RadarHub watches any video feed and turns it into traffic intelligence: it detects every vehicle with AI, tracks each one, measures its real speed in km/h, counts traffic by type, and flags speeders with photo evidence. Everything runs on your own hardware: no cloud, no per-camera fees, no subscription.
Full Python source code, web dashboard, calibration tool, CLI, full documentation, test suite. Lifetime license, updates included.
python -m venv .venv and activate itpip install -r requirements.txtpython main.py dashboard --source 0 and open http://localhost:8080python main.py calibrate --source your-video.mp4 --width 3.5 --length 20Full guide: see docs/documentation.html inside the package or the online docs.
No. CPU works with the default nano model. Any NVIDIA GPU gives real-time HD.
RadarHub is a measurement and analytics tool. Official fines usually require type-approved devices; use it for studies, private sites and monitoring, and check local regulations.
Never. Detection, dashboard and snapshots are all local.
Run one instance per camera on different ports.