Human Computer Interface systems are extending their boundaries in our daily life and become an important subject in biomedical engineering. Electrooculogram signal as an input for such systems stems from the corneo-retinal standing potential, which can be used for monitoring human eye rotation. Higher amplitude, better signal-to-noise ratio, and much easier recording conditions compared to EEG, make it an important input modality for HCI systems. In the present study, a real-time processing and cost effective (<100$) HCI system was designed and developed based on the EOG signals. The required electrodes were embedded in updated eyeglasses for easy electrode placement over the subject’s face. EOG signals were acquired by the subject’s eye movement toward the four middle parts of screen edges of a laptop placed in front of them. The system training for each subject ameliorated system withstanding against the blink and wrinkle artifact. Finally, the required commands for quadcopter navigation (up, down, left and right) generated with 0.6s total delay and 94.8% of system accuracy in detecting the correct eye movements. A real quadcopter navigation experiment based on the standard navigation interface and the developed HCI system represented the system accuracy, robustness and usefulness.

Keywords