Dynamic FingerPrint based on human motion and posture

Author

Hesami, Amir ; Naghdy, Fazel ; Stirling, David

Author_Institution

Comput. & Telecommun. Eng., Univ. of Wollongong, Wollongong, NSW, Australia

fYear

2013

fDate

9-12 July 2013

Firstpage

716

Lastpage

721

Abstract

The feasibility of generating a Dynamic FingerPrint (DFP) for an individual is explored. DFP is a unique signature generated based on a combination of body part movements. The body movements are obtained using a sensor suit recording inertial signals that are subsequently modeled on a humanoid frame with 23 degrees of freedom (DOF). Measured signals include position, velocity, acceleration, orientation, angular velocity and angular acceleration. DTW (Dynamic Time Warping) is used to classify the individual´s identity. The approach is described and the characteristics of the algorithms are presented. It is anticipated that these approaches will have applications in surveillance and security, medical science and animation modeling. Classification results show an accuracy rate of 100% for the 10 subjects studied during validation.

Keywords

data visualisation; fingerprint identification; gait analysis; DFP; DTW; angular acceleration; angular velocity; animation modeling; body part movements; dynamic fingerprint; dynamic time warping; human motion; human posture; humanoid frame; individual identity classification; inertial signal recording; medical science; security; sensor suit; signal measurement; surveillance; unique signature; Buildings; Dynamics; Feature extraction; Fingers; Foot; Joints; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Advanced Intelligent Mechatronics (AIM), 2013 IEEE/ASME International Conference on

Conference_Location

Wollongong, NSW

ISSN

2159-6247

Print_ISBN

978-1-4673-5319-9

Type

conf

DOI

10.1109/AIM.2013.6584177

Filename

6584177