Driver Fatigue Detection using Computer Vision

Prevent car accidents that happen due to fatigue or drowsiness

The objective of this project is to build a device that will monitor the driver’s alertness by observing his/her eyes and head movements. If the unit deduces that the driver’s eyes are drooping down due to sleepiness or his/her head is tilting in forward direction, the unit will sound an audible alarm which will alert the driver and he can take corrective action.

    Developers
  • Sushil Patil
    Funded By
  • Self Funded

India ranks among the top in the world in road accidents and death. In 2016, India witnessed over 480,000 accidents, resulting in over 150,000 deaths. Breaking it down further, on average, there are 55 accidents resulting in 17 deaths every hour.
Most accidents are due to driver error. The road conditions, vehicle maintenance, differential between vehicle speeds, lack of discipline by drivers and citizens all lead to high stress levels in drivers. The stress induces sleepiness and fatigue. With fatigue, the drivers are more prone to making error leading to a road accident.
Our solution aims to detect stress and sleepiness of the driver. It will monitor and alert the driver before it is too late.
This solution could immensely benefit the following entities

  • State Transport Corporations
  • Commercial Vehicles carrying goods
  • Private Fleet Owners
  • Trains
  • Oil / Chemical Tankers
  • Insurance Companies

How it works?

This unit can mounted in front of the driver – behind the steering wheel of the vehicle. It is an application which can run on a dedicated hardware unit. The unit will continue to silently monitor the driver. The unit will monitor the driver’s alertness by observing his/her eyes and head movements. If the unit deduces that the driver’s eyes are drooping down due to sleepiness or his/her head is tilting in forward direction, the unit will sound an audible alarm which will alert the driver and he can take corrective action.