CAM-102
Fight in the day room
The day room of a wing, seen from above. In the middle of the floor a group is locked together, struggling, and just below it one person is face down on the ground with another leaning over them. Four officers are crossing from the left towards it, one of them in the foreground. On the upper landing people are leaning on the railing watching what is going on below, and back on the left some are still sitting at the tables.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- people struggling
- person on the ground
- officers crossing the room
- officer moving in
- people watching from the landing
What this camera measures
An alert says something is happening. The reading says how much.
- 12 sGoing on for
- 5People in the scuffle
- 1On the floor
- 4Officers in the room
- 9Watching from above
- 17:05Time of the alert
The figures on this screen are an example from the illustration, not any client's measurement.
The rule
You decide once. IRIS applies it always.
If
there are sudden violent movements between people packed close together in the day room, or someone is on the ground and does not get up
Then
- alert the control room with the camera, the wing and the time
- put the live feed of that room on the screen of whoever needs to see it
- keep the video from a few seconds before, for the incident report
The instruction that set it up«Tell me the moment two people come to blows in the day room, and tell me too if someone goes down and does not get back up.»
What changes
What changes for whoever answers for the facility
A fight seen at ten seconds and the same fight seen at three minutes are not the same event. In three minutes there are people on the ground, more people involved and a far longer report to write. Today it all depends on that camera being on the screen someone happens to be watching right then, and with hundreds of cameras and two people in the control room that is luck, not craft. With IRIS the camera puts itself in front of whoever needs to see it, with the wing and the time. Who decides what to do and who walks into that room is a person: IRIS does none of that and it does not stop a fight. And the video of the exact moment is kept, which is what later backs the report and protects the officer who went in to break it up as well.
What people usually ask
Does IRIS recognise the people who are fighting?
No. IRIS does not identify anyone: it does not read faces, it does not compare them against any database, it does not store who anyone is and it keeps no file on anyone. What it detects is that something is happening at a point in the room, and what it hands over is the camera, the time and the video. It cannot be used to sanction a particular person: that comes from the video, the report and the people who answer for the facility, exactly as it did before.
Can't it warn earlier, when you can see it coming?
No, and promising that would be a lie. IRIS does not predict, it does not score anyone and it does not classify anyone by how they behave. It reports what is happening in front of the camera while it happens. What you gain here is not foresight: it is finding out at ten seconds instead of at three minutes, and that alone changes the event.
What if it is horseplay, or two people hugging?
It happens, and that is why the alert closes nothing and accuses nobody: it puts the camera in front of a person so they can look. If it is a hug, it is dismissed in two seconds and it cost nothing. The threshold is tuned by the facility with its own staff, because a wing day room and an exercise yard do not behave the same, and nobody outside knows that better than the people inside.
A different case every time
This isn’t a video. It’s a slice of the real screen.
A summary of the real interface, played hands-free. You see one case from start to finish; what you do not see are the other screens, and there are many.
- The rule gets built
- It goes to every camera
- Published, it watches alone
- And then it fires
Next step
Who watches the cameras of a whole city?
23 cases: 18 raise an alert when something happens and 5 only measure. You see each image as it is, and then with what IRIS understands on top.
Who watches an entire motorway?
10 cases: 8 raise an alert when something happens and 2 only measure. You see each image as it is, and then with what IRIS understands on top.
And the best part
You do not need to buy cameras:you need yours to understand what they see.
Nothing gets built, mounted or rewired. It plugs into what is already up and running, and the only thing that changes is what those cameras understand.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.