Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 2

INTRODUCTION

Driver Drowsiness Detection.


What Is Drowsy Driving?
Continuous driving can be tedious and exhausting. CDC defines drowsy driving as
a dangerous combination of driving and sleepiness or fatigue. Due to the lack of
allowed body movement, a driver might start to have droopy eyes, feel sleepy and
eventually fall asleep while driving.

Our goal is to create drowsy driver detection application that detects and alerts
users if their eyes are closed for a long time.

Our Approach to Drowsy Driver Detection System:


Our requirements for DDD are:
1. Access to a camera.
2. An algorithm to detect facial landmarks.
3. An algorithm to determine what constitutes “closed eyelids.”

Solution for above points are:

 We can use any camera capable of streaming. For demonstration purposes, we will
use a webcam.
 We will use the pre-built Mediapipe Face Mesh solution pipeline in python.
 We will use the simple yet robust Eye Aspect Ratio (EAR) technique

An eye blink is a speedy closing and reopening action. For this, we use an SVM [support
vector machine]classifier to detect eye blink as a pattern of EAR values in a short temporal
window.

How can we detect drowsiness?


We are not looking to detect “blinks” but rather whether the eyes are closed or
not. To do this, we won’t even need to perform any training. We will make use
of a simple observation, “Our eyes close when we feel drowsy.”

 To create a driver drowsiness detection system, we only need to


determine whether the eyes remain closed for a continual interval of time .

To detect whether the eyes are closed or not, we can use the Eye Aspect Ratio (EAR) formula:

You might also like