Description as a Tweet:

Our project inputs the live video stream of a driver and uses image detection to indicate whether the driver is drowsy and paying attention to the road.


We were inspired to make this project because one of the leading causes of death in the US are road accidents due to dangerous driving. We hope our application can provide a new direction to drowsiness detection.

What it does:

Our project is based on driver drowsiness detection, it monitors different behaviors and actions of the driver such as the frequency of blinking and regularly checks if the driver is paying attention to the road. The program identifies if the driver has closed his eyes for a long enough period of time that would cause harm to others and himself. Furthermore, we used the correlation between higher blinking frequency and drowsiness to identify if the driver is tired.

How we built it:

We used the OpenCV which is a very popular image detection and recognition library and some open source code. We proceeded to understand the open source code and implemented our own parameters and modules to prevent driver drowsiness.

Technologies we used:

  • Python

Challenges we ran into:

We took of time understanding the open-source code.

Accomplishments we're proud of:

We were able to finish.

What we've learned:

We learnt how to import libraries and understand predefined libraries.

What's next:

Image detection for blinking is used in many fields such as medicine, where it is used to identify patients with MS and Parkinson etc. And we would love to implement many more characteristics such as frequency of yawning of other passengers in the car, to make the algorithm more accurate. And blink frequency can also be used to help people communicate through Morse code.

Built with:

We used python libraries and OpenCV image recognition libraries

Prizes we're going for:

  • Best use of Transposit

Team Members

Rohit Rangan
Ajey Gowda
Pawan Sarma
Malyaka Imran
Mauricio Angelone

Table Number

Table 11