlation among features and to reduce the feature vector dimension. Experiments using leave-one-subject-out and 10-fold cross validation protocols revealed a classification accuracy of ~80%, which was comparable with that obtained by a Decision Tree classifier.

AB - In this study, a single tri-axial accelerometer placed on the waist was used to record the acceleration data for human physical activity classification. The data collection involved 24 subjects performing daily real-life activities in a naturalistic environment without researchers' intervention. For the purpose of assessing customers' daily energy expenditure, walking, running, cycling, driving, and sports were chosen as target activities for classification. This study compared a Bayesian classification with that of a Decision Tree based approach. A Bayes classifier has the advantage to be more extensible, requiring little effort in classifier retraining and software update upon further expansion or modification of the target activities. Principal components analysis was applied to remove the correlation among features and to reduce the feature vector dimension. Experiments using leave-one-subject-out and 10-fold cross validation protocols revealed a classification accuracy of ~80%, which was comparable with that obtained by a Decision Tree classifier.

U2 - 10.1109/IEMBS.2009.5334925

DO - 10.1109/IEMBS.2009.5334925

M3 - Conference contribution

SN - 978-1-4244-3296-7

SP - 6107

EP - 6110

BT - Proceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Long X, Yin B, Aarts RM. Single-accelerometer-based daily physical activity classification. In Proceedings of the EMBC' 09; 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2-6 September 2009, Minneapolis, Minnesota. Piscataway: Institute of Electrical and Electronics Engineers. 2009. p. 6107-6110. Available from, DOI: 10.1109/IEMBS.2009.5334925