Inspiration
Every year, thousands of people are killed in car accidents due to drowsy drivers. It is clear from this fact that a sleep sensor application could help prevent such accidents and save lives as a result.
What it does
Our objective in this project is to create a system for detecting driver drowsiness while driving and alerting the driver automatically so that he is able to drive safely.
How we built it
we built it using CNN machine learning algorithm . we added maxpooling Dropout , Input , Flatten , Dense layers from the keras library . Pre processing was done use image data generator . Preprocessing was converting images to grayscale.
Challenges we ran into
we couldn't get higher accuracy as we couldn't train for higher epochs , training stopped after 10 epochs as there was no significant change in validation -accuracy and there was no change even after the learning rate was altered.
Accomplishments that we're proud of
Our model detects the opening and closing of eyes . If eyes are closed alarm is triggered which alret the driver to wake up.
What we learned
We learned the implementation of the machine learning algorithm which is Convolutional neural network.
What's next for Driver Drowsiness Detection System
Our primary aim to next goal is to refine more data to get accurate results and implement the model using IOT to implement in vehicles to reduce the accident.
Log in or sign up for Devpost to join the conversation.