This project uses a ConvNet architecture to classify traffic signs. The dataset to train are traffic sign imgaes from the German Traffic Sign Dataset.
For more information about this project visit the Wiki page
model_architecture.py
Model architecture
model_calls.py
Functions to train and use the model
preprocess_augmentation.py
Split, balance and augmentation functions
helper_Functions.py
Functions to visualize data
This project requires Python 3.5 and the following Python libraries installed:
Run this command at the terminal prompt to install OpenCV. Useful for image processing:
conda install -c https://conda.anaconda.org/menpo opencv3
Data set for this project can be downloaded from German Traffic Sign Dataset. NOTE: Data set images used on this project have been already resized to 32x32. You may want to do that if you you are using different image size.
Clone the project and start the notebook.
git clone https://github.com/cuevas1208/Traffic-Sign-Recognition
cd Traffic-Sign-Recognition
python main.py
This project is part of Udacity Self-Driving Car Engineer Nanodegree program. Tools, techniques and knowledge learned in class about deep neural networks and convolutional neural networks were used to classify traffic signs.
To learn more about convolutional networks I recommend this book