Cover Image

Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin)

This study aimed to assess the motion accuracy of Baduanjin and recognise the motions of Baduanjin based on sequence-based methods. Motion data of Baduanjin were measured by the inertial sensor measurement system (IMU). Fifty-four participants were recruited to capture motion data. Based on the moti...

Full description

Saved in:
Bibliographic Details
Main Authors: Li, Hai, Khoo, Selina Phaik Lin, Yap, Hwa Jen
Format: Article
Published: MDPI 2022
Subjects:
GE Environmental Sciences
GF Human ecology. Anthropogeography
RA0421 Public health. Hygiene. Preventive Medicine
Online Access:http://eprints.um.edu.my/33422/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.um.eprints.33422
record_format eprints
spelling my.um.eprints.334222022-08-17T04:56:54Z http://eprints.um.edu.my/33422/ Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin) Li, Hai Khoo, Selina Phaik Lin Yap, Hwa Jen GE Environmental Sciences GF Human ecology. Anthropogeography RA0421 Public health. Hygiene. Preventive Medicine This study aimed to assess the motion accuracy of Baduanjin and recognise the motions of Baduanjin based on sequence-based methods. Motion data of Baduanjin were measured by the inertial sensor measurement system (IMU). Fifty-four participants were recruited to capture motion data. Based on the motion data, various sequence-based methods, namely dynamic time warping (DTW) combined with classifiers, hidden Markov model (HMM), and recurrent neural networks (RNNs), were applied to assess motion accuracy and recognise the motions of Baduanjin. To assess motion accuracy, the scores for motion accuracies from teachers were used as the standard to train the models on the different sequence-based methods. The effectiveness of Baduanjin motion recognition with different sequence-based methods was verified. Among the methods, DTW + k-NN had the highest average accuracy (83.03%) and shortest average processing time (3.810 s) during assessing. In terms of motion reorganisation, three methods (DTW + k-NN, DTW + SVM, and HMM) had the highest accuracies (over 99%), which were not significantly different from each other. However, the processing time of DTW + k-NN was the shortest (3.823 s) compared to the other two methods. The results show that the motions of Baduanjin could be recognised, and the accuracy can be assessed through an appropriate sequence-based method with the motion data captured by IMU. MDPI 2022-02 Article PeerReviewed Li, Hai and Khoo, Selina Phaik Lin and Yap, Hwa Jen (2022) Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin). International Journal of Environmental Research and Public Health, 19 (3). ISSN 1660-4601, DOI https://doi.org/10.3390/ijerph19031744 <https://doi.org/10.3390/ijerph19031744>. 10.3390/ijerph19031744
institution Universiti Malaya
building UM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaya
content_source UM Research Repository
url_provider http://eprints.um.edu.my/
topic GE Environmental Sciences
GF Human ecology. Anthropogeography
RA0421 Public health. Hygiene. Preventive Medicine
spellingShingle GE Environmental Sciences
GF Human ecology. Anthropogeography
RA0421 Public health. Hygiene. Preventive Medicine
Li, Hai
Khoo, Selina Phaik Lin
Yap, Hwa Jen
Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin)
description This study aimed to assess the motion accuracy of Baduanjin and recognise the motions of Baduanjin based on sequence-based methods. Motion data of Baduanjin were measured by the inertial sensor measurement system (IMU). Fifty-four participants were recruited to capture motion data. Based on the motion data, various sequence-based methods, namely dynamic time warping (DTW) combined with classifiers, hidden Markov model (HMM), and recurrent neural networks (RNNs), were applied to assess motion accuracy and recognise the motions of Baduanjin. To assess motion accuracy, the scores for motion accuracies from teachers were used as the standard to train the models on the different sequence-based methods. The effectiveness of Baduanjin motion recognition with different sequence-based methods was verified. Among the methods, DTW + k-NN had the highest average accuracy (83.03%) and shortest average processing time (3.810 s) during assessing. In terms of motion reorganisation, three methods (DTW + k-NN, DTW + SVM, and HMM) had the highest accuracies (over 99%), which were not significantly different from each other. However, the processing time of DTW + k-NN was the shortest (3.823 s) compared to the other two methods. The results show that the motions of Baduanjin could be recognised, and the accuracy can be assessed through an appropriate sequence-based method with the motion data captured by IMU.
format Article
author Li, Hai
Khoo, Selina Phaik Lin
Yap, Hwa Jen
author_facet Li, Hai
Khoo, Selina Phaik Lin
Yap, Hwa Jen
author_sort Li, Hai
title Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin)
title_short Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin)
title_full Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin)
title_fullStr Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin)
title_full_unstemmed Implementation of sequence-based classification methods for motion assessment and recognition in a traditional Chinese sport (Baduanjin)
title_sort implementation of sequence-based classification methods for motion assessment and recognition in a traditional chinese sport (baduanjin)
publisher MDPI
publishDate 2022
url http://eprints.um.edu.my/33422/
_version_ 1744649153443528704
score 13.211869

Similar Items