/Traffic-Sign-Classifier

A convolution neural network code to classify over 43 traffic signs.

Primary LanguageJupyter Notebook

Traffic Sign Classifier

This model can classify over 43 different traffic signs with a validation accuracy of around 99%.

Google Colab code:

https://colab.research.google.com/github/CleanPegasus/Traffic-Sign-Classifier/blob/master/Traffic_Sign_Classifier.ipynb

Dataset :

German Traffic Sign

Requirements :

Use Google Colab for running the model

The following libraries are required to run the code. But if you are using Google Colab, use need not install these libraries.

numpy

matplotlib

OpenCV

tensorflow

Keras

Pickle

Random

Pandas

Requests

Pillow

Procedure :

Run every cell of the notebook. Feel free to modify the code to match your needs.

You can try it with a different dataset containing more classes of signs and change the parameters.

You can try different preprocessing methods for better accuracy.

Loss and Accuracy graph :