How abhijitshukla/automatic-numberplate-detection works
Project type and purpose
• This is a full-stack Python application for Automatic Number Plate Detection.
• It captures video frames, detects license plates via YOLOv8, performs OCR with PaddleOCR, then sends results to a FastAPI backend which persists them in a MySQL database.
File structure
• Top-level scripts:
• main.py – orchestrates video input, detection, OCR, and API calls
• server.py – defines FastAPI routes and database logic
• Model assets:
• license_plate_detector.pt – YOLOv8 weights
• coco1.txt – class labels for YOLO
• Sample media:
• sample/sample.mp4 – test video input
• Config and dependencies:
• requirements.txt, .gitignore
• README.md describes setup and usage
• .env (not in file tree but mentioned) holds DB connection details
Key insights from README
• Real-time video processing with OpenCV and CVZone
• Detection via YOLOv8 (Ultralytics)
• OCR via PaddleOCR
• Backend: FastAPI + Uvicorn
• Database: MySQL accessed via environment variables