Ayonix Video Analytics

Industry

Video analytics for car parks without a sensor in every bay

In short

Parking video analytics reports live and historic occupancy from existing cameras, using Ayonix vehicle detection over bay or area zones. It removes the per-bay sensor cost and civil works, and keeps plate recognition as a separate decision with its own lawful basis.

Elevated view of vehicles moving through a modern multi-bay parking facility
Conceptual illustration of a monitored parking facility. AI-generated conceptual image, not a customer deployment.

The problem

What parking teams are dealing with

  • Drivers circulate hunting for spaces while whole areas sit empty, and the operator cannot tell them where to go.
  • Per-bay sensors were quoted and the capital and installation cost ended the business case.
  • Occupancy is known only from barrier counts, which miss anyone who entered before the count was reset.
  • Overstay is suspected but no record exists of how long bays are actually held.
  • Multi-storey decks have no per-level visibility, so signage sends drivers to full levels.

Camera placement

Where cameras need to be, and why

Camera position decides more of the outcome than any software setting. These are the placements that matter in this sector.

Recommended camera placement forParking
Location Purpose What matters
Each deck or surface area Area occupancy Height is the single biggest factor — a low view cannot see past the front row when the area is full.
Entry and exit lanes Vehicle counting and direction Across the lane rather than along it, away from habitual queueing points.
Controlled access lanes Plate reading where justified A dedicated ANPR camera per lane with controlled approach speed and dedicated illumination.
Restricted and accessible bays Presence and dwell These bays usually justify per-bay treatment where the general area does not.
Pedestrian routes Conflict visibility where required Only where a shared-route safety question genuinely exists.

Workflow

From event to acknowledged action

Event workflow

  1. 1 Deck and lane cameras are ingested and monitored for availability.
  2. 2 Vehicles are detected and tracked; bay and area polygons are evaluated with a stationary threshold.
  3. 3 Occupancy is derived per area and compared against reconciled capacity.
  4. 4 Threshold states drive entrance and deck signage in near real time.
  5. 5 Occupancy and dwell series are retained for utilisation planning.
  6. 6 Where ANPR is deployed, reads are handled under separate retention and access rules.

Operator workflow

  1. 1 Entrance signage shows per-deck availability so drivers are routed before they enter.
  2. 2 Operations sees which areas fill first and adjusts routing or pricing.
  3. 3 Restricted-bay dwell beyond the permitted period is flagged for attendance.
  4. 4 Utilisation reports support pricing and capacity decisions.
  5. 5 Any enforcement action follows a separate documented process with its own evidence standard.

Operations

Dashboards, integration and deployment

Dashboard metrics

  • Live occupancy per deck and site total
  • Peak occupancy and time of peak per area
  • Entry and exit counts by interval
  • Bay dwell distribution
  • Hours above a utilisation threshold
  • Camera availability per area

Integration options

  • REST and webhook delivery into parking guidance, signage and app platforms
  • Barrier and access-system integration where ANPR is deployed
  • VMS event delivery for security rules
  • ONVIF and RTSP ingestion from the existing camera estate
Compatibility states →

Deployment options

  • Edge processing at the car park so signage keeps working during a WAN outage
  • On-premise processing where a site aggregates cameras locally
  • Multi-site management across a parking portfolio
  • Air-gapped deployment where required by the host site
Compare architectures →

Evidence

Deployment pattern

Car park guidance pattern

Per-deck area occupancy driving entrance and deck signage, entry and exit counting, and restricted-bay dwell, with ANPR only where separately justified.

Occupancy accuracy is validated against a physical audit on your own site. No accuracy figures from other car parks are published, because camera height and layout determine the result.

This is an anonymised pattern describing how a deployment of this kind is structured. It is not a named customer reference. No customer names, logos, testimonials or measured results are published on this site, because none have been supplied for publication. More on how evidence is handled .

Governance

Privacy and governance in this sector

  • Occupancy and counting outputs are aggregate with no plate reading and no keeper identification.
  • ANPR is a materially different decision: plate data is personal data in many jurisdictions and requires a lawful basis, impact assessment, retention rule and audited access.
  • Where plate text alone satisfies the workflow, plate images should not be retained.
  • Signage must inform drivers where plate recognition operates, to the local standard.
  • Aggregate occupancy series can be published to apps and open data because they contain no personal data.

Frequently asked questions

Is per-bay accuracy achievable from existing cameras?

For the front rows usually, for rear rows in a low oblique view often not. Where per-bay state is unreliable, area-level occupancy is used for that camera. Reporting an area figure honestly is better than a per-bay figure that is wrong for half the spaces.

Do we need ANPR?

Only if the workflow needs the vehicle identity — pre-authorisation, season-ticket validation or enforcement. Occupancy, counting and guidance need none of it, and avoiding plate processing removes a substantial compliance burden.

What about multi-storey decks?

Each deck needs its own coverage and its own reconciled capacity. A camera cannot report on a level it cannot see, and the readiness review confirms coverage level by level rather than assuming whole-site coverage.

Other industries

Author
Gabriel Bamola, Chief Marketing Officer, Ayonix
Technical review
Dr Sadi Vural, Founder and Chief Executive Officer, Ayonix
Published
Last reviewed

Design a parking pilot around your own cameras

Start with a camera readiness review and a log-only baseline, then measure against acceptance criteria we agree before the pilot begins. That is what turns an analytics evaluation into a decision.