This project focuses on developing neural networks structures that are capable of perform brazillian traffic signs classification. With that porpouse, OpenCV was used to create dataset and TensorFlow framework was used to build the neural networks.
In order to work locally with the files, you should clone the repository with the following command:
$ git clone https://gitlab.com/ML-Intel/traffic-signs.git
- Anaconda >= 4.1.11
This work uses Intel Distribution for Python and Full Intel Performance Libraries wich contains all the necessary packages to run correctly the jupyter notebooks. Please follow the instructions provided on the link.
To run a Jupyter Notebook, navigate to the directory and run
$ jupyter notebook
Once in the Jupyter Notebook environment, navigate to Report_v1.ipynb.
- Thiago Daros Fernandes - CNN Network
- Frederico Damian Preve - Initial work