Deep Learning
Project: Build a Traffic Sign Recognition Program
My neural network model trained to decode traffic signs from natural images by using the German Traffic Sign Dataset.
Click here to see full source code with visualizations in Jupyter notebook!
- Download the dataset. This is a pickled dataset in which we've already resized the images to 32x32.
- Original notebook from Udacity:
git clone https://github.com/udacity/CarND-Traffic-Signs
cd CarND-Traffic-Signs
jupyter notebook Traffic_Signs_Recognition.ipynb
I have used Vivek Yadav's transform image function. It applies random affine transformation using angle,shear and translation ranges. Here is sample what generated images look like
Images count per label after adding generated images
This project requires Python 3.5 and the following Python libraries installed:
- Jupyter
- NumPy
- SciPy
- scikit-learn
- TensorFlow
- Matplotlib
- Pandas (Optional)
Click here to see full source code with visualizations in Jupyter notebook!