Abstract
The primary goal of this thesis is to collect and compare the touch parameters of finger movements on touch devices and build personal profile with good-recognition parameters to indicate the touch characteristics of individual users using the touch devices. In order to study the possibility of implementation of touch-style identification for touch devices, this work mainly focuses on finding and testing the possible touch parameters which could be used to compose a profile to verify the users.
A full test with an developed anroid application on tablet was performed by 20 subjects to collect touch information, including location of finger points, finger pressure force and speed of finger movements. Statistical analysis was applied on each dataset of the users. The finding has shown that each user can be identified by the discriminative information of finger movements on the touch screen. The results show huge difference in mean, standard deviation and skewness for the dataset of each user giving a reason to hope the implementation of finger movements based on biometric authentication for touch devices. Hopefully, the result of this project will be valuable for further research of implementation of biometric authentication on touch devices based on the finger movements.