Driver Fatigue Detection | OpenCV

This repo can detect and track user's eyes and mouth and alert when the user is feeling drowsy using video processing techniques.

Applications

The AAA says that 20% of all fatal accidents in the USA are due to drowsiness! We can only imagine what the stats are like for India which has a higher road accident rate. This system can be used by riders who tend to drive for a longer period of time to warn them and and prevent accidents.

Code Requirements

The example code is in Python (version 2.7 or higher will work). The code is tested under Ubuntu 18.04

Dependencies

• import cv2
• import immutils
• import dlib
• import scipy

Description

A computer vision system that can automatically detect driver drowsiness by tracking closing of eyes and yawing in a real-time video stream and then alert if the driver appears to be drowsy.

Execution

To run the code, perform the below command

python3 driverFatigue_v2.py

or

python driverFatigue_v2.py