Drive Safe - fatigue driving detector

Detects fatigue driving using OpenCV, dlib (face recognition and facial landmark detection), imutils (image processing).

Supported fatigue symptoms:

• Slow blinking or struggles to keep eyes open
• Nodding off or struggles to keep head up
• Yawning

Installation

• Python 3.9.10
• CMake 3.22.1
• Install requirements: pip install -r requirements.txt

Usage

python monitor.py
    -p / --face_landmark_predictor [path to shape_predictor_68_face_landmarks.dat]
    -v / --verbose [turns on verbose mode]
    -i / --input [optional path to input video file]

Note: By providing an optional input video, the app will run detection on the provided video instead of a live camera feed.

GUI usage

Press 'p' to pause or continue. Press 'q' to exit the app.

---

Credit

• Real-Time Eye Blink Detection using Facial Landmarks by Tereza Soukupová and Jan Cech
• 68-point facial landmarks model by Davis E. King.