CAM-87
Entering with the face covered
A person comes in through the street entrance with their face covered, among other passengers walking through at the same time. IRIS sees that the face is covered: it does not recognise faces and does not know who anybody is.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- covered face
- street entrance
- passengers walking by
The rule
You decide once. IRIS applies it always.
If
a person comes in with their face covered
And alsoit is one of the entrances and one of the time bands the operator has marked
Then
- notify whoever watches that specific entrance, and nobody else
- tag the clip so it can be found later in the archive
- nothing else: no identifying, no matching against any list, no storing who anybody is
The instruction that set it up«At the service entrance, and only outside opening hours, tell me if someone comes in with their face covered.»
What changes
What changes for whoever runs the station
Having your face covered is not suspicious. A mask in winter, a scarf, a helmet: if this rule flagged every covered face in the station, the security manager would switch it off on day two, and he would be right. Where it does earn its place is narrowed down: one specific entrance that has a rule of its own, one specific time band, or combined with another condition the operator decides. And above all it earns its place afterwards, when someone has to search the archive: instead of watching eight hours of one doorway, that manager types what he is looking for and gets the clips.
What people usually ask
Does IRIS recognise faces?
No. Never. IRIS does no facial recognition, compares no face against any database and stores nobody's identity. What it detects here is a situation — there is a covered face at an entrance — exactly as it detects a suitcase, a bicycle or a puddle. If what you want is a system that tells you who that person is, IRIS is not it and will not be.
So does IRIS flag every covered face?
No, and saying so matters. A mask, a scarf in January, a motorbike helmet: there are a thousand ordinary reasons to have your face covered, which is why a rule that flags them all is a useless rule nobody leaves running for two days. This detection only makes sense in combination: at one specific entrance, in one specific time band, or alongside another condition set by whoever runs the station. On its own, it does not get switched on.
What is this detection actually good for?
For two concrete things. The first: entrances where a written rule already exists — a restricted access, a service door outside opening hours — where the station itself has already decided what is and is not allowed. The second, and the one most used: searching afterwards. When something has happened and you need to review who came through that door between two and three, having the archive tagged turns an afternoon's work into a one-minute search. In neither case does IRIS say who anybody is.
The real interface, step by step
Everything comes in. Only what matters gets through.
This sums up what IRIS does with one alert, from the moment it appears until it is closed. The real screen holds far more; this is the main thread.
- Four cameras at once
- Each understands its own view
- The numbers climb by themselves
Next step
Who sees the sales that never reach the till?
10 cases: 4 raise an alert when something happens and 6 only measure. You see each image as it is, and then with what IRIS understands on top.
Who is watching when the load swings over somebody?
11 cases: 9 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
Finding something in yesterday’s videoshould take as long as asking for it.
No more scrubbing back through six hours with your finger on the mouse. You type what you are after, in plain words, and the matching moments come up. Same cameras as always.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.