A comprehensive vehicle tracking system that utilizes advanced computer vision and OCR technology to detect and log vehicle number plates from video feeds. The system provides real-time number plate recognition, database logging, route mapping, and performance analysis through confusion matrices.
Real-time Number Plate Detection - Advanced OCR-based detection using EasyOCR for accurate text recognition from video frames Database Logging - Automatic storage of detected plates with timestamps in SQLite database Image Archival - Saves detected plate images for future reference and analysis Performance Analytics - Generates confusion matrices to evaluate detection accuracy Route Visualization - Interactive mapping functionality to display vehicle routes from point A to G Frame Processing Optimization - Processes every 5th frame for efficient performance without compromising accuracy.
OpenCV (opencv-python) - Computer vision and image processing EasyOCR (easyocr) - Optical Character Recognition for text detection Tesseract (pytesseract) - OCR engine support PyTorch (torch, torchvision) - Deep learning framework for EasyOCR
SQLite3 - Database for vehicle logs and timestamp storage NumPy - Numerical computing and array operations Matplotlib - Data visualization and confusion matrix plotting Scikit-learn - Machine learning metrics and performance evaluation
Folium - Interactive map generation and route visualization Google Colab patches - Image display compatibility
bash# Install required packages pip install opencv-python pytesseract matplotlib scikit-learn folium easyocr torch torchvision
apt-get install tesseract-ocr libtesseract-dev
detected_plates/ - Directory containing captured plate images
route_map.html - Interactive route visualization
vehicle_tracking.db - SQLite database with detection logs
Google Collab or Jupyter Notebook Python 3.7+ Sufficient storage for video processing and image archival GPU support recommended for faster PyTorch operations
