How much resolution is really needed for LPR

[Nidstang]

We have folks come here all the time thinking that more MP is better, whether it be for general purposes or for LPR.

Those of us that have been around long enough know that sensor size is more important than MP.

Those that have been here awhile know that I share a representative sample of plates I get at night of vehicles traveling about 45MPH at 175 feet from my Z12E that is on the 2nd story soffit, My angle is about 40 degrees vertical, 50 degrees horizontal. Camera is 35 feet above street at this location.

View attachment 119342

So I was playing tonight and wanted to see how low resolution could you go before a 3rd party plate reader couldn't read it.

I am shocked, but this is D1 resolution substream from the Z12E at 175 feet away and the 3rd party plate reader has read every plate the same as the 2MP stream. The 2MP is a little more defined, but with the idea of getting plates is to bounce IR off a reflective plate, the differences are not as great as we may think.


View attachment 119340

Now I can see why OpenALPR says in many instance 720P is more than enough to read plates and saves a ton of CPU processing compared to 2MP or 4MP.

Hey wittaj and others. I'm new here, and have been reading the forums for a couple of weeks trying to learn everything Apologies for reviving an older thread but I'd like to clarify a couple of things which might also help other noobs like me.

1. You mention sensor size is more important than MP, but the recommended Z12E camera for LPR has the smallest sensor from the generally recommended cameras here. Is it more correct to chase PPM (pixels per m) / PPF (pixels per feet) at a given distance? That is how I've been comparing cameras for LPR but want to make sure I'm not missing something.

2. Entering your details into the JVSG calculator, and assuming a 60mm focal length, I get 402 ppm (123 ppf) for your first pic. Is this correct?

3. Changing the camera resolution in JVSG to 640x480 (closest to D1 that it has) gives about 134 ppm (41 ppf). Would you say that is an OK approximation?

Based on richardgohths post for LPR on an Australian Plate (12mm stroke):
minimum 2 pixels per stroke width = 168 ppm (51 ppf) - this roughly lines up with your D1 resolution
desirable 3 pixels per stroke width = 250 ppm (76 ppf)
Australian Standard (2016) recommendation is 410 ppm (125 ppf) - @60mm focal length this lines up with your 2MP resolution

4. The reason I'd like to clarify these figures, is that I'm trying to set the minimum PPM at which to compare LPR cameras, and I can't achieve 400 ppm with the Z12E. I have the following options:
a. Z12E (2MP) the most I'll achieve is 384 ppm (117 ppf) with my parameters
b. B5842-Z4E S3 which costs 15% more will achieve 489 ppm (149 ppf)
My question is would you still recommend the Z12E with the lower PPM, and for what reasons, in terms of night performance, cpu usage, etc? I'm assuming the higher PPM option won't add any value?

Thanks!

 

[wittaj]

Hey wittaj and others. I'm new here, and have been reading the forums for a couple of weeks trying to learn everything Apologies for reviving an older thread but I'd like to clarify a couple of things which might also help other noobs like me.

1. You mention sensor size is more important than MP, but the recommended Z12E camera for LPR has the smallest sensor from the generally recommended cameras here. Is it more correct to chase PPM (pixels per m) / PPF (pixels per feet) at a given distance? That is how I've been comparing cameras for LPR but want to make sure I'm not missing something.

2. Entering your details into the JVSG calculator, and assuming a 60mm focal length, I get 402 ppm (123 ppf) for your first pic. Is this correct?

3. Changing the camera resolution in JVSG to 640x480 (closest to D1 that it has) gives about 134 ppm (41 ppf). Would you say that is an OK approximation?

Based on richardgohths post for LPR on an Australian Plate (12mm stroke):
minimum 2 pixels per stroke width = 168 ppm (51 ppf) - this roughly lines up with your D1 resolution
desirable 3 pixels per stroke width = 250 ppm (76 ppf)
Australian Standard (2016) recommendation is 410 ppm (125 ppf) - @60mm focal length this lines up with your 2MP resolution

4. The reason I'd like to clarify these figures, is that I'm trying to set the minimum PPM at which to compare LPR cameras, and I can't achieve 400 ppm with the Z12E. I have the following options:
a. Z12E (2MP) the most I'll achieve is 384 ppm (117 ppf) with my parameters
b. B5842-Z4E S3 which costs 15% more will achieve 489 ppm (149 ppf)
My question is would you still recommend the Z12E with the lower PPM, and for what reasons, in terms of night performance, cpu usage, etc? I'm assuming the higher PPM option won't add any value?

Thanks!

The concept of "chase sensor size" is as it relates to the MP/sensor ratio. A 2MP on the ideal sensor size will outperform a higher MP camera on a smaller than ideal sensor size. See this chart:



While the ppm/ppf is a good tool, it is just a tool. I can tell you that any images you see from these tool calculators after about 25 feet are not really good representations of what you would actually see.

For reliable LPR the idea is to get the camera with the strongest (largest/highest) focal length to OPTICALLY zoom to make the field of view as tight as possible to make the plate as large as possible.

The Z12E is good to up about 175-200 feet, although some have stretched it to 200 feet.

The Z4E is good to about 50-60 feet. Many people here that are around 60 feet bought the Z4E and then replaced it with the lower MP Z12E due to the more zoom capability of the Z12E.

For LPR, many multi-billion dollar companies use LPR cameras that are 720P, so more MP isn't necessary for plates. 2MP is more than enough.

And that is why I created this thread in that 3rd party plate readers could read D1 resolution if the plate is large enough!

That camera has one job and that is to capture plates. Get a camera that allows the shutter to be set fast, with enough focal length for the distance you are covering, and the correct angle and you are set.

Now being from AUS, you all have big @$$ plates the size of bumpers, so the distance can be stretched a bit. But you will find that the bigger you make the plate, the better chance you have to read the plate. I have several Z12E and Z4E used in a variety of situations, and I will pick the Z12E every time for plates.

 

[samplenhold]

That camera has one job and that is to capture plates. Get a camera that allows the shutter to be set fast, with enough focal length for the distance you are covering, and the correct angle and you are set.

This right here ^

All of the guidelines that we state about sensor size and pixel density are just that, guidelines. They are really meant 'in general', for facial ID purposes, with movement, in low light. They are not meant explicitly for LPR uses.

But for LPR, in low light, most will be running fast shutter speed with IR and high zoom. So one needs to pick the cam that can do that. The first question should be 'how much zoom do I need'. The next question is 'what cams deliver that amount of zoom'. 'What are their costs' and ''what cost suits my needs'.

For most of us we need the optical zoom amounts of the Z12E. The B5241 Z12E fits that bill at a price point that most will find reasonable. While the B5442 Z4E has a larger sensor, it does not have the optical zoom that the Z12E has. But if that is all you need, then it is a great cam for that purpose.

 

[Flintstone61]

also available for LPR is the 5442z z4e

 

[Flintstone61]

Just putting out some examples to consider.

 

[Hans*2]

Opname in de nacht met de SD6CE225U-HNI en ik moet zeggen dat ik vind het nog knap