Machine Control via Real Time Eye Detector

As the population ages, the number of people dependent on others who are paralyzed or losing their self-movement is increasing. This paper is focusing on the development of a smart robot based on wireless vision control which is designed for physically challenged individuals. The research employs a human pupil movement hands-free control machine for the moderate or severe physically disabled individuals by applying Convolution Neural Networks (CNNs) method to pre-train the model. The dataset contains the annotation information. By using the coordinates of the eye to identify the Iris’ location. This feature is to ensure the alignment of the eyes to identify the dominant eyes for Strabismus users. The result of this paper works as per expected with a preferable accuracy which fulfilled the objectives of doing this project. The delay time of the transmission data is negligible. Therefore, the synchronizing between the prototype and the eyeball movement of the user’s intention is nearly perfect in which the eyeball moves upward, the robot will go forward, and so on. In short, this project is to have a contribution to society in a small way by presenting an idea for a system that can truly improve the lives of physically disabled people around the world.