Drowsiness Detection Using 3b+ Model of Raspberry Pi: A Step into the Word of Deep Learning

Authors

  • Krishan Kumar Malik student

Keywords:

Person Drowsiness Detection, Raspberry Pi 3, Raspbian camera, Open CV, Feature Extraction, Eye Aspect Ratio (EAR)

Abstract

Person drowsiness is a first cause of respective catastrophe leads to
delay of work, loss of money. The execution of person somlolence
in real-time will aid in avoiding major delay on work. The system is
designed for workers wherein the person’s fatigue or drowsiness
is detected and alerts the person. The suggested method will use
Raspbian camera that captures person’s face and eyes and processes
the images to detect person’s fatigue. On the detection of somlolence
the programmed system cautions the person through an alarm to ensure
attentiveness. The suggested method compose of varied stages to work
out insomlolence of the person. In step with this output, the warning
message is generated. Cascade Classifiers is employed to detect the
blink duration of the person and Eye Aspect Ratio (EAR) is calculated.
Finally, the alert message and you are drowsy is displayed on screen .
For this Raspberry Pi 3 model 3b+ with Raspbian (Linux Based) package
is employed.

Author Biography

Krishan Kumar Malik, student

Global Institute of Technology,Jaipur, Rajasthan

Published

2020-05-04