Five industries worked hands on, so we arrive knowing where the gaps usually are.
All industries →Request a demoFour ERP platforms in house, and fifteen capability modules. The recommendation is never tied to a licence quota.
All solutions →Request a demoOne team from platform selection through to the reports on your desk.
All services →Request a demoGuides, how-tos, comparisons and a plain-English glossary, written by the consultants who deliver the projects. No form in front of any of it.
Nobody in UAE ERP publishes their tools or their pricing. We do both.
Free tools →Request a demoAn ERP and IT consultancy in Business Bay, Dubai, implementing Oracle, Microsoft, Odoo and SAP across the UAE and GCC.
Consultants in the UAE, technical team in South Asia. Cost stays sharp without cutting corners.
About QZ Infomatics →Request a demoYour cameras are already watching.
CogniVision makes them intelligent.
We are a CogniVision implementation partner. We design, deploy and integrate AI-powered vision solutions for manufacturing, facilities management, retail, logistics and more across the GCC. It works with the cameras you already have on site.
Definition
A typical warehouse, mall or plant has dozens of cameras recording continuously, watched by nobody, reviewed only when something has already gone wrong.
AI video analytics turns camera feeds into structured data rather than footage somebody reviews after an incident.
How it works
Plugs into your existing CCTV or IP camera setup. Works with any standard 4MP or 5MP cameras you already have on site. No new hardware needed in most cases.
Your existing camerasAdvanced AI models watch every camera feed in real time. They detect people, behaviour, compliance violations and safety data automatically.
Every feed, in real timeYou get instant alerts, visual dashboards, heatmaps and compliance reports delivered to the people who need them.
To the people who actProducts
CogniVision ships as four distinct products. Each solves a specific operational problem, and each can run on its own or alongside the others on the same camera infrastructure.
Digital tokens, counter routing and virtual queuing for branches, clinics, government service centres and telecom stores. Customers take a ticket from a kiosk, mobile or QR code and get called to a specific counter, while the system records waiting and service time for every visit.
ANPR & Number Plate RecognitionAutomatic number plate recognition for gates, car parks, logistics yards and gated communities. Reads plates as vehicles approach, matches against a registered list, and triggers barriers or logs movements. Handles the seven UAE emirate plate formats and both Arabic and English numerals.
People Counting & FootfallEntrance and zone counting for retail, malls, branches and buildings. Produces conversion rate against transactions, occupancy by hour, and the staffing curve that matches actual traffic rather than assumed traffic.
Face Recognition AttendanceContactless workforce attendance without cards or fingerprints, feeding overtime, late deductions and unpaid absence straight into payroll. Suits construction and manufacturing workforces where fingerprint readers struggle with worn hands.
Industries
01
See exactly where your customers walk, which aisles get ignored, and when queues are getting too long. Get heat maps, automated footfall counting, staff tracking, and shoplifting detection all in one system.
02
If someone walks into a zone without their hard hat, vest, or gloves, the system flags it immediately. It also detects falls, tracks idle time, monitors restricted zones, and creates audit-ready compliance logs.
03
Monitor crowd density in real time, spot unattended bags, detect people too close to platform edges, and flag anyone entering restricted areas. Queue analytics help optimize security and boarding flows.
04
See how your space is being used with occupancy heatmaps, track staff attendance through movement analytics, and catch intruders or suspicious behavior automatically.
05
Automate attendance using face recognition, detect fights and bullying, reduce delays, track ward compliance, and use video summarization to review incidents in minutes instead of hours.
06
Detect patient falls the moment they happen, monitor OPD wait times to reduce delays, track ward compliance, and use video summarization to review incidents in minutes instead of hours.
07
Automate PPE checks in hazardous zones, detect unsafe behavior before accidents happen, and monitor restricted area access 24/7. Safety audits that used to take days can now pull visual evidence instantly.
08
Detect cargo theft and suspicious movement in real time, map yard activity with heatmaps, track manpower efficiency across shifts, and find exact incident moments in minutes with video search.
Detection
Nine detection types cover the requests we receive most often. Each one runs on standard camera feeds and produces a timestamped, exportable record rather than footage that has to be searched.
Hard hats, high-visibility clothing, gloves and eye protection checked at entry to a zone rather than audited afterwards.
A person on the ground detected and escalated in seconds rather than at the next patrol.
Entry into a restricted or hazardous area, by time of day and by whether the person is authorised.
Where people go and how long they stay, aggregated rather than tracked individually.
Queue length and crowd density, with an alert before a threshold rather than a report afterwards.
Activity against the line or the work area, measured consistently rather than sampled.
Hours of footage reduced to the moments something happened.
Loitering, tailgating and movement patterns that differ from the norm for that location and time.
Whether a documented procedure was actually followed, with a timestamped record either way.
Detection accuracy in every one of these is determined more by camera placement, angle and lighting than by the model. A site survey establishes what your existing cameras can support and where new ones are needed.
Privacy
Any system processing images of people needs a documented position on four questions: what is retained, for how long, who can access it, and on what basis.
Face recognition warrants particular care. In employment contexts, staff should be notified before enrolment and the policy documented — including what is stored, usually a mathematical template rather than a photograph.
These questions belong at design stage, not after installation.
Deployment
| Cloud | On-premise |
|---|---|
| No additional server hardware | Runs on a local server you already own |
| Scales across sites, deploys in days | Footage never leaves the building |
| Suits retail chains and SMEs | Suits government, defence, oil and gas |
Where the data goes is usually a policy decision rather than a technical one. Government, defence and oil and gas clients typically require on-premise; retail chains and SMEs almost always take cloud. We will implement either.
CogniVision FAQs
CogniVision is a computer vision product line applying video analytics to practical operational problems: queue and token management, number plate recognition, people counting and footfall analytics, and face recognition for attendance. Each product runs on standard camera hardware with the analysis handled in software.
Computer vision is software that interprets images and video the way a person would — counting people, reading a number plate, recognising a face, detecting an event. Businesses use it to automate observation tasks that would otherwise need someone watching a screen or standing at a door.
Video analytics turns camera feeds into structured data: how many people entered, how long they waited, which vehicle arrived when, whether an area was occupied. That data then drives reporting and system actions, rather than only existing as footage someone reviews after an incident.
Four: a queue management system with tokens and counter routing, ANPR for number plate recognition at gates and car parks, people counting and footfall analytics for retail and branches, and face recognition attendance for workforce time capture. They run independently or together.
Often yes, depending on camera resolution, positioning and frame rate. Analytics has stricter requirements than recording — a camera adequate for evidence may be mounted at the wrong angle or resolution for reliable counting or plate reading. A site survey establishes what can be reused.
Both deployment models are available. Edge processing keeps video on site and transmits only the resulting data, which suits organisations with data residency or bandwidth constraints. Cloud processing simplifies multi-site management. The choice usually follows your privacy position rather than technical preference.
Any system processing images of people needs a clear position on what is retained, for how long, who can access it and on what basis. Face recognition in particular warrants explicit policy and, in employment contexts, employee notification. These questions belong at design stage, not after deployment.
Two to six weeks per site depending on the product and whether cameras need installing. Site survey, camera positioning and calibration take most of the effort, because accuracy in vision systems is determined largely by physical setup rather than by software configuration.
Talk to a consultant
Book a free 30 minute call with QZ Infomatics in Dubai. You will leave it with a platform recommendation, the reasoning behind it, a realistic timeline and an indicative budget band — before you commit to anything.
