NotCrash

Safety starts from within

Comment

Inspiration

Distracted driving has been, and continues to harm society everyday. NotCrash is our first step into making the world a safer place.

  • Distraction was a factor in nearly 6 out of 10 moderate-to-severe teen crashes.
  • 80% of collisions and 65% of near crashes have some form of driver inattention as contributing factors
  • Economic losses caused by traffic collision-related health care costs and lost productivity are at least $10 billion annually

What it does

A connected vehicle application that brings a new safety system which acts from the inside. Leveraging machine learning classifiers and speech recognition software from OpenCV, dlib, and Google Cloud APIs to check and restore driver alertness while providing aditional hands-free functionalities. When a driver is determined to be inalert, an auditory cue is initiated and can only be disengaged by verbal confrimation by the driver. Each incident is recorded in a MongoDB collection, past a certain threshold an SMS message is sent to the user that includes the timestamped incidents.

How we built it

The image recognition component of the project was completed using OpenCV and dlib. The voice recognition was completed through the use of Google Cloud Speech Recognition which provided the data to be stored in a MongoDB (Atlas) database. When documents accumulated in the database, an SMS message is sent through the MessageBird Stdlib API to the user as a report. This API service was also used for the voice command text message feature.

Challenges we ran into

Due to the processing power required to run OpenCV, in combination with the Google Cloud Speech Recognition API, there was severe lagging occurring between the processes which resulted in frequent crashing. There was also the issue of actually integrating our separate Python modules together which caused flow conflicts which needed to be solved.

Built With

Share this project:

Updates