Summary
the 2014 International Symposium on Nonlinear Theory and its Applications
2014
Session Number:D1L-B
Session:
Number:D1L-B2
The footstep classification using continuous DP matching and threshold
Hiroshi Yasukawa, Akitoshi Itai,
pp.664-667
Publication Date:2014/9/14
Online ISSN:2188-5079
DOI:10.34385/proc.46.D1L-B2
PDF download (188.1KB)
Summary:
Human footsteps include various characteristics determined by the gait, the footwear and the floor. In these years, the personal identification using footstep has been investigated. The footstep identification method based on continuous dynamic programing (DP) matching and short term Fourier transform (STFT) is proposed[1]. The conventional method shows the recognition rate of 95.1% for 10 subjects, however, all the footstep is classified as the registered footstep. This means that the nonregistered subject, whose footsteps not used as the reference data, is recognized as one of the reference data. In this paper, we propose the footstep classification using a continuous DP matching and threshold to reject a nonregistered pedestrian.