How sumukh73/ai-smart-parking-system works
This is a Flask-based smart-parking demo with a separate video-detection process. The main implemented workflow reads video, detects and counts vehicle classes with YOLO, and stores aggregate occupancy in MySQL; the Flask app reads the latest records for an admin dashboard and live-data API. Parking-location data and detailed detection history are simulated. The README describes recommendations, but the available source does not show an allocation or recommendation engine; frontend templates are also absent from the tree.
Main entry points: app.py Flask routes · detect.py Detection process · video.mp4 Video input · yolov8n.pt YOLO weights
Components 10
Detection
- Video input
video.mp4 - Detection process
detect.py - YOLO weights
yolov8n.pt
Web application
- Flask routes
app.py - Dashboard view
app.py - Live data API
app.py - History API
app.py - Locations API
app.py
Parking data
- MySQL parking_data
Other components
- Operator / user
Connections 15
- Operator / user → Flask routes: submits login
- Flask routes → Dashboard view: renders
- Flask routes → Live data API: dispatches
- Flask routes → History API: dispatches
- Flask routes → Locations API: dispatches
- Video input → Detection process: supplies frames
- Detection process → YOLO weights: runs inference
- Detection process → MySQL parking_data: writes counts
- Dashboard view → MySQL parking_data: reads latest
- Live data API → MySQL parking_data: reads latest
- History API → MySQL parking_data: reads records
- Dashboard view → Operator / user: shows occupancy
- Live data API → Operator / user: returns counts
- History API → Operator / user: returns simulated history
- Locations API → Operator / user: returns demo locations