This document describes a driver drowsiness detection system that uses computer vision and machine learning. The system captures video of the driver's face in real-time using a webcam. It detects facial landmarks and calculates the eye aspect ratio to determine if the driver's eyes are open or closed. If the eye aspect ratio indicates the eyes are closed for too long, an alert is generated to warn the driver. The system is designed to help prevent accidents by detecting drowsiness early through an affordable, real-time analysis of the driver's face and eyes.
5. Abstract
• The detection system can detect sleeping individuals by using a web camera to obtain real-time continuous images.
• The developed system has detects the eye, opening and closing conditions. It varies the condition of eye the distance
between two consecutive dips of the light intensity.
• It will detect and localize whether the eyes are open or close based on the real time video stream of drivers. Through the
live video streaming, a frame is extracted for image processing.
• Images are captured typically at a fix frame rate of 20fps based on brightness and camera quality.
• The OpenCV and Dlib support the android platform to detect the faces from the live frame and predict the driver
drowsiness.
• It is also affordable as it can process incoming video streams in real-time and does not need any expensive hardware
support.
6. Modules
• Live Face Detection
• Facial Points Collection
• Eye Aspect Ratio
• Detection
7. Modules
• Live Face Detection : OpenCV library is used to capture the real time video of the users where the frames are collected and
haar-cascade is an algorithm that can detect objects in images, irrespective of their scale in image and location.
• Facial Points Collection : Facial landmarks are used to localize and represent important regions of the face, such as: Mouth,
Eyes, Eyebrows, Nose, Jawline etc.. Nearly 68 facial points are captured.
• Eye Aspect Ratio: Eye Aspect Ratio (EAR) is a measure used in computer vision and image processing to detect changes in
eye behavior. It is a mathematical representation of the ratio of the width of an eye to its height. EAR is commonly used in
driver drowsiness detection systems, where it is used to determine if the driver is becoming drowsy or distracted by
measuring the changes in the shape of their eyes. EAR is calculated using the following formula:
• EAR = (|p2-p6| + |p3-p5|) / 2|p1-p4|
8. Cont.…
• Where p1, p2, p3, p4, p5, and p6 are the six key points on the eye, detected using facial landmarks detection algorithms
such as dlib.
• The EAR value is then compared to a threshold value, which is determined by training a machine learning model on a large
dataset of images and videos of eyes in various states of drowsiness and distraction.
• If the EAR value falls below the threshold, it indicates that the driver is becoming drowsy or distracted, and an alert can be
generated to warn the driver to take a break or rest before continuing to drive.
• In conclusion, EAR is a useful measure in computer vision and image processing for detecting changes in eye behavior. It
is commonly used in driver drowsiness detection systems to detect when the driver is becoming drowsy or distracted, and
to generate alerts to prevent accidents.
9. Technology
S.No Existing Work Proposed Work
1 In conventional method, driver
drowsiness cause a death or
accident issues
The system will detect the
Driver’s eye and face for further
processing to detect whether the
driver is active or drowsy. The alert
system also be there which will
alert the driver about the
drowsiness.
10. System Requirement
• Hardware Requirements
• System: Intel core I3 3.80 GHz 64 bit.
• Monitor: LED.
• Mouse: Logitech.
• Ram: 4.00 GB.
• Camera
• Software Requirements:
• Operating system : Windows 10
• Language : python
• Platform : Visual Studio
• Frontend :python IDE