Misc.

Sep 7, 2009

Quantitative Classification of Pedaling Skill of Power-Assisted Bicycles Using Discriminant Analysis based on Mahalanobis Distance

IEICE technical report
  • INADA Hiroaki
  • ,
  • KIRYU Tohru
  • ,
  • MAEDA Yoshinobu
  • ,
  • HAYASHI Toyohiko

Volume
109
Number
194
First page
1
Last page
6
Language
Japanese
Publishing type
Publisher
The Institute of Electronics, Information and Communication Engineers

In this report, we evaluated pedaling skills from muscle activation patterns of both vastus lateralis of agonist and biceps femoris of antagonist, to classify them quantitatively. To this end, four parameters to classify muscle activation were prepared by means of calculating cross-correlation coefficients between the above two muscles. In the four-parameter space, we classified the pedaling skills into four groups in terms of the discriminant analysis based on Mahalanobis distances. As a result, the classification method presented in this report corresponded about 70.6% accuracy with a qualitative classification based on the human subjectivity. Furthermore, the groups classified by this method were statistically more significant than the qualitative classification.

Link information
CiNii Articles
http://ci.nii.ac.jp/naid/110007387366
CiNii Books
http://ci.nii.ac.jp/ncid/AN1001320X
URL
http://id.ndl.go.jp/bib/10390264
ID information
  • ISSN : 0913-5685
  • CiNii Articles ID : 110007387366
  • CiNii Books ID : AN1001320X

Export
BibTeX RIS