Skip to content

Latest commit

 

History

History
111 lines (82 loc) · 3.77 KB

README.md

File metadata and controls

111 lines (82 loc) · 3.77 KB
Raw

🚗 CarVisionProject

A graduation project developing an innovative solution for car recognition and intelligence using artificial intelligence (AI) technology.

License: MIT Flutter TensorFlow Node.js

CarVision Logo

🎓 About This Project

CarVision is a graduation project that showcases the power of AI in automotive recognition. This mobile application provides instant car identification and price prediction capabilities, processing images in under 2 seconds. The project demonstrates practical applications of deep learning and mobile development technologies in solving real-world problems.

Key Achievements

  • 🎯 70% Test Accuracy across 899 car models
  • 📊 1.45M Training Images from DVM-CAR 2.0 dataset
  • ⚡ 40% Reduced Training Time through optimization and Transfer Learning
  • 🚀 <2s Processing Speed for real-time results

✨ Features

  • Instant Car Recognition: Upload or capture images for immediate classification
  • Price Prediction: Get market value estimates based on historical data
  • Cross-Platform Support: Built with Flutter for iOS and Android
  • User-Friendly Interface: Intuitive design for seamless experience

🛠️ Technology Stack

AI/ML

  • TensorFlow for deep learning
  • Transfer learning optimization
  • Custom data preprocessing pipeline

Mobile App

  • Flutter framework
  • Real-time camera integration
  • Responsive UI design

Backend

  • Node.js server
  • MongoDB database
  • RESTful API architecture

📦 Installation

Prerequisites

# Install Flutter
flutter doctor

# Install Node.js dependencies
cd nodejs
npm install

# Install Flutter dependencies
cd flutter
flutter pub get

Running the Application

# Start backend server
cd nodejs
npm run dev

# Launch Flutter app
cd flutter
flutter run

📁 Project Structure

CarVisionProject/
├── nodejs/           # Node.js server and API
├── flutter/          # Flutter application
└── preprocessing/    # Data preprocessing scripts

🧪 Model Performance

  • Classification Support: 899 car models
  • Test Accuracy: 70%
  • Processing Time: <2 seconds
  • Dataset: DVM-CAR 2.0 (1.45M images)

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.

👤 Author

Taha Khamessi & Mohammed Louai Lamsi

  • Graduation Project (2024)
  • Computer Science
  • Coursework: Software Engineering and Information Systems

LinkedIn LinkedIn GitHub GitHub

Made with ❤️ as a graduation project by Taha Khamessi and Mohammed Louai Lamsi