CrashChannel

Inspiration

In order to improve public safety, we wanted to create a program that would provide up-to-date data on traffic incidents occuring right now.

Our Project

Our project uses the City of Calgary's Open Data API to fetch images from traffic cameras all over the city, and determine if an accident has taken place. This information is used to populate a map of traffic cameras, and individual images can be tested as well, to test the machine learning model.

How we Built our Project

We built this project using multiple technologies, with Tensorflow powering our machine learning (ML), Python/Flask powering our REST API to handle testing images against our ML model, and Vue.js on the front end to build a user-friendly experience.

Challenges we faced

Handling packages, dependencies, across multiple different aspects of the project was a huge challenge, and required a lot of Googling and digging through StackOverflow.

Accomplishments that we're proud of

This is the first time any of us have ever worked with machine learning, and for some of us, it was our first experience building a full-stack web app - and it works! To us, we're really proud of that.

What we learned

A whole whack of new technologies, how to plan out and design an application from start to finish, and how to work in a team of sleep-deprived university students.