CAM-581
Eighteen minutes in one line, four in the one beside it
Check-in hall at six thirty. On the left, the line of staffed desks with their signs above; on the right, the bank of self-service machines for dropping a bag. In front, a belted snake with two hundred people inside it. The thing is, the two queues are nothing alike: the desk queue is running at eighteen minutes and the self-service one at four. And five desk positions have their screens turned off. IRIS measures the wait in each queue, how many people are inside it and how many positions are open while people wait.

Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognizes in this scene
- staffed desks
- self-service bag drop machines
- line to the desks
- line to self-service
- staff behind the desk
- closed desk position
- passenger waiting in line
- apron seen through the window
What this camera gives back
Nobody gets an alert here. Here you get numbers.
- 18 min+6 minWait in the desk line
- 4 minWait in the self-service line
- 210People in both lines
- 6 / 11Desk positions open
- 9 / minPassengers served per minute
- 06:40Last updated
The figures on this screen are an example from the illustration, not any client’s measurement.
How the counting works
Counting isn’t something you ask for. You draw a line.
Each queue has two lines: one at the mouth, where somebody joins it, and one right at the desk, where they’re served. The time between the two is the wait. And there’s a zone painted over the floor of each queue, telling you how many people are inside it at any moment. What IRIS follows is a silhouette within the frame, with no name and no face, and that trace is lost the moment it leaves the image. Neither the lines nor the zones are asked for in words: they’re drawn once and there they stay, so a Monday in August and a Tuesday in February are measured the same way.
- line at the mouth of the desk queue · Counting line
- line at the mouth of the self-service queue · Counting line
- desk queue zone · Dwell zone
- self-service queue zone · Dwell zone
And also, one rule
You decide once. IRIS applies it always.
If
a line goes over fifteen minutes of waiting
And alsothere are still desks closed on that same line
Then
- alert the hall supervisor with the line, the minutes and the open positions
- show the self-service wait next to it, so people can be redirected
- log the episode for the report on desks open per hour
The instruction that set it up«Tell me if a line goes over fifteen minutes and there are still desks closed.»
What changes
What changes for operations, for the passenger and for maintenance
The hall supervisor doesn’t need to be told there’s a line: he can see it. What he needs to know at six thirty is how many minutes it really is, in which line, and how many positions he still has closed. With that he opens another one, sends someone to the mouth of the line to move people to self-service, and sorts out in five minutes what otherwise takes forty. And at the end of the month he has the curve per time slot to agree with the airlines how many desks have to be open at each hour, with a measured number instead of an argument.
What people usually ask
How does IRIS know how many minutes the wait is?
With two lines drawn on the image: one where you join the queue and one where you’re served. The time between the two is the wait, and because the lines are always drawn in the same place, today’s figure and the one three months from now are measured the same way. That’s the difference between measuring and estimating: nobody has to time anything or ask anyone how long they have been there.
Does it recognize passengers in order to follow them through the line?
No. IRIS does not recognize faces and does not know who anybody is. What it follows is a silhouette inside this frame while it’s in view, with no name and no identifying feature, and that trace is lost as soon as the person leaves the image. There’s no record of anybody, nothing is matched against any list, and the route of a particular person through the airport can’t be reconstructed.
If I open another desk, does the figure pick it up?
Yes, and that’s why the number of open positions is counted too. A waiting figure without that beside it’s no use for deciding anything: fifteen minutes with eleven desks open and fifteen minutes with six are two different problems and they’re fixed differently. The wait and the open positions are measured together and shown together.
Other cameras
- CAM-580 · The curb jams before the terminal does
- CAM-582 · How long the security wait really is, measured not guessed
- CAM-583 · An open lane nobody has gone through in nine minutes
- CAM-584 · Fifty-two minutes of line and seven booths turned off
- CAM-585 · The aisle is clear and thirty feet on, nobody moves
- CAM-586 · Three minutes standing at the board isn’t walking slowly
Demo · IRIS at work
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 don’t see are the other screens, and there are many.
- Checks the whole system
- Flags what is failing
- Puts it on one screen
Next step
The school
Who sees the child who’s fallen at the far end of the schoolyard?
14 cases: 13 raise an alert when something happens and one only measures. You see each image as it is, and then with what IRIS understands on top.
Critical infrastructure
Who is watching a site where there’s almost never anybody?
44 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
A whole afternoon in front of a screennow fits into one written sentence.
The video hasn’t changed: it’s all still there, in the same place. What has changed is that you can ask it questions, and the answer comes with its camera and its time.
Tell us your problem and we’ll say whether IRIS understands it, or not yet.
A person answers, the same working day.