People Identification Using Gait Via Floor Pressure Sensing and Analysis

This paper presents an approach to people identification using gait based on floor pressure data. By using a large area high resolution pressure sensing floor, we were able to obtain 3D trajectories of the center of foot pressures over a footstep which contain both the 1D pressure profile and 2D position trajectories of the COP. Based on the 3D COP trajectories a set of features are then extracted and used for people identification together with other features such as stride length and cadence. The Fisher linear discriminant is used as the classifier. Encouraging results have been obtained using the proposed method with an average recognition rate of 94% and false alarm rate of 3% using pair-wise footstep data from 10 subjects.