Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 8

Driver Drowsiness Detection System

Using Image Processing And IR


sensing Mechanism

By- Gargi Dutta


Farhan Abdul Salam.

Image Processing
Image Processing is a technique used to process
images from live video or to detect and track
different features and objects. This technique can
further be used and programmed for different
purposes.
Here we are trying to detect facial features form a
live video feed with the help of Viola Jones
Algorithm and KLT algorithm.

Steps

Fetching the video from webcam


Grabbing frames
Detecting the eye
Tracking the eye
Fetching eye from rest of the image
Detect blink

ViolaJones
HAAR features

Algortihm
1. Connect web camera to MATLAB
2. Extract images from video frames to get frame size.
3. Convert video frame into grey scale image.
4. Detect feature points in image and mark them.
5. Form Bounding box around the Region Of Interest.
6. Convert Bounding box data in form of coordinate points.
7. Monitor the number of points as entries to check whether detection is done or not.
8. If detection done then
9. keep track of bounding box points using geometric transformation.
10.Run loop that verify the conditions to ensure continuous monitoring.
11.Crop Bounding box and convert into black-n-white image.
12.Extract the black n white image and plot the data into graph.
13.The data in graph monitor the pixels of black portion and white portion in the
bounding box image.
14.When the eye is open the intensity of black pixel will be higher than when the eye is
closed.
15.If the intensity of the black pixels in the image will determine whether the person is
sleepy or not.

OUT PUT

You might also like