IMPROVING EYE MOVEMENT BIOMETRICS USING REMOTE REGISTRATION OF EYE BLINKING PATTERNS

In this paper, the biometric potential of eye movement and eye blinking for human recognition task is investigated. These modalities might be useful for specific biometric applications like driver authentication for law enforcement. For this purpose, a database of 22 subjects was build where eye movements and blinks were recorded using Gazepoint GP3 while users were watching real driving sessions. Eye movement features were extracted from eye fixations and saccades separately. Eye blinking features include the blink pattern, its speed and acceleration patterns, and time delineation features. Evaluation of each modality was investigated first, then, both modalities are combined in a multi-modal setup for performance improvement. Although the employment of eye movement or eye blinking separately as a biometric trait might not be secure enough, the fusion of both traits achieves higher levels of identification which are comparable to that of other conventional biometric traits like fingerprint.