Drivers' Drowsiness Detection
A technology of automotive safety with the aim to prevent accidents caused by drivers' drowsiness
Did you know?
Euros
Estimated cost of a road traffic fatality
Euros
Annual economic impact in EU as a result of traffic accidents
Traffic accidents
Estimated to be related with driver's fatigue
Driver's Drowsiness detection
Technology behind our App
There are several functions used to detect fatigue through images, described below:
EAR (Ratio of the eye)
This metric allows to detect blinks of this eye when the value goes lower than the median. The length and frequency of blinks can help detect symptoms of fatigue.
.gif)
MAR (Mouth Aspect Ratio)
This metric gives the ratio of mouth opening and it is used to identify yawns with 97% accuracy. Yawn is another symptom of drowsiness used by our program.
Methodology
Image Capture
Take color images or infrared camera images, and analyze videos in real time
.gif)
Eye & Mouth Segmentation
Face Detection of the Subject Using the OpenCV Library
Image Processing
Image processing, resizing, and grayscale transformation
Algorithm
Analysis of the images by passing them to trained networks of both eyes and mouth
.gif)
Drowsiness Prediction
Weighting of the estimated values to conclude the level of drowsiness of the driver
Alarm
Once the system has detected a level of drowsiness of the driver, it activates an alarm to prevent accidents
MATRIX SWOT
The Driver Drowsiness Detection project uses programming algorithms to improve road safety by identifying signs of driver fatigue. A SWOT analysis is a valuable tool for evaluating the project's internal strengths and weaknesses, as well as external opportunities and threats.
.png){kind=small}
.png){kind=small}
