CAM-914
Insects on the dome
A square at night, seen from the pole on the corner. Below there are full terraces, people sitting at the tables, lit shopfronts and, on the right, the road with cars, a motorbike and the crossing. And in front of all of it, stuck to the dome, there are four or five large insects. They are so close to the glass that they never come into focus: they read as shadows with open wings covering the left edge, the bottom centre and the right corner. Half the terrace is behind them. IRIS sees bodies attached to the dome, always in the same spot, not moving with the scene. It does not know what insect it is, or whether it will still be there tomorrow.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- bodies on the dome
The rule
You decide once. IRIS applies it always.
If
bodies stuck to the dome are covering part of the frame and are still in the same place after the time maintenance has set
Then
- alert maintenance with the camera, the time and the video of how the picture ended up
- flag the camera as degraded while the dome stays covered
- log how many nights in a row it has happened, to tell whether it is the night or the location
The instruction that set it up«Tell me if something is stuck to a camera’s glass, covering part of the frame, and it is still there the next night.»
What changes
What changes for whoever answers for the cameras working
Insects come to the infrared light, and the infrared light is on the camera. That is why they land exactly there, and why they come back every summer night. The problem is not the insect: it is that the night’s recording is worthless. If something happens on the terrace at three in the morning, the video exists, but half the frame is a shadow with wings. And nobody finds out until that video is needed, which is the worst possible moment. With IRIS somebody knows the same night, with the camera, the time and the picture. Sometimes moving the infrared lamp fixes it for good; sometimes a cloth does. Both are cheap if done beforehand.
What people usually ask
Will it alert me every summer night, when the insects stick to the light?
Not unless you want it to. You write the rule: you can ask for the alert only when the covered area is still there after sunrise, or only when it repeats several nights running. An insect that settles for a while and leaves is not a fault. One that turns up every night in the same spot is information: it means the infrared lamp should be switched off or moved. IRIS tells the two apart because it measures how long it lasts and how often it comes back.
How does it tell an insect on the glass from something happening in the street?
By distance and by movement. What is stuck to the dome stays out of focus, always in the same part of the frame, and does not travel with the scene. A person or a car is in focus, moves and leaves. IRIS looks at the whole picture and sees an area that has not changed for a long while, while everything else does. That is why the box does not outline an insect: it outlines the part of the frame that has stopped being useful.
So does somebody have to go up and clean the camera every morning?
Almost never. Most of these cases are fixed once: move the infrared lamp away from the dome or reposition it, and it stops happening. What IRIS does is give you what you need to decide: which cameras get covered, in which months and for how many nights in a row. Without that, cleaning runs on a calendar, and the round ends up wiping the camera that was fine while the one that is actually blocked stays blocked.
The real interface, step by step
A new case each time. Always the same screen.
It plays by itself and shows one case from start to finish. It is a summary: the real screen has far more functions than fit in this box.
- Checks the whole system
- Flags what is failing
- Puts it on one screen
Next step
What does an airport see before the queue forms?
14 cases: 7 raise an alert when something happens and 7 only measure. You see each image as it is, and then with what IRIS understands on top.
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.
And the best part
What happens outside the framedoes not exist for IRIS. No camera sees the whole world.
So the first thing we do is check what each camera covers and write it down: from here you can understand this, and not that. A dull hour that saves a lot of misunderstandings.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.