PGAK

Vehicles

An ANPR system that turns the gate register into something you can search

An ANPR system earns its place at exactly one moment: when somebody asks which vehicle left with the material, and the only record is a security guard's notebook in handwriting nobody can read. Automatic number plate recognition logs every vehicle at the gate with its plate, type, direction, timestamp and a snapshot — so the question becomes a search rather than an argument. This page covers deploying one across real gates; for how plate recognition works on a camera feed, see vehicle and number plate recognition.

Book a free demo →Get a free camera audit

  • Per lane

    one camera and sized hardware for each

  • Indian plates

    tuned for local formats, not imported models

  • Snapshot kept

    every logged plate keeps its image

  • Searchable

    across gates, instead of per notebook

Sound familiar?

  • The gate register is a paper notebook, filled in inconsistently, and unreadable after a month.
  • Material left on a truck and reconciling which one took three weeks and never fully resolved.
  • Resident and staff vehicles stop at the gate every single day for a check nobody actually performs.
  • A previous ANPR quote assumed one camera could cover a two-lane gate and it never worked.

Where an ANPR system actually pays for itself

The value is concentrated in a few situations. Outside them, plate recognition is an expensive way to produce a log nobody reads.

Dispatch and material movement

Factories and warehouses where goods leave on vehicles. Every movement timestamped against the dispatch schedule turns 'which truck' from an investigation into a query, and an unbooked vehicle at a bay after hours becomes an alert.

Gates with heavy repeat traffic

Housing societies, office parks and campuses where the same few hundred vehicles come and go daily. Registered vehicles pass without stopping; everything unregistered is logged with a photo.

Contractor and fleet oversight

Sites where vehicles belong to other people. Arrival and departure times become evidence for billing disputes rather than a matter of recollection.

Multi-gate estates

Large sites where a vehicle entering gate one and leaving gate three is currently two unrelated notebook entries. One searchable record across gates is the whole point.

The camera is the project

ANPR is the least forgiving capability PGAK deploys, and almost every failed installation traces back to the same few placement decisions rather than to the software.

Plate height, not general coverage

A camera mounted high for a wide view of the gate area will see vehicles perfectly and read no plates at all. ANPR needs a camera aimed at plate height, along the direction of travel.

Enough pixels across the plate

The plate must occupy enough of the frame to resolve characters. This is a lens, distance and angle calculation done at survey — not something a better model rescues later.

One camera per lane

A two-lane gate needs two cameras. A single camera asked to cover both lanes reads plates well in one and poorly in the other, which is the most common reason a previous ANPR install disappointed.

Headlights, rain and night

Night is where cheap ANPR falls apart, because headlight glare washes out the plate. Placement and exposure handling are settled during the survey, in the conditions the gate actually experiences.

Indian plates, and why generic models struggle

Off-the-shelf plate recognition is generally trained on European and American plates and degrades noticeably on Indian ones — different aspect ratios, varied fonts, state-code formats, decorative and non-standard plates, and a meaningful share of vehicles whose plates are damaged, obscured or simply not to spec. PGAK's recognition is tuned for Indian plate formats for exactly this reason. It is still not magic: a plate that a person cannot read in the snapshot is a plate the system will not read either, which is why every logged event keeps its image alongside the text.

Sizing the hardware, honestly

Plate recognition is the most computationally demanding thing PGAK runs at the edge, and this is where quotes get optimistic. Measured on real hardware, a low-power edge box handles roughly one lane of ANPR comfortably; the very small form factors that are fine for person and vehicle detection across several cameras do not cope with plate reading on two. A multi-gate site is therefore sized per lane rather than per site, and that arithmetic is done before quoting rather than discovered during commissioning. It is also why an honest ANPR survey takes longer than a general CCTV survey.

ANPR system — common questions

What is an ANPR system?

Automatic Number Plate Recognition — cameras and software that read vehicle registration plates automatically and log each movement with the plate, vehicle type, direction, time and a snapshot. It replaces the manual gate register with a record you can search.

How accurate is ANPR on Indian number plates?

Good when the camera is placed correctly, and poor when it is not — placement matters far more than the model. Indian plates vary in font, format and condition, so PGAK uses recognition tuned for local plates rather than an imported general model. Non-standard, damaged or obscured plates remain genuinely hard, which is why every event keeps its image so a human can confirm.

Can ANPR run on our existing gate camera?

Only if that camera is aimed at plate height along the direction of travel with enough resolution across the plate. Many existing gate cameras are mounted for general coverage and will need repositioning or a dedicated lane camera. The feasibility check gives a straight yes or no per camera before anything is quoted.

Do we need one camera for each lane?

Yes, for reliable reading. A single camera covering two lanes reads one well and the other badly. Budgeting per lane from the start avoids the most common disappointment with ANPR installations.

Can ANPR open a boom barrier automatically?

Recognition of registered vehicles can trigger an action at the gate where suitable barrier hardware is in place and can accept a trigger. Barrier supply and its electricals are a separate scope from recognition, so it is worth confirming what your gate already has before assuming automation is included.

Related solutions

Read more on this

Start with a free audit of your own cameras.

Your number and camera count is all it takes. We call within one working hour and send the report within 48 hours.

Get my free audit →See pricing
Who you are dealing with

A registered Ludhiana company that builds the software it sells.

PGAK Innovations Pvt. Ltd. is incorporated in Punjab, engineers its own AI layer, and runs the free camera audit itself — the people who answer the phone are the people who built the product. About the company →

Founded
2023, Ludhiana
Registered office
BK Towers, 2480/2, Gill Rd, Ludhiana 141003
CIN
U62013PB2023PTC058631
Leadership
Aditya Mittal · Puneet Garg · Ankur Kaplesh
Talk to us
+91 62839 93600WhatsAppOpen Mon–Sat 09:00-19:00
Free — no obligation

Get your free camera audit.

Tell us your WhatsApp number and how many cameras you have. We call within one working hour, 9 am to 7 pm, Monday to Saturday. Your audit report follows within 48 hours — on the cameras you already own.

What are you protecting? (optional)

No spam, no new hardware, no pressure to buy anything.