DriveBuddy

Inspiration

Drowsy-driving is related to at least 100,000 motor-vehicle crashes and more than 1,500 deaths per year. Drowsy Driving was responsible for 72,000 crashes, 44,000 injuries, and 800 deaths in 2013. According to a survey among nearly 150,000 adults in 19 states and the District of Columbia, 4% reported that they had fallen asleep while driving at least once in the previous 30 days. Here is where we come in. Drive buddy is an inexpensive effective solution to this problem that makes the best use of machine learning.

What it does

Using Machine Learning Algorithms it detects open and close eyes, This project delivers an Early-warning system for Drowsy driving thus raising the safety standards and help avoid fatal road accidents.

How we Built it

Built using a Convolutional Neural Network that classifies images according to the dataset. It works in tandem with a webcam and esp that can be installed in a car which buzzes when the driver closes their eyes for a long time

Tools

The tools we used were the TensorFlow library, OpenCV, webcam, buzzer, esp8266.

Challenges we ran into

We first ran into the challenge of the esp being able to communicate the information to the program. We also faced a small dataset issue due to which the program did not detect closed eyes accurately but we were able to solve it

Accomplishments that we are proud of

Creating a project which helps in avoiding fatal accidents. Working in unison despite being far away from each other. Completing this awesome project within a very short amount of time.We were able to accurately detect closed eyes and were able to trigger the buzzer. Interfacing the entire system was a challenge for us and we were quite proud of ourselves to have accomplished it.

What we learned

We have learned to use Python Machine learning libraries more efficiently and using the external board for signal transmitting.

What's Next for DriveBuddy

Drive Buddy in the future will be able to record patterns of behavior among drivers hence improving the overall system and it will also work as a black box if ever the system fails to work.

Contributors

atharwa-24
Atharwa_24
Om Khairate
Om Khairate
Dhawal Khapre
Dhawal Khapre
Tejas Khairnar
Tejas Khairnar

Built With

Share this project:

Updates