CAM-919
Bird droppings
Midday in a square by the sea, with the sun straight ahead. Everything reads well: people crossing the promenade, the café terrace with its parasols, the planters, the cars pulled up at the kerb. Everything except one corner. On the right-hand side of the frame there is a thick white splat, lumpy, stuck to the glass of the dome: it is not a shadow and not a reflection, it sits in front of the picture. It covers the pedestrian crossing and the opposite pavement completely. IRIS sees that this patch has not changed for hours while the rest of the frame moves, and reports that the camera has lost part of what it was supposed to watch.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- splat on the dome
The rule
You decide once. IRIS applies it always.
If
part of the frame is blocked by something stuck to the dome and is still blocked hours later
Then
- alert maintenance with the camera, the time and the video of the moment it appeared
- say which watched area has gone blind and since when
- log how many times the same thing has happened to that camera this year
The instruction that set it up«Alert me if something sticks to the dome and blocks a piece of the frame, and tell me whether that piece is one of the areas I watch.»
What changes
What changes for whoever answers for the cameras working
It is the silliest fault and one of the most expensive. The camera is fine, the picture is fine, and nobody is going to take it out of service. But the pedestrian crossing it was meant to watch is no longer in the video, and the splat always blocks the same patch: this is not a fault that comes and goes. If somebody is knocked down right there tomorrow, the recording exists and does not show it. With IRIS the alert arrives the same day and says which patch has been lost and whether that patch was one of the watched areas. Somebody goes up to clean that camera, not the thirty in the square. And both dates stay on record: the day it was covered and the day it saw again.
What people usually ask
By the sea this happens nearly every day. Will I not end up ignoring the alert?
It depends on what the alert brings with it. If it is only a light going on, yes, it ends up ignored. This one brings two more things: which part of the frame has been lost, and whether that part was one of the watched areas. A splat on a corner of sky gets logged and waits for the next visit; a splat over the pedestrian crossing gets dealt with today. Same camera, same dirt, two different priorities.
Does it notice that it has been cleaned, or do I have to close the alert by hand?
It closes by itself. IRIS compares what it sees with that camera’s reference: when the blocked patch starts moving and giving detail like the rest again, the alert closes with its own date and time. Nobody has to remember to mark it as resolved. And both dates end up side by side, the one it got covered and the one it got cleaned, which is exactly what you need in order to know whether the cleaning round arrives in time or late.
Is there any point in knowing that the same thing keeps happening to one camera?
It is useful because it lets you stop cleaning it. If one camera turns up covered every couple of weeks and the ones beside it do not, the problem is no longer cleaning: there is a perch for birds right above it. That does not get fixed by going up more often; it gets fixed with a spike or a different mount. IRIS decides none of that: it just counts how many times it has happened to each camera. That count asks the question by itself.
The real interface, step by step
You set the rule once. IRIS applies it every time.
You will see a summary of the interface, played step by step and hands-free. Each round starts with another case. The complete tool does not fit in a demo.
- The rule gets built
- It goes to every camera
- Published, it watches alone
- And then it fires
Next step
Who is watching a whole quay at three in the morning?
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.
Who is watching the cameras of a whole branch network?
13 cases: 12 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.
And the best part
There are nights of thick fogwhen the image gives you nothing. We do not dress that up.
Backlight, heavy rain, a street lamp that has blown. IRIS works with what the camera puts in front of it; when there is no image, it says so instead of inventing an answer.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.