Skip to content
IRIS NEURALContact

Solutions by typeEnvironment and cleanliness

How many times a month is that corner like this, and since when?

The bin was full, so the sidewalk became the bin

It’s 7:15 a.m. on a Monday. The truck came on Saturday at 5:40 a.m. and isn’t back until tomorrow. The trash cans are full and around them sit eighteen hours of things left beside them: open cardboard boxes, tied bags, a loose pile that already takes up half the sidewalk and the corner of the crossing. Someone with a stroller steps into the road to get around it. The intersection camera has watched it grow all weekend.

12 cameras across 6 sectors

This is how you ask for it

«Tell me if trash is piling up outside the trash cans and how long it has been there.»

A pile like that isn’t paid for in one go: the city pays for it in installments. The street crew sweeps where the calendar says, not where it’s needed, and that corner gets cleaned two days late. Meanwhile the neighbors phone in, the photos go online and the complaint reaches the councillor. And at the end of the month nobody can say whether that corner is a one-off or a weekly problem.

You decide once

You say which stretch of sidewalk has to stay clear around each bin island, how long something can sit there and what happens then: if there are items outside the bin at the intersection island for more than two hours, a job is opened for the cleaning service with the camera attached; if they also block the corner of the crossing, the alert goes out right away. And every day, without anyone asking, it’s recorded how many hours each island spent like that.

A full trash can is an invitation. The first person who can’t elevator the lid leaves the box beside it, propped up, with the best intentions in the world. By the fourth bag it is nobody’s slip anymore: it’s a dumping spot that created itself, and it’ll happen again next Saturday at the same hour, because what failed isn’t the people, it’s that the trash can was full when it should have been empty.

The same thing happens where nobody’s looking. At a waste plant, a dumpster filled to the rim and left uncovered sheds bags and film with the first north wind; they cross the mesh fence and end up in the ditch and on the neighbor’s sidewalk, and the neighbor is the one who calls. It’s the same scene: something that had a place and is no longer in it, in a route people walk, growing slowly and without a sound.

IRIS watches the space that has to stay clear — the sidewalk around the island, the strip in front of the fence, the aisle by the cell — and says two things. First, that something’s outside it and what minute it got there. The second is the one that really changes next year: how many hours a month that spot has looked like this. With the first you clean today; with the second you argue about collection frequency with a number on the table, instead of with the photo a resident sent in this morning.

What changes

This used to arrive by phone. Nobody knew how long it had been there, and at the end of the year collection frequency was argued over with impressions.

Now the alert comes from the intersection itself, with the time the first item appeared and the camera to see it. And every trash can island has its own record: how many hours a month it overflows, which days and at what time. The crew goes where it’s needed, and the next contract is negotiated with figures.

Where we have seen it

  • It’s the middle of the night and the square is full: one large group in the center and smaller ones spread around it. There are bottles across the ground, the trash cans are overflowing and the patio at the back is still open.

    CAM-22Night gathering in the square

    «Tell me if people gather in the square above the threshold at night, and how full the trash cans are.»

    See this cameraThe city

  • Early afternoon in machining. You can barely see the floor: parts racks full of parts are stacked everywhere, in front of the machines and across the walkways. At the bench at the back, six people are reworking parts by hand, one after another. The outbound conveyor, meanwhile, runs almost empty. IRIS counts how many parts racks are in the area, compares that with the count at the same hour yesterday, and times how long the pile has been growing.

    CAM-248The rework area has swallowed the aisle

    «Tell me if, mid-afternoon, there are more parts racks than usual in the rework area.»

    See this cameraIndustry

  • A car parked at an angle right across the substation gate and the technical door beside it. This is an ordinary street, with traffic, sidewalks and a crossing. The car has done nothing but park there. The problem is that this is where a truck has to get in on the day a machine has to be replaced.

    CAM-160A car parked across the access

    «Tell me if anything stays parked across the gate for more than a few minutes: a car, a dumpster or a pile of materials, it doesn’t matter.»

    See this cameraCritical infrastructure

  • Mid-afternoon in the container block. On the bottom tier, a door is open and the cargo in boxes can be seen inside. Two people are standing at the doorway. A truck passes down the aisle without stopping. That row has no operation open at this hour. IRIS detects the open door, the position of the container in the stack and the exact minute, and compares that time with the working window the port has set for the block.

    CAM-444A container opened outside the working window

    «Tell me if a container door is opened in the block outside operating hours.»

    See this cameraPorts

  • A windy morning at the north fence. A metal dumpster filled to the brim and left uncovered is shedding bags and film, which clear the chain-link fence and land on the other side. The sidewalk and the gutter are carpeted with plastic right up to the curb, and a car and a van are going past on the road. Inside, the stacked bales and a forklift carry on working. IRIS sees light waste leaving the site toward the public road.

    CAM-436The bags clear the fence and end up on the neighbor’s sidewalk

    «Tell me if plastics start blowing over the fence toward the street.»

    See this cameraIndustry

  • Mid-afternoon at a battery site: six white containers on gravel, fans along the roof. On the front one the door has been blown off its latch and orange flame shows through the opening; a column of black smoke rises from the roof and already covers half the site. The other five are untouched and nobody is inside the fence. IRIS sees the smoke and the blown-open door. What happens inside the cells is reported by the battery management system, and the gases by the detector.

    CAM-353Black smoke pouring out of a battery container

    «Alert me right away if smoke or flame comes out of a battery container, or if a door opens on its own.»

    See this cameraIndustry

  • The trash cans are full and the trash has been left beside them: boxes, bags and loose waste now cover the sidewalk and part of the road. The corner of the crossing is blocked.

    CAM-24Waste dumped outside the trash can

    «Tell me if trash is piling up outside the trash cans, and how long it has been there.»

    See this cameraThe city

  • Early morning at the robot cell. The arm has stopped in the middle with nothing to pick. In front of the fence the big gray bin hasn’t a single part in it: you can see the bottom. The steel hopper beside it’s empty too, and so is the bin at the back next to the pallet. An operator stands looking at the cell, hands on his hips, waiting for material. IRIS times from the moment the cell ran out of anything to process.

    CAM-243The cell is ready and the bin is empty

    «Tell me if the bin at the cell runs empty and the robot goes more than five minutes without putting out a part.»

    See this cameraIndustry

  • The mouth of the spillway seen from above. Against the screens of the bays on the right there’s a raft of logs, branches and a bit of plastic, crossed over one another and resting on the concrete; underneath, white water is still coming through. The bay next to it has a clear screen and the water goes in unobstructed. On the walkway above you can see the gate hoists. Nobody is in frame.

    CAM-146Logs piling up against the spillway

    «Tell me when wood or trash builds up against the spillway screens, and tell me whether the pile is still growing.»

    See this cameraCritical infrastructure

  • Early afternoon in the reefer row. One of the containers has both doors open, with the empty interior in plain view. In front of it, a power cable is lying loose on the ground, out of its socket. A technician in a hard hat and safety vest stands beside it with a tablet. The limit is worth stating right here: the temperature comes from the container’s own unit, not from the camera. IRIS sees the open door, the cable on the ground and how long it’s been like that.

    CAM-452A reefer with its doors open and the cable on the ground

    «Tell me when a reefer has been more than five minutes with a door open or with its cable out of the socket.»

    See this cameraPorts

What it doesn’t do

IRIS does not say who left the bag. It does not identify people and does not recognize faces: it says something is outside the bin, what minute it got there and which camera sees it. Whether that adds up to a case is decided by a person, under your city’s rules.

And if the pile is behind the bin itself, hidden from the camera, it is not there.

Questions

Can it be used to fine whoever leaves the bag outside?
No. IRIS reports that trash is outside the trash can, what minute it got there and which camera sees it. It does not identify anyone, does not recognize faces and does not open cases. If a city wants to go further, that’s its own decision, under its own rules and with a person looking at the video. And in practice the saving isn’t in the fine: it’s that the corner gets cleared the same day instead of two days later, and that you know which corners keep coming back.
Can it count how many times a month each trash can island overflows?
Yes, and for a cleaning service this is worth more than the alert. Every island builds up its own hours: how many times a month there are items outside, which days of the week, what time it starts. That shows right away what everybody suspected but couldn’t prove: that three spots always overflow on Sunday afternoon and the rest of the city doesn’t have that problem. Asking is for searching; counting is for measuring, and to measure you draw the area on the image rather than describe it.
Does it tell a bag from a mattress or a pile of cardboard?
Yes, well enough for the decision at hand. IRIS separates a bulky item — a mattress, a sofa, a builder’s debris bag — from a heap of bags or stacked cardboard, because they aren’t collected the same way or by the same crew. What it doesn’t do is classify the waste stream or say whether the cardboard was properly flattened: that can’t be seen from a light pole. For the alert it is enough to know that something’s outside, roughly how big it is and how long it’s been there.
Does it alert every single time somebody leaves a bag?
No, unless you ask for that. An alert per bag would be noise, and within three days nobody would read them. The normal setup is a threshold: how much space taken up, or how long you tolerate it, before it counts as an alert, and during which hours. One bag at eleven at night, a quarter of an hour before the truck comes, isn’t a problem. The same bag on Sunday at ten in the morning, with two boxes on top and no truck until Monday, is. You decide that difference once, and IRIS applies it always.

See all the questions

Demo · IRIS at work

A new case each time. Always the same screen.

It plays by itself and shows one case from start to finish. It’s a summary: the real screen has far more functions than fit in this box.

  1. Cameras on the map
  2. You draw an area
  3. What passed shows up
IRIS NEURAL
30ES
IRISDirectoGrabacionesGISIncidencias3996SituaciónCasosLPRAutomatizacionesIRIS DATA
Ask a question or make a request…Send
IRIS · Infinity Neural

Not seeing yours here?

These words are the ones we have built, not the ones IRIS understands. The list falls short on purpose: you ask for it in a sentence, and sentences don’t run out. Tell us what is in front of your cameras and we’ll tell you whether it understands it, or not yet.

See all solutionsTell us what you see

And the best part

Nobody was hired to stare at a screen.They were hired to know what to do.

Watching is what a machine does best; deciding is what only a person does. IRIS watches all four thousand at once and hands over only what calls for a decision.

Tell us your problem and we’ll say whether IRIS understands it, or not yet.

A person answers, the same working day.