Driver's Drowsiness Detection and Prevention System

The project creates a computer vision application that makes use of facial landmarks, Python, and OpenCV to detect eye closure in video streams from a camera. The Eye Aspect Ratio (EAR) is a far more elegant solution that uses a very basic calculation based on the ratio of distances between facial landmarks of the eyes to detect eye closure, in contrast to traditional image processing methods. This eye blink detection technique is quick, effective, and simple to use. In a video stream, it detects whether or not a person is closing his eyes. This indicates that there is a relationship between these coordinates' width and height. When the eye is open, the eye aspect ratio remains relatively constant; however, it quickly returns to zero during blinking or closing of the eye. In our project, the driver will receive an alert from the system if the EAR ratio drops significantly for a predetermined amount of time, say 5000 ms. The alerting signal stays on for at least ten seconds, allowing the driver to wake up and prepare to steady the car.