A dictionary learning based approach for gait classification

To foster diagnosis of gait abnormalities as well as tracking the recovery rate in the course of healing, automated gait classification methods have great added value. Therefore, gait classification based on a dictionary learning approach was developed and tested. With an average classification rate...

Full description

Saved in:
Bibliographic Details
Published in2017 22nd International Conference on Digital Signal Processing (DSP) pp. 1 - 4
Main Authors Poschadel, Nils, Moghaddamnia, Sanam, Alcaraz, Javier Conte, Steinbach, Marc, Peissig, Jurgen
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.08.2017
Subjects
Approximation algorithms
Approximation error
Dictionaries
dictionary learning
gait classification
Image coding
Machine learning
Meters
Sparse coding
Online AccessGet full text

Cover

More Information
Summary:To foster diagnosis of gait abnormalities as well as tracking the recovery rate in the course of healing, automated gait classification methods have great added value. Therefore, gait classification based on a dictionary learning approach was developed and tested. With an average classification rate of about 93%, the proposed method offers great potential to be deployed in support of digital healthcare and therapy. Moreover, by providing efficient data storage as well as low runtime, it is ideal for use in portable diagnostic tools.
ISSN:2165-3577
DOI:10.1109/ICDSP.2017.8096121