Skip to content

Latest commit

 

History

History
35 lines (30 loc) · 1.78 KB

README.md

File metadata and controls

35 lines (30 loc) · 1.78 KB
Raw

Name

Drowsiness Detector Python model to detect the drowsiness of the Driver.

Description

This model has following key steps: - It detects the human face for the live vedio by processing each frame. - Then with the help of dlib library of python and a dataset file we get the location of points on the face. - After this it calculates the eye aspect ratio to get how much the eye of the person is open. - This process continues till the aspect ratio is smaller than a desired one for particular frames. - On encountering above condition the alert is raised

Prerequisites

Pyhton3 installed with following libraries on your system.

Libraries needed

  1. scipy to compute the Euclidean distance between facial landmark points in the eye aspect ratio.
    • Install with following command pip install scipy
  2. imutils for vedio and image processing
    • Install with following command pip install imutils
  3. thread for playing alert without intrupting or pausing the script
    • Install with following command pip install threaded
  4. playsound for playing alert
    • Install with following command pip install playsound
  5. dlib most important of all for returning the facial points
    • Install with following command pip install dlib and make sure you have cmake installed.
  6. cv2 and numpy for computer vision on images
    • Install with following command pip install opencv-python and pip install numpy

How to run

After the installation of all packages run with the following command python detect_drowsiness.py \--shape-predictor face_shape_predictor.dat \--alarm alert.wav

Disclaimer

This project was by me to detect drowsiness in drivers and raise a appropriate alert so as to minimise road accidents due to drowsiness.