There are thousands of car crashes every year and in order to help one of the aspects that result in crashes, drowsiness, we wanted to build an app that can detect drowsiness and alert the driver. The program will detect when the driver is drowsy by detecting whether their eyes are closed or open. If the program detects the eyes are closed then a chosen song will play out loud to alert the driver. We used Keras and TensorFlow platforms in order to build our program. We did not have enough time to implement what we originally intended to do, which was to create an app that can play a song. We are very proud to say that our accuracy rate is 97%, which measures whether the driver is falling asleep or not. We learned new functions in python and the use of data to build an accurate program. We also realized that we need to allocate more time because we may run into problems. Since we could not build an app, the next step for AISol would be to implement a functional app that can successfully play a chosen song when the driver is dozing off.

Built With

Share this project: