Sleep Onset Period Detection Using Slow Eyelid Movement (SEM) Through Eye Aspect Ratio with Electroencephalogram (EEG)

Abstract

This study presents the development of sleep onset period detection using SEM through eye aspect ratio with EEG. The researchers made use of a camera module, Neurosky Mindwave headset and a microcontroller coupled up with an improvised alarm system composed of a buzzer and vibration motors, to detect drowsiness of a subject and to alert the same. Raspberry Pi Camera Module was utilized for eyelid movement detection, Neurosky Mindwave headset for brain wave monitoring and a microcontroller to manage and activate the alarm system of the device. The results of the study showed that the integration of eyelid movement and electroencephalogram provides a more accurate method of determining sleep onset period compared to previous studies. The integrated SEM and EEG parameters provided 97.5% accuracy. This research will greatly benefit the safety of the drivers. Also, this will be beneficial to companies which require its employees to have a high level of alertness as demanded by certain occupations.