Critical infrastructureProduction efficiency
CAM-182
Gas cylinder on the sorting belt
The inside of a sorting hall. The stream of material comes down the belt — packaging, cans, plastic — and among it rides a red gas cylinder, whole and closed. Four workers are spaced along the line, with the reject bins underneath. IRIS sees a shape that shouldn’t be in that stream.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognizes in this scene
- gas cylinder
- material stream
- sorting station
- reject bins
What this camera measures
An alert says something’s happening. The reading says how much.
- 8 sOn the belt for
- 1Objects flagged
- 4Stations manned
- 2Stations ahead of the object
- 2Reject bins in view
- 07:35Time 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
an object goes down the belt whose shape does not fit the stream of material being sorted
Then
- alert the line supervisor with the camera and the frame showing the object
- alert the station the object is heading for, because that is who will have it in hand
- keep the video, so the plant can trace which delivery it came in with
The instruction that set it up«Tell me if something comes down the belt that is not shaped like what we are sorting, such as a gas cylinder.»
What changes
What changes for whoever answers for the site
In a sorting plant, nobody chose what arrived in the truck. Every so often a gas cylinder turns up, and a cylinder in a shredder or a furnace isn’t an incident, it’s a serious accident. Today the only filter is the eyes of whoever is at the station, eight hours in, with a full belt. IRIS watches the same belt without tiring and raises the alert while the object is still upstream, with the frame and with the station about to receive it. That gives the worker something he doesn’t have today: a few seconds of warning.
What people usually ask
Does IRIS know what’s inside the cylinder?
No, and that matters. IRIS analyzes nothing: it doesn’t sniff, it doesn’t measure and it opens nothing. What it sees is a shape that doesn’t fit what’s coming down that belt, and that alone is reason enough to stop and look at it. Whether it’s full, empty or holds something else is decided by a person, with the object in front of them.
What if the odd object is legitimate and comes past often?
Then you tune the rule. Every line sorts a different stream, and what’s out of place on one is routine on another. What counts as out of place is defined by the plant on its own belt, and it can differ per line and per shift. An alert that fires at everything is one nobody looks at within a week.
Does IRIS identify the workers on the line?
No. IRIS does not identify anyone: it does not recognize faces and it does not read license plates. It sees that a station is manned and where along the belt it is, which is what it needs in order to warn whoever is about to receive the object. Who is standing there it neither knows nor stores.
Every simulation
Emergency gear blocked
Tell me if something blocks an exit, an emergency shower or a fire lane, or if the equipment is not where it belongs.
- The sentence
- Where it applies
- What it finds
- The same rule
Next step
Who watches a whole station at rush hour?
17 cases: 11 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 sees the sales that never reach the register?
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.
And the best part
Stop scrubbing through six hours.Ask what happened and watch only that.
You type what you’re looking for in plain words — a white van, a door left open — and the moments it appears come back. What used to take an afternoon now takes minutes.
Tell us your problem and we’ll say whether IRIS understands it, or not yet.
A person answers, the same working day.