People Identification Using Floor Pressure Sensing and Analysis

This paper presents an approach to people identification based on gait using floor pressure data. By using a large-area, high-resolution pressure sensing floor, we are able to obtain both the 1D pressure profile and 2D position trajectories of the centers of pressure (COP) of both feet to form a 3D...

Full description

Saved in:
Bibliographic Details
Published in:IEEE sensors journal 2010-09, Vol.10 (9), p.1447-1460
Main Authors: Qian, Gang, Zhang, Jiqing, Kidane, Assegid
Format: Article
Language:English
Subjects:
Access control
Algorithms
Application software
Biometrics
Detection
False alarms
Feature extraction
Feet
Foot
gait recognition
Human computer interaction
Legged locomotion
pressure analysis
Pressure measurement
Recognition
Studies
Terrorism
Testing
Three dimensional
Trajectories
Two dimensional
Walking
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents an approach to people identification based on gait using floor pressure data. By using a large-area, high-resolution pressure sensing floor, we are able to obtain both the 1D pressure profile and 2D position trajectories of the centers of pressure (COP) of both feet to form a 3D COP trajectories over a footstep. From the 3D COP trajectories of a pair of footsteps, a set of features are extracted and used together with other features such as the mean pressure and stride length for people identification. The Fisher linear discriminant is used as the classifier. The speed-invariant properties of different features are examined and we have shown that the pressure features have good speed-invariance properties, much better than the stride length feature. The proposed algorithm has been extensively tested using a floor pressure dataset collected from 11 subjects in different walking styles, including varying speed walking and free style walking. Decent people identification results have been obtained using the proposed method with an average recognition rate of 92.3% and false alarm rate of 6.79% using the best performing feature set.
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2010.2045158