This program can be used to detect drowsiness for a given person. In this program we check how long a person's eyes have been closed for. If the eyes have been closed for a long period i.e. beyond a certain threshold value, the program will alert the user by playing an alarm sound.
IMPORTANT
Download shape_predictor_68_face_landmarks.dat.bz2
from Shape Predictor 68 features
Extract the file in the project folder using
bzip2 -dk shape_predictor_68_face_landmarks.dat.bz2
numpy==1.15.2
dlib==19.16.0
pygame==1.9.4
imutils==0.5.1
opencv_python==3.4.3.18
scipy==1.1.0
Use pip install -r requirements.txt
to install the given requirements.
Put your file to be detected in images folder with name test.jpeg or change the file path in Line : 14 face_and_eye_detector_single_image.py
to your image file.
Run script using:
python face_and_eye_detector_single_image.py
Run script using:
python face_and_eye_detector_webcam_video.py
Run script using:
python drowsiness_detect.py
The algorithm for Eye Aspect Ratio was taken from pyimagesearch.com blog, by Adrian RoseBrock.