AI Vision Smart Parking
The camera that only recorded now counts, analyses and recommends.
- Niadad solution
- Iran
- 2025 – 2026
- Pilot
Challenge
Car-park management usually relies on manual counting, expensive per-bay sensors or guesswork. Existing cameras record footage but produce no data for capacity management, vehicle guidance or planning.
Our approach
Niadad built the solution on the cameras already in place: the video stream reaches Bina, computer-vision models detect vehicles and bay states, and the result becomes structured data for analysis. Models are trained and evaluated on Rayon so they stay adapted to each site's conditions.
Solution
A complete chain from camera to decision: entry, exit and bay-occupancy detection, dwell-time and peak-hour analytics, free-capacity forecasting and outputs for guidance signs and management reporting. The solution is currently in pilot.
Technology
Computer vision on existing camera streams
Vehicle and bay-state detection
Occupancy, dwell-time and peak-hour analytics
Model training and monitoring on Rayon
Outputs for vehicle guidance and management reporting
Results
Existing cameras have become the data source for occupancy and capacity.
The detection-to-analytics chain runs end to end in the pilot environment.
Models are retrained on each site's data and become more accurate.
Next phase
Moving from pilot to operation in multi-storey car parks and connecting to smart-city and payment systems.
Platforms used
Solutions
Related projects
Let's build together.
If your organisation, bank or industry is ready to turn data into decisions, start the conversation here.