Inspiration
The inspiration for this project was brought by Sebastian Zech.
What it does
This project consists in identification and classification of problems concerning traffic signs such as covered, damaged or in other bad conditions.
How we built it
To build this infrastructure, we used concepts of web development and artificial intelligence. For constructing the neural network that classifies the traffic signs, we utilized Python alongside TensorFlow, which is a framework that helps build better neural networks. We also developed a website prototype where users can visualize the information gathered by the neural network through a map and perform analysis on top of it.
Challenges we ran into
The biggest challenge we ran into is the lack of data. We gathered and searched as much as we could and to complement the data we collected, we used another neural network to generate images similar to the ones we had, so we could train ou network better.
Accomplishments that we're proud of
We are very proud on how our neural network is performing and how we managed to build the interface where the user can interact with our application.
What we learned
We learned how to tackle a real problem and learned how difficult and challenging it can be to find and categorize your own training data.
Log in or sign up for Devpost to join the conversation.