🚗 Automatic Vehicle Number Plate Recognition

An intelligent system for real-time vehicle license plate detection and recognition using YOLOv8 and EasyOCR.

🎯 Features

• Real-time vehicle detection and tracking
• License plate detection using custom-trained YOLOv8 model
• Optical Character Recognition (OCR) for plate text extraction
• Vehicle tracking using SORT algorithm
• CSV output generation with detection results
• Result visualization tools

⚙️ Requirements

• Python 3.10+
• CUDA-compatible GPU
• Dependencies:

  ultralytics
  opencv-python
  easyocr
  numpy
  pandas
  filterpy
  scikit-image
  

🚀 Getting Started

1. Clone the repository:

   git clone https://github.com/SakshamJr/Automatic-Vehicle-Number-Plate-Recognition
   cd Automatic-Vehicle-Number-Plate-Recognition

2. Install dependencies:

   pip install -r requirements.txt

3. Train the model:

   python main.ipynb

4. Run detection:

   python main.py

📁 Project Structure

└── 📁 Automatic-Vehicle-Number-Plate-Recognition
    ├── 📁 dataset
    │   └── 📁 train
    │       ├── 📁 images
    │       ├── 📁 labels
    │       └── 📄 dataset.yaml
    ├── 📄 main.py
    ├── 📄 main.ipynb
    ├── 📄 utils.py
    ├── 📄 sort.py
    ├── 📄 visualize.py
    └── 📄 README.md

🛠️ Components

• Vehicle Detection: YOLOv8 model for vehicle detection
• License Plate Detection: Custom YOLOv8 model
• Tracking: SORT algorithm implementation
• OCR: EasyOCR integration
• Data Processing: Automated CSV generation and result handling

📊 Output Format

Generated CSV includes:

• Frame number
• Vehicle ID
• Vehicle bounding box coordinates
• License plate location
• Detected license plate text
• Confidence scores

✨ Future Improvements

• Multi-GPU support
• API integration
• Web interface
• Real-time streaming support
• Multiple camera support

👤 Author

Saksham Sapkota

🙏 Acknowledgments

• Ultralytics YOLOv8
• EasyOCR
• SORT Algorithm
• Computer Vision Engineer