CAM-287
Three vials have left the capper with the cap wrong
Sterile packaging room. On the accumulation table dozens of glass vials with blue caps wait, already closed, before entering the belt. Three are not right: one has no cap at all and its neck is open, one has it crooked and a third has it only half seated. An operator in blue coveralls, hood and mask has picked up a vial and is looking at it closely. To the right the capper keeps turning. IRIS sees the cap that is missing and the one that is not straight.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- vial with no cap
- vial with a crooked cap
- vial with a half-seated cap
- operator checking a vial
- capper at the end of the table
- line control screen
What this camera measures
An alert says something is happening. The reading says how much.
- 3Vials flagged
- 96Vials on the table
- 2 min 40 sTime since the first flagged vial
- 840Vials passed in the last hour
- 10:18Time 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
a vial leaves the capper with no cap, or with the cap tilted or half seated
Then
- alert the line operator with the camera, the time and the image of the vial
- save the video of the vial from the moment it left the capper
- count how many vials like that turn up across the shift
The instruction that set it up«Tell me if a vial leaves the capper with no cap or with the cap crooked, and save me the image.»
What changes
What changes for whoever answers for the plant
The limit comes first: the integrity of a closure is proven by a test, not by a camera. What the camera sees is the cap that is missing, the crooked one and the one left half in place, and it sees them while the vials are still on the table. For the line manager that changes when the problem is found: pull three units now, or hunt for the fault once the batch is already in the warehouse.
What people usually ask
Does IRIS confirm that the vial closure is tight?
No, and it must not be claimed. The integrity of a closure is demonstrated by a laboratory test. What the camera sees is what is visible: a cap that is missing, one that is tilted and one left half seated. It is an early warning on the line, with image and time, not a quality result and not a substitute for the test.
How many vials can it watch at once?
As many as fit in the shot. On this table there are close to a hundred vials in view and the camera looks at all of them in every frame, the same in the first hour of the shift as in the seventh. That does not replace the line inspection: it goes alongside it, flagging what stands out to the naked eye before the tray moves on.
Who decides what counts as a badly placed cap?
The plant does. IRIS does not write the rules: the line defines what tilt is acceptable, what gets pulled and who gets told. The system applies that rule to the image and hands over the evidence, which is the camera, the time and the video. The decision about the product is always taken by a person, and the record stays for the batch review.
Other cameras
- CAM-280 · Through the gowning room with no mask and street shoes on
- CAM-281 · Both airlock doors open at the same time
- CAM-282 · Eight people in the room and the routes crossing
- CAM-283 · The previous batch is still on the table
- CAM-284 · A green-lidded drum where it does not belong
- CAM-285 · A pink carton moving among identical blue ones
The real interface, step by step
We won't just tell you. We'll show you.
It is a summary of the real screen: it plays by itself, step by step, and shows only the path of one case. The actual tool has far more than fits in here.
- You ask
- IRIS searches
- It shows you
Next step
What can a camera see without pointing at anyone?
13 cases: 9 raise an alert when something happens and 4 only measure. You see each image as it is, and then with what IRIS understands on top.
What if the outage does not start with an attack?
13 cases, and in every one of them IRIS raises an alert when something happens. You see each image as it is, and then with what it understands on top.
And the best part
Stop scrubbing through six hours.Ask what happened and watch only that.
You type what you are 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 will say whether IRIS understands it, or not yet.
We reply the same working day.