Presentation 2011-07-29
Human Gait Recognition using 2 Frequency CW Radar
Kouki FUKAMACHI, Takayuki INABA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In earlier study, recognition of human walking using a radar have been investigated. Conventionally, there are available features to separate human and other objects, which are (1)Mean Velocity, (2)Step Frequency and (3)Appendage Ratio. This paper presents a new feature that is (4)Normalized Power. Normalized Power uses estimated distance available from 2 Frequency CW Radar. In purpose of evaluating effectiveness of new feature, their features are extracted from measured data which is obtained by human, bicycle and dog. After that, our approach is to apply Fisher's linear discnminant and Support Vector Machine to extracted features. It has been showed that presented feature is available.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Radar / 2 frequency CW radar / feature extraction / Support Vector Machine
Paper # SANE2011-51
Date of Issue

Conference Information
Committee SANE
Conference Date 2011/7/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Space, Aeronautical and Navigational Electronics (SANE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Human Gait Recognition using 2 Frequency CW Radar
Sub Title (in English)
Keyword(1) Radar
Keyword(2) 2 frequency CW radar
Keyword(3) feature extraction
Keyword(4) Support Vector Machine
1st Author's Name Kouki FUKAMACHI
1st Author's Affiliation Graduate School of Electro-Communications, The University of Electro-Communications()
2nd Author's Name Takayuki INABA
2nd Author's Affiliation Graduate School of Electro-Communications, The University of Electro-Communications
Date 2011-07-29
Paper # SANE2011-51
Volume (vol) vol.111
Number (no) 166
Page pp.pp.-
#Pages 6
Date of Issue