A 3-D Hand Gesture Signature Based Biometric Authentication System for Smartphones

Abstract

Most of the smart phones are equipped with user authentication mechanism such as entering a 4-digit password or drawing a simple pattern. These system are easy to be hacked and more importantly, they authenticate the password or the pattern, not the real user. In this paper, we describe the design and implementation of a 3-D hand gesture signature (HGS) based biometric authentication system. We take advantage of the on-phone accelerometer to capture the 3-D acceleration information when user holds the phone and makes a gesture in order to gain access to the phone. This data will go through a sequence of signal processing, namely data smooth, gesture spotting, sequence alignment and interpolation. Then the processed data will be compared with the genuine user's registered pattern to determine whether access to the phone should be granted to the user. We have implemented the proposed 3-D HGS authentication system on real smart phones of different brands and recruited volunteers to perform on-phone experiments to test the performance of the system. The authentication process is tested for a total of 76,520 times by 19 users. The results show very low false acceptance (0.28%) and false rejection rates (5.13%). The system is user friendly (the acceptance/rejection decision is made instantaneously) and does not require any addition hardware on the phone. We further export the real gesture samples from the phones to a desktop PC, where we implement two existing gesture based authentication systems similar to ours. The simulation results reveal that our system has the best authentication accuracy.