Gait recognition using linear time normalization

被引：47

作者：

Boulgouris, NV ^{[1
]}

Plataniotis, KN

Hatzinakos, D

机构：

[1] Kings Coll London, Dept Elect Engn, Div Engn, London WC2R 2LS, England

[2] Univ Toronto, Edward Rogers Sr Dept Elect & Comp Engn, Toronto, ON, Canada

来源：

PATTERN RECOGNITION | 2006年 / 39卷 / 05期

关键词：

gait; angular analysis; time normalization; recognition; verification;

D O I：

10.1016/j.patcog.2005.10.013

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present a novel system for gait recognition. Identity recognition and verification are based on the matching of linearly time-normalized gait walking cycles. A novel feature extraction process is also proposed for the transformation of human silhouettes into low-dimensional feature vectors consisting of average pixel distances from the center of the silhouette. By using the best-performing of the proposed methodologies, improvements of 8-20% in recognition and verification performance are seen in comparison to other known methodologies on the "Gait Challenge" database. (c) 2005 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

引用

页码：969 / 979

页数：11

共 19 条

[1] BenAbdelkader C., 2002, IEEE INT C AUT FAC G
[2] BOULGOURIS NV, 2004, IEEE INT S MULT SIGN
[3] BOULGOURIS NV, 2004, IEEE INT C IM PROC S
[4] COLLINS RT, 2002, IEEE C AUT FAC GEST
[5] Cunado D, 2003, COMPUT VIS IMAGE UND, V90, P1, DOI [10.1016/S1077-3142(03)00008-0, 10.1010/SI077-3142(03)00008-0]
[6] Automatic gait recognition using area-based metrics
Foster, JP
Nixon, MS
Prügel-Bennett, A
[J]. PATTERN RECOGNITION LETTERS, 2003, 24 (14) : 2489 - 2497
[7] GROTHER P, 2002, 7083 NISTIR
[8] KALE A, 2003, THESIS U MARYLAND CO
[9] Lee L., 2002, P 5 IEEE INT C AUT F
[10] LEE L, 2003, INT C COMP VIS

← 1 2 →