Xentim Book a demo
Parking and traffic

How many spaces are free? The camera already knows

A driver circles the car park for the third time, because the sign at the entrance says “free”, while the free spaces are two at the far end of level minus two. A sensor in every bay would solve the problem, but it costs money and means breaking up the floor. The camera has been hanging there for years.

The Xentim team ·
Parking lot from a high camera: a counter of occupied and free spaces
Illustrative visualization: a frame generated for this article, with boxes and labels in the Xentim style. This is not footage from a client site.

How the Danes did it

At the Køge Nord railway station near Copenhagen, 25 cameras serve a car park of 650 spaces. A single camera sees up to 90 bays. The driver gets information about free spaces on signs by the car park, in the town and by the motorway. The vendor reports accuracy above 98 percent. No one drilled a single hole in the surface.

Free spaces in your car park

We mark the bays on the image, and every few seconds the system checks which are occupied. The result is a simple number for each level or sector, which can be shown on a sign at the entrance, in an app or in a report for the operator.

how many spaces were free on level -1 on Friday at 17:00

After a month you know which hours the car park is full, which sectors stand empty and whether it pays to set aside spaces for season tickets.

Congestion at the entrance and at the gate

The same counting works for a queue. A line of cars before a shopping centre barrier, or trucks before a plant gate, is a loss on both sides: the customer drives off, the driver loses the unloading window. When there are more vehicles in the approach zone than we set, and they stand longer than a few minutes, staff get a signal to open a second lane or route traffic through another gate.

Operational alert15:22, camera Main gate. 9 trucks in the approach zone, the first standing for 11 minutes.

A vehicle that stopped where it should not

In Norwegian tunnels, cameras have detected stopped vehicles, pedestrians and wrong-way driving since the 1990s, because a call from a driver usually comes in a few minutes after the event. In Colorado, sensors on motorway ramps notify the traffic centre 2 to 3 seconds after a wrong-way entry, and in the first months of testing they helped prevent four such events. There thermal imaging was used, which is worth knowing: in bad weather an ordinary camera sees worse.

At the scale of a site, the same mechanism watches the garage entrance, the exit ramp and the emergency lane.

What you get

Revenue

A customer who finds a space right away walks into the shop instead of driving off to a competitor. Occupancy data lets you price season tickets and peak hours sensibly.

Savings

No sensor in every bay and no breaking up the floor. In Køge, one camera per about 26 spaces was enough.

Safety

A stopped vehicle in a tunnel, on a ramp or going the wrong way, reported in seconds, not after a call from another driver.

What can go wrong

A pillar hiding half the bays, a camera at a sharp angle, snow on the lines, a night with no lighting. Each of these lowers accuracy, and none of them shows in a sales brochure. That is why we first count spaces on your footage and compare the result with what the naked eye sees.

This is a deployment scenario. We set the rules on your cameras and verify them during a two-week trial, on real footage, at different times of day. Only then do we tell you what works and how well.

See it on your own footage

The demo takes up to an hour and runs on footage from your cameras.

Book a demo
Read also
The case was dropped because no one had time to watch the footageNight shift, two people, forty camerasA sofa bed by the bins and four days of footageThe trail camera took the photo. Who will look at it?A car stalls on the tracks. The signaller knows within secondsYou can see the smoke before the detector under the roof senses itThe security queue builds at 5:40, and only two lanes are open