Development of electrooculogram based human computer interface system using deep learning

The patients with diseases that cause severe movement disabilities was noticeably increasing. These made unable to carry out their daily activities or interact external environment. However, the existence of human-computer interfaces (HCI) gave those a new hope be able once again. HCI enabled these communicate environment by recognizing eyes. Eye movements are recorded an electro-oculogram (EOG) through some electrodes put vertically and horizontally on In this paper, EOG vertical horizontal signals were analyzed detect six eye (up, down, right, left, double blinking, center). Three deep learning models namely convolution neural network (CNN), visual geometry group (VGG), inception had been examined filtered signals. experimental results reveal superiority model in providing best average accuracy 96.4%. Accordingly, writing system is presented based detected movements.