Driver drowsiness detection
- 0 Collaborators
A computer vision system made with the help of OpenCV that can automatically detect driver drowsiness in a real-time video stream and then play an alarm if the driver appears to be drowsyion, can be used by riders who tends to drive the vehicle for a longer period of time that may lead to accidents ...learn more
Project status: Concept
Overview / Usage
This can be used by riders who tends to drive the vehicle for a longer period of time that may lead to accidents
Methodology / Approach
A computer vision system made with the help of opencv that can automatically detect driver drowsiness in a real-time video stream and then play an alarm if the driver appears to be drowsy.
● We utilized pre-trained a pre trained frontal face detector from Dlib’s library which is based on a modification to the Histogram of Oriented Gradients in combination with Linear SVM for classification.
● The pre-trained facial landmark detector inside the dlib library is used to estimate the location of 68 (x, y)-coordinates that map to facial structures on the face. The 68 landmark output is shown in the figure below. However, we utilized the 70 landmark model.
● We then calculate the aspect ratio to check whether eyes are opened or closed.
● The eye is open if Eye Aspect ratio is greater than the threshold. (Around 0.3)
● A blink is supposed to last 200-300 milliseconds.
● A drowsy blink would last for 800-900 ms.
Technologies Used
Technologies:
OpenCV
Libraries:
cv2 immutils dlib scipy playsound queue time sys
Repository
https://github.com/shubham769/Driver-drowsiness-detection