CAM-57
Someone climbing the site fence
A fenced building site in the middle of the city, seen from above. At the corner of the timber hoarding a person is halfway up: one foot against the boards, both hands on the top edge. Inside you can see the debris skip, the churned-up ground and, further back, two workers in helmets and hi-vis. IRIS sees that a person is crossing the fence; it does not know who they are.
Turn on IRIS Vision
AI-generated image
What IRIS understands
What it recognises in this scene
- climbing the fence
- site hoarding
- debris skip
- workers inside
- barrier on the pavement
The rule
You decide once. IRIS applies it always.
If
a person crosses the perimeter fence somewhere that is not the gate
Then
- alert whoever is on duty right away, with the camera and the point of the fence they came through
- keep the video from before and after, to see where they went once inside the site
- log the point on the fence, because a way in that works once gets used again
The instruction that set it up«Tell me if someone climbs the fence or gets in where nobody is meant to, whatever time it is, and show me the camera so I know where they came through.»
What changes
What changes for whoever answers for the site
A site that has stopped for the day is not empty: it has open shafts, trenches, machines and stacked material. Whoever climbs the fence almost never knows what they are walking into, and often it is a teenager who came in for a look. Today you find out the next morning, when someone sees the fence pushed aside or notices material missing; if something worse happened, you find that out just as late. With IRIS, whoever is on duty gets the alert while it is happening, with the camera and the exact point on the fence. There is time to talk to the person who came in, to call whoever needs calling, and to make sure nobody is left inside a dark site. IRIS closes nothing and stops nothing: it warns, and a person decides.
What people usually ask
Does IRIS know who the person climbing the fence is?
No. IRIS does not know who it is: it detects that a person is crossing the fence, and nothing else. It does not recognise faces, it does not compare them against any database and it does not store who anyone is. The alert says what happened, on which camera and at what time, and that is enough for whoever is on duty to go and look. Putting a name to someone is not the camera's job.
Won't the alert fire for a cat or a plastic bag blowing past?
The alert does not fire for a cat or a plastic bag, and the reason is that the rule gets tuned. It looks at the size of what crosses, where it crosses and how long it stays inside, and those three things are adjusted with the site running during the first days. That is the most important part of the installation: if the alert arrives a hundred times a day, whoever is on duty switches it off within a week and then the camera is worth nothing.
And during the day, with the site working, doesn't it alert constantly?
During the day, with the site working, the rule is not the same as at night. The fence is watched in time bands: during working hours it only alerts on someone crossing where nobody should cross, and outside working hours it alerts on anyone entering the perimeter. It is the same camera with two different rules depending on the hour, and someone walking in through the gate at seven in the morning raises nothing. The site sets the time bands, not us.
Every simulation
Someone on the ground
Tell me if someone stays on the ground for more than twenty seconds and does not get up.
- The sentence
- Where it applies
- What it finds
- The same rule
Next step
Who sees the sales that never reach the till?
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.
Who is watching when the load swings over somebody?
11 cases: 9 raise an alert when something happens and 2 only measure. You see each image as it is, and then with what IRIS understands on top.
And the best part
A wall of monitors is impressive.Nobody can read forty tiles at once.
The more tiles there are, the less each one gets looked at. IRIS watches all four thousand cameras at once and points at a single one: the one that needs someone right now.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.