Driver Drowsiness Detection System

A real-time drowsiness and yawn detection system built with Python, OpenCV, and dlib to help prevent accidents caused by driver fatigue.

Features

• Eye Drowsiness Detection: Monitors eye aspect ratio (EAR) to detect when a person's eyes are closed for too long
• Yawn Detection: Measures lip distance to identify when a person is yawning
• Real-time Alerts: Visual alerts on screen and audio alerts via system speakers
• Easy-to-use Interface: Simple command-line application with webcam support

Requirements

• Python 3.6+
• OpenCV (cv2)
• dlib
• imutils
• numpy
• scipy
• espeak (for audio alerts)

Installation

1. Clone this repository:

   git clone https://github.com/9keystrokes/Driver-Drowsiness-Detection-System.git
   cd Driver-Drowsiness-Detection-System
   

2. Install the required packages:

   pip install opencv-python dlib imutils numpy scipy
   

3. Download the shape predictor file:

   • Download the 68 facial landmark predictor from dlib's model repository
   • Extract the file and place it in the project directory

4. Install espeak for audio alerts:

   • Windows: Download and install from espeak official website
   • Linux: sudo apt-get install espeak
   • macOS: brew install espeak

Usage

Run the program with:

python drowsiness_yawn.py --webcam 0

Where:

• --webcam (or -w): specifies the webcam index (default is 0 for the built-in webcam)

How It Works

1. Face Detection: Uses a Haar Cascade classifier to detect faces in the video stream
2. Facial Landmarks: Uses dlib's 68-point facial landmark detector to identify key points on the face