EOG-based visual navigation interface development

This paper presents an EOG-based (Electrooculography) interface for Human Computer Interface (HCI) purposes. The solution enables the filtering of the recorded signals and identification of characteristic peak amplitudes associated with eye saccades, blinks or winks by using a classifier based on a set of fuzzy logic rules and a deterministic finite automaton. The identified eye saccades were assigned to six low-level commands for navigation purposes. An experiment study was conducted in order to check the accuracy and the performances of the proposed interface compared with three traditional input control interfaces. Experimental results show that the developed interface has good performance and can be used for online communication and control in EOG-based HCI systems or even for first-person navigation metaphors in games industry.

► A classifier based fuzzy logic and a deterministic finite automaton was used.
► Eye saccades were assigned to six low-level commands for navigation purposes.
► The proposed interface was used for a navigation task in a virtual environment.
► The EOG interface has comparable performances with the classical desktop interfaces in terms of number of commands used.
► The system accurately detects the control signals defined by eye saccades.