Computer Control with Face and Eye Movements Using Deep Learning and Image Processing Methods

Today's computing is one of the basic needs every human being. Many actions are done with mouse in use computer. Individuals physical disabilities, paralysis from neck down, or ALS patients who have difficulty making contact computer having using computers. In this study, an artificial intelligence-assisted system has been developed for these individuals, where they can control head and eye movements. system, facial movements eyes detected real-time through library Haar Cascade, Dlib, Open CV images acquired camera. When Cascade used to detect face region, Dlib acquire right left region image. These areas provided as introduction CNN model, which trained 2874 data (https://github.com/iparaskev/simple-blink-detector), it determined that closed open. The model 1500 on a public image dataset representing open 1374 closed-eye conditions. opened state allows click close open, click. addition, location motion. hybrid combination different methods tested users. According test results, was observed correctly identified eyes, classifying blink event both correctly. However, there slowness modeling movement poor fit movement. next study will focus issue improve by fine-tuning many people.