Skip to content
IRIS NEURALContact

The prisonIntrusion detection

CAM-116

A hole in the wing wall

The wing seen from above. On the ground floor there’s a hole in the wall: a run of masonry blown out from top to bottom, the mesh inside bent and exposed, and rubble and dust spread across the floor. An officer is standing in front of the gap. Around it the wing carries on: the fixed tables with their stools, the upper tier with people leaning on the rail, and the cells open on both sides.

The wing seen from above. On the ground floor there’s a hole in the wall: a run of masonry blown out from top to bottom, the mesh inside bent and exposed, and rubble and dust spread across the floor. An officer is standing in front of the gap. Around it the wing carries on: the fixed tables with their stools, the upper tier with people leaning on the rail, and the cells open on both sides.
CAM-116IRIS VisionLive

Turn on IRIS Vision

AI-generated image

CAM-116 · WING WALL · STRUCTURE BREACHED

What IRIS understands

What it recognizes in this scene

  • hole in the wall
  • exposed mesh
  • rubble on the floor
  • officer in the wing
  • fixed tables in the wing

The rule

You decide once. IRIS applies it always.

If

a hole, a break or a deformation appears where there used to be a wall, a grille or a mesh fence

Then

  1. alert the duty governor with the camera, the point on the wall and the time
  2. keep the video from before the damage appeared
  3. log it as a maintenance report, with its picture
The instruction that set it up«Tell me when damage appears on a wall, a grille or a mesh fence in this area, and tell me while it is happening, not on the night round.»

What changes

What changes for whoever answers for the facility

The officer on the three o’clock round walks past the wall, sees the hole and reports it. Until that moment nobody knew it was there. Hours can pass between the wall breaking and somebody walking by, and anything fits in that gap: that it’s been getting bigger, that nobody has the video, that the report ends up written from memory. With IRIS the alert arrives when the wall breaks, with the camera, the spot and the video from just before. It works the same for a forced grille or a panel somebody has been loosening for weeks. IRIS doesn’t say who did it or why: it says there’s a gap where there used to be a wall, while it’s happening. What’s done next is the facility’s decision.

What people usually ask

Does IRIS identify whoever broke the wall?

No. IRIS does not identify anyone: it does not recognize faces, it does not compare them against any database and it does not store who anyone is. It detects the situation — there’s a hole where there used to be a wall — and reports it. Nor does it predict or classify anyone: it doesn’t say who might do it and it doesn’t single out any of the people held there. Who, how and why is an investigation for the facility, and that’s what the video of the moment is for — which is exactly what nobody has today.

What about small damage, the kind that’s done bit by bit?

Those are the hardest to spot on a round, because every day they look the same. The camera compares what it sees now with what it normally sees at that exact spot, so a plate that shifts a little each week or a mesh fence that keeps giving way does come out in the end. How big something has to be before it is reported is the facility’s call: if the threshold is very fine, it will also flag a chipped edge, and that isn’t a failure either, it’s a setting to be adjusted.

Can the same camera watch the wall and the wing’s headcount?

Yes, and that’s the advantage of the rule sitting on top of the camera that’s already there: the same image carries several rules at once, each with its own hours and its own recipient. Here one watches the wall and another can count how many people are in the wing. You don’t need a camera for every thing you want to know: you need to write another rule.

A different case every time

A camera looks. IRIS understands what it sees.

An extract of the real screen, no voice-over and hands-free. Each round poses another situation. What sits behind it is a good deal bigger than what you see.

  1. Checks the whole system
  2. Flags what is failing
  3. Puts it on one screen
IRIS NEURAL
30ES
IRISDirectoGrabacionesGISIncidencias3996SituaciónCasosLPRAutomatizacionesIRIS DATA
Ask a question or make a request…Send
IRIS · Infinity Neural

And the best part

You write the rule yourself, in your own words.Nobody has to program it for you.

“Tell me if a fire door is left open.” That’s the whole rule. From then on it applies on every camera and at any hour, without ever touching it again.

Tell us your problem and we’ll say whether IRIS understands it, or not yet.

A person answers, the same working day.