🛑 Smart Traffic Sign Recognition System

🔍 Overview

Smart Traffic Sign Recognition System is an AI-powered application that classifies traffic signs from uploaded images using a deep learning model. Built using Streamlit and TensorFlow, this system is ideal for autonomous vehicle systems, road safety monitoring, and smart traffic solutions.

---

✨ Features

• ✅ Deep Learning-Powered – Built with Convolutional Neural Networks (CNNs) for high performance
• ✅ High Accuracy – Achieves 96.5% accuracy on test data
• ✅ Streamlit UI – Interactive, user-friendly web interface
• ✅ Auto Preprocessing – Grayscale conversion, histogram equalization, normalization
• ✅ Instant Results – Real-time predictions on uploaded images
• ✅ Optimized Input Pipeline – Uses OpenCV for efficient image processing
• ✅ Lightweight & Fast – Quick deployment without requiring a heavy backend

---

🛠️ Tech Stack

• Frontend/UI: Streamlit
• Backend/Model: TensorFlow, Keras
• Image Processing: OpenCV, NumPy, Pillow
• Deployment Ready: Can run locally with minimal setup

---

🚀 How It Works

1. Upload an image of a traffic sign using the file uploader.
2. The system preprocesses the image (grayscale, normalize, resize).
3. The trained CNN model classifies the image.
4. The predicted sign name is displayed instantly.

---

🧠 Model Info

• Architecture: Convolutional Neural Network (CNN)
• Input Size: 32x32 (grayscale)
• Layers: Conv2D, MaxPooling, Dropout, Dense
• Optimizer: Adam
• Loss Function: Categorical Crossentropy

---

📌 Future Improvements

• 🧭 Real-time webcam support
• 📱 Mobile app version
• 🔁 Retraining option with custom datasets
• 📈 Add prediction probability graph (bar chart)