Average speed cameras on A-roads - how do they work?

facade

I think the OP's question is more "How do the cameras know precisely when to start and stop timing?"

The answer must be "When the numberplate gets to a specific position in the video frame"

I suppose they could start & stop when the numberplate is n pixels wide. Then if the cameras are the same height and at the same angle the numberplate would be the same distance from each camera.

Then they could check calibration by driving through dragging a bicycle wheel to measure distance whilst timing themselves between camera poles, and see how closely the camera matched their average speed.

born_again

Okell said:

Why are you interested in their "capture zone" and their "range"?  Neither really means anything with average speed cameras.

You have two precisely located points - Point A and Point B - on the road which are a known distance apart.  I don't know what the minimum* distance between them must be, but they could be several miles apart.

Camera A simply records your car passing Point A at a precise time, and Camera B records it passing Point B at a precise time.  Because the distance between Points A and B is known to a high level of accuracy, your average speed is easily - and accurately - calculated.  The greater the distance between A and B the greater the accuracy of the calculation.

The distance between Points A and B will vary from one set of cameras to another.  So "capture zone" and "range" are irrelevant.

Perhaps I've misunderstood but you seem to be thinking you have two cameras at different locations which take instantaneous speed readings and then take the average of the two readings.  If you are, that isn't how they work.



*I'd have thought it really needed to be > one mile but it might be less than that.

Wonder if the OP is planning to drive like a F1 driver coming into the pits & having to brake very hard to avoid over speeding in the pit lane Average camera area. 

Aretnap

facade said:

There is, of course, no reason why they can't site a mobile camera within an average speed zone, to catch the people who (as advised by Proper Top Gear) pull over for 5 minutes or whatever just past the first camera so they can drive at 180 through the rest of the zone.

Or to catch motorcyclists who consider themselves to be "exempt" from the forward facing variety.

Car_54

facade said:

I suppose they could start & stop when the numberplate is n pixels wide. Then if the cameras are the same height and at the same angle the numberplate would be the same distance from each camera.

That would work if all plates were the same width. But they're not.

facade

Car_54 said:

facade said:

I suppose they could start & stop when the numberplate is n pixels wide. Then if the cameras are the same height and at the same angle the numberplate would be the same distance from each camera.

That would work if all plates were the same width. But they're not.

It would always work no matter the width of the plate. A square plate would have to be closer to the start camera than a rectangular one to still be n pixels wide, and then the same amount closer to the finish one, so the distance between start and finish points would still be the same.

The camera could be configured with a "hotspot" and time from when the plate enters the hotspot, and rely in the same way on the orientations of the cameras being exactly the same.

If someone crashes into a camera pole and bends it,( or there is ground movement- e.g. earthquake) the camera view would change and the plate would be at different distances from each camera, which is why there really ought to be a fixed marker in camera view, like a line across the road, a diamond on the road, or a little post at the edge of the carriageway, then timing could be between two precise points- the hotspot could be placed next to the marker and re-adjusted when necessary, or dynamically.

Or the cameras could just use LIDAR rangefinders in the first place.

Even with the full power of t'interweb, I can't find what they do, which makes it an interesting question.

B0bbyEwing

400ixl said:

What is the reason behind the question as that may give some clearer answers.

To avoid getting done for speeding, basically. 

I'm not going to lie like some just to avoid some people trying to torch me. Sure I'll speed. I'm not talking about doing 70 in a 30 or going 20 over on blind bends or anything stupid like that but I'll speed for sure, especially if I set off late in a morning for work.

"Just get up on time then". 

Why didn't I think of that.

I'm sure I'll be no different to the majority reading this thread in terms of going over the limit. The difference will be that I have no problems whatsoever in admitting it.

Car_54 said:

Are you sure about the camera ‘facing you’? AFAIK they all face the same way.

Two cameras, set a known distance apart, capture you, and it’s easy to get the average speed from the time difference. Assuming they’re identical cameras, the ‘capture point’ (x metres from the camera) should also be identical. Any practical differences of a few feet in setup will be negligible compared to the distance between the cameras, and indeed in relation to the 110% + 2 enforcement threshold.

Yep I'm sure about some cams facing. Some have a single cam on top of a post, others have two facing opposite directions. 

I know how average cameras work in terms of taking the time travelled between A & B and using the formula to calculate the average speed

I was more wondering whether these things have hawk like vision & can detect you from an absolute mile away. 

Some that have gone up around here, you can see them a good distance away on a long straight. I wasn't sure whether there's a standard distance that they detect you from or not.

user1977 said:

Why are you asking? 

To be informed.

daveyjp said:

Tens of metres as they rely on illumination to read plates in the dark,

Thanks for answering the question.

molerat said:

The capture distance from the actual camera ain't gonna make that much difference on the 31 mile single enforceable section of the A9 !

Good thing I never said it would.

born_again said:

Look at the angle of the camera, that will give you a idea of how far away it will measure from.

But why just A roads? M Ways are just the same.

I rarely travel M ways & when I do I don't hit them, very very very rarely. 

LunaLater said:

Why do you ask?

Because I'd like to know.

Okell said:

Why are you interested in their "capture zone" and their "range"?  Neither really means anything with average speed cameras.

Because there's absolutely nothing wrong with me wanting to know.

As I get to this point in my response, I get the impression people are thinking I'm talking about moving through A-B-C-D.....
Which I never said.

I'm talking about approaching A for example. How far you get picked up.

And as I say, I've passed these cameras where a single one on top of the post faces you. I know for sure what I've seen. Whether it's been installed wrong based on the responses here I couldn't possibly tell you but I know what I've seen.

SiliconChip said:

B0bbyEwing said:

With the static cameras with the white lines painted on the road you knew exactly where the 'capture zone' was. 

Isn't it true that the white lines are now irrelevant as static cameras don't work in the same way as they used to?

Pass.

Are you saying then that the old yellow box on a stick cameras detect beyond the white lines painted in the road then?



Based on page 1 I'm not going to page 2 & 3 as I suspect it'll just be more "why do you want to know", which although irrelevant because I highly doubt my reasoning changes the detection range any.... I've now answered. 

Unless of course, I want to know because of reason X means they detect 100 metres but if I answered reason Y then it suddenly changes the detection range to 500 metres??!

TooManyPoints

I was more wondering whether these things have hawk like vision & can detect you from an absolute mile away. 

No they can't and there would be no point in them doing so if they could. You have said that you understand how they work (i.e. camera A taking a picture of the vehicle at the start of the stretch and Camera B doing so at the end). So they obviously need to capture you at those points (+/- a metre or two, as explained). Capturing you elsewhere would be pointless.

RichardD1970

chrisw said:

I don't think I've ever seen average speed cameras other than on the motorway. Are they just used on higher speed A roads and I presume there are signs saying you are now entering an average speed zone?

There's a set near me on a quarter mile stretch of 30mph road. Roughly staring and ending at the edges of the picture. It was perplexing as to why they ended up there.


 Of course it nothing to do with the fact it's the stretch of road where the councillor responsible for highways lived. 

Okell

B0bbyEwing said:

... I know how average cameras work in terms of taking the time travelled between A & B and using the formula to calculate the average speed

I was more wondering whether these things have hawk like vision & can detect you from an absolute mile away. 

Some that have gone up around here, you can see them a good distance away on a long straight. I wasn't sure whether there's a standard distance that they detect you from or not...

I think the first sentence in the above quote is contradicted by what then follows....(?)

The first camera records your car when it goes past whatever part of the road the camera is aimed at.  The same with the second camera.

While the aiming point on the road must be some distance from the camera lens (I suppose it could reasonably be between perhaps 20m and 50m from the camera) it is a fixed point and won't move.  

If you think about how the average speed calculation works (ie time taken to travel between two fixed points) then it must follow that the cameras do not work within a range and are not detecting your car from 200m, 500m, 1000m or one mile away.  Your car is being detected as it passes the fixed points the start and finish cameras are trained at.  These will be close to the cameras, not far away from them.

LunaLater

You seem to be confused. If you speed, on average, from when you pass one to when you pass another then they will catch you.

Did you think that it was the average of your speed at the two points, rather than between them?