🚦 Traffic Sign Recognition using CNN & Genetic Algorithm

An intelligent system that detects and classifies traffic signs using Convolutional Neural Networks (CNN) combined with Genetic Algorithm (GA) for optimization.

---

📌 Project Overview

Traffic Sign Recognition is a crucial application in:

• Autonomous Driving 🚗
• Driver Assistance Systems (ADAS)
• Road Safety Monitoring

This project uses Deep Learning + Evolutionary Optimization to improve classification performance.

---

🎯 Aim

To develop a model that:

• Accurately detects and classifies traffic signs
• Optimizes model parameters using Genetic Algorithm
• Improves performance and generalization

---

🧾 Description

The system processes traffic sign images and predicts their class using:

• CNN for feature extraction and classification
• Genetic Algorithm for hyperparameter tuning and optimization

---

✨ Features

✔ Traffic sign classification ✔ CNN-based feature extraction ✔ Genetic Algorithm optimization ✔ High accuracy model ✔ Scalable and efficient system

---

🛠 Technologies Used

• Language: Python

• Libraries:

  • TensorFlow / Keras
  • OpenCV
  • NumPy
  • Matplotlib

• Concepts:

  • Deep Learning (CNN)
  • Evolutionary Algorithms (GA)
  • Computer Vision

---

🧠 Model Architecture

🔷 CNN

• Convolution Layers
• Pooling Layers
• Fully Connected Layers
• Output Layer (Traffic Sign Classes)

🔷 Genetic Algorithm

• Population initialization
• Fitness evaluation
• Selection
• Crossover
• Mutation

---

📂 Dataset

• German Traffic Sign Recognition Benchmark (GTSRB)
• Contains multiple traffic sign classes

---

⚙️ Workflow

1. Data Collection
2. Data Preprocessing
3. CNN Model Training
4. Apply Genetic Algorithm
5. Optimize hyperparameters
6. Final prediction

---

🔄 Methodology

🔹 Data Preprocessing

• Resize images
• Normalize pixel values
• Data augmentation

🔹 CNN Training

• Train model on traffic sign dataset
• Extract features automatically

🔹 Genetic Algorithm Optimization

• Tune hyperparameters such as:

  • Learning rate
  • Number of layers
  • Batch size

• Improve model accuracy

---

📊 Evaluation Metrics

• Accuracy
• Precision
• Recall
• F1 Score

---

▶️ How to Run

1. Install Dependencies

pip install tensorflow opencv-python numpy matplotlib

2. Run the Project

python train.py
python optimize_ga.py

---

📁 Project Structure

Traffic-Sign-Recognition/
│── dataset/
│── model/
│── train.py
│── optimize_ga.py
│── utils.py
│── README.md

---

📸 Output

• Traffic sign image
• Predicted class label
• Model performance results
• German Traffic Sign Recognition Benchmark (GTSRB) Training: 80% of the dataset. Testing: 20% of the dataset.

Input:

Output:

---

✅ Conclusion

This project demonstrates:

• Application of CNN in image classification
• Use of Genetic Algorithm for optimization
• Improved accuracy through hybrid approach

---

🚀 Future Enhancements

• Real-time traffic sign detection
• Integration with self-driving systems
• Use advanced models (YOLO, Faster R-CNN)
• Deploy as mobile/web app

---

👨‍💻 Author

• Ganesh Sesha Sai Akhil
• Venkatesh
• Arun Kumar

---

⭐ Support

If you like this project, give it a ⭐ on GitHub!