Perimeter protection

Smart perimeter protection — a fence that reports itself

Smart perimeter protection replaces the physical sensors nobody maintains with virtual boundaries drawn on the cameras you already own. Mark a line along the compound wall, the yard edge or the roof access, and PGAK raises a real-time alert the moment a person or vehicle crosses it — while filtering out the animals, weather and moving foliage that made your old perimeter alarm unusable.

Book a free demo →Get a free camera audit

1 min
To draw a new virtual boundary
0m
Of new cable trenched
< 3s
From boundary crossing to alert

Sound familiar?

Beam sensors and fence wires fail constantly and cost a fortune to trench in.
The old perimeter alarm fires so often at night that the guard disabled it.
A one-kilometre boundary is covered by one guard at one gate.
You learn about the breach from the damage, not from an alert.

Virtual fencing, drawn in a minute

A perimeter intrusion detection system used to mean buried cable, beam pairs and a maintenance contract. Now it means drawing a line on a camera view. You can change the boundary whenever the site changes, at no cost, on any camera that can see the area.

Line-crossing detection

Alert when an object crosses a boundary — with direction, so entering the site raises an alert and leaving it doesn't.

Zone intrusion

Enclosed areas — a yard, a substation, a roof, a fuel store — alert on any presence rather than on crossing.

Loitering along the boundary

Someone standing at the fence for two minutes is a stronger pre-breach signal than someone walking past it.

Schedules per boundary

The dispatch gate is busy until 8pm and should be silent after; the substation should be alarmed all day. Each boundary has its own hours.

Filtering is what makes it survivable

The reason most perimeter systems end up switched off is not that they miss intruders — it's that they report everything. PGAK classifies before alerting: animals, rain, insects on the lens, headlights sweeping the wall and branches in the wind are suppressed. A boundary alert means a person or a vehicle.

Response, not just notification

A boundary breach can trigger more than a phone buzz — an on-site siren, a strobe, a floodlight, or an announcement. Deterrence during the approach is worth more than evidence after the entry.

Long perimeters, no new cabling

Because it uses cameras that are already installed and powered, extending coverage along a long boundary costs a camera, not a trench. Sites typically start with the two or three most exposed stretches and widen from there.

smart perimeter protection — common questions

How is virtual fencing different from beam sensors?

Beams detect that something broke the beam — a bird does that as well as a person. Virtual fencing on camera classifies the object first, so it tells you a person crossed rather than that something did. It also needs no trenching, no power runs along the fence and no field maintenance.

Will wind, rain or animals trigger it?

They're filtered out. Object classification runs before the alerting logic, so weather, foliage and animals are suppressed by default. This is the single biggest reason perimeter systems get abandoned, and it's the part we tune hardest during the first fortnight.

Can it trigger a siren or light, not just my phone?

Yes. Boundary alerts can drive a siren, strobe, floodlight or public announcement at the site, alongside the phone notification. Deterrence during the approach beats evidence after the entry.

How long a perimeter can this cover?

As long as your cameras can see. Coverage is limited by camera placement rather than by the software, and extending it means adding a camera to an existing stretch — not laying new sensor cable.

See it working on your own cameras.

Book a free demo and watch PGAK turn one of your live feeds into intelligent protection — in about 20 minutes.

Book a free demo →See pricing