Traffic Sign Classification using CNNs

This project classifies German traffic signs using three deep learning models:

Models Used:

• Model 1: MobileNetV2-based Model
• Model 2: Custom CNN with Dropout and Batch Normalization
• Model 3: Deeper Custom CNN with Batch Normalization and Dropout

The models are trained and evaluated using the GTSRB dataset (German Traffic Sign Recognition Benchmark).

Setup Instructions

1. Install Dependencies

Ensure the following libraries are installed. You can install them using pip:

pip install tensorflow numpy pandas opencv-python matplotlib seaborn scikit-learn pillow

2. Run Jupyter Notebook

To open Jupyter Notebook, run this in your terminal or command prompt:

jupyter notebook

How to Run the Code in Jupyter Notebook:

1. Make sure the repository contains the dataset (gtsrb).
2. Open the Jupyter Notebook and run Final_Project_Traffic.ipynb.
3. Run the cells sequentially:
   • Preprocess and normalize images.
   • Define, train, and evaluate three models (Model 1, Model 2, Model 3).
   • Generate performance metrics such as:
     • Training and Validation Metrics (accuracy/loss plots).
     • Confusion Matrices for all models.
     • ROC Curves for model comparison.
   • Visualize test predictions with actual and predicted labels.
   • Trained models will be saved in the Model/ directory.

Dependencies

Ensure the following Python libraries are installed:

• TensorFlow (2.x)
• NumPy
• Pandas
• OpenCV (opencv-python)
• Matplotlib
• Seaborn
• Scikit-learn
• Pillow (PIL)