Wide-Awake: A drowsy Driver Detection System.

Abstract

The number of accidents caused has increased rapidly over the past few years. According to recent researches, many of these accidents are caused by drowsy driving. At present, in response to this situation, certain types of drowsy driver identifying systems have been introduced to the world. However, due to some of the technical loopholes of such existing systems, the functionalities of those systems are questionable. With these considerations in mind, Wide-Awake, The Drowsy Driver Detection System has been developed to answer these shortcomings. Through this system, a drowsy driver will be immediately identified, and actions will be taken to bring the driver back to a normal alert state within seconds. The concepts of computer vision and image processing were mainly used in the development of this Wide-Awake system. The system consists of three components: a neural network-based component, a mobile application, and a web application. The concepts of deep learning, machine learning, python, and Keras were used to develop the core component of the system. The Android framework was used to develop the mobile application with the Google vision API. Finally, Flask micro web framework, OpenCV libraries, HTML, and CSS were used to develop the third component, the Web application. The Wide Awake system plays a vital role as a high quality system that enhances the quality of human life and provides greater security for human lives.