Exploring how user routine affects the recognition performance of a lock pattern

Abstract

To protect an Android smartphone against attackers, a lock pattern can be used. Nevertheless, shoulder-surfing and smudge attacks can be used to get access despite of this protection. To combat these attacks, biometric recognition can be added to the lock pattern, such that the lock-pattern application keeps track of the way users draw the pattern. This research explores how users change the way they draw lock patterns over time and its effect on the recognition performance of the pattern. A lock-pattern dataset has been collected and a classifier is proposed. In this research the best result was obtained using the x- and y-coordinate as the user's biometrics. Unfortunately, in this paper it is shown that adding biometrics to a lock pattern is only an additional security that provides no guarantee for a secure lock pattern. It is just a small improvement over using a lock pattern without biometric identification.