German Traffic Sign Classification

In this project, I used Python and TensorFlow to classify traffic signs.

Dataset used: German Traffic Sign Dataset. This dataset has more than 50,000 images of 43 classes. Download the dataset from here.

I was able to reach a +99% validation accuracy, and a 97.6% testing accuracy.

Pipeline architecture:

• Load The Data.
• Dataset Summary & Exploration
  • turning data into tensors
  • Getting images and their labels
  • turning images into tensors
• Data Preprocessing.
  • Normalization
  • resize
• Design a Model Architecture. -CNN Model
• Model Training and Evaluation.
• Testing the Model Using the Test Set.

Environement:

• Python 3.7
• TensorFlow 2.4.1 (GPU support)