As we all know, the number of vehicle accidents at intersections in urban cities are on the rise. We decided to create this application to reduce travel times, make the roads more safer and provide the users a comprehensive report about the events near them.

What it does

Our program collects data from various cameras and then sends it into the cloud. The video data is then processed using OpenCV, TensorFlow, Keras and Google Cloud Platform to generate sensible data about the road and traffic conditions. This data is compiled and sent to our private back-end server where we have utilized Mapbox's services to deploy real time, live maps of those locations.

How we built it

We have built it by applying Machine Learning concepts and developing various algorithms for sorting through and organizing data. We used python for all our basic application of concepts like TensorFlow,etc. We finally hosted our live python script on []

Challenges we ran into

Our team had a lot of brilliant ideas to work on when we started development on this project. It was hard to settle for things due to time constraints. Another challenge we faced was connecting the back-end to the front-end, however with the use of a Python script and an online service, we were able to overcome these challenges with ease.

Accomplishments that we're proud of

We are proud that the application program worked at the end and performed the way it was supposed to. We are also proud that our team was able to produce a working prototype of the idea within 36 hours from scratch.

What we learned

We learned that you can achieve anything if you are passionate and dedicated for it. We also acquired new skills like Mapbox map generation, Machine Learning, etc.

What's next for TraffIQ

We think we can optimize our model by training it with real world data. This would help us improve the accuracy of the application, and make it more reliable.

Share this project: