{"id":1524,"date":"2017-03-27T15:52:31","date_gmt":"2017-03-27T13:52:31","guid":{"rendered":"http:\/\/www.asuralpr.com\/?p=1"},"modified":"2017-03-27T15:52:31","modified_gmt":"2017-03-27T13:52:31","slug":"lpr-systems-working-150-200-250-kmh-speed","status":"publish","type":"post","link":"https:\/\/asuratechnologies.com\/de\/2017\/03\/27\/lpr-systems-working-150-200-250-kmh-speed\/","title":{"rendered":"LPR systems working up to 150-200-250 km\/h speed"},"content":{"rendered":"

Marketing materials often claim that an LPR system can read license plates of vehicles traveling at high speed e.g. the motorway speed limit. Is this something for an LPR system to be proud of or gibberish?<\/em>
\nWhat is the maximum traffic speed that an LPR system can handle? What does this parameter depend on? In any case, what is the requirement for reading license plates of vehicles at a certain speed? Read on and you will understand.<\/em><\/p>\n

In order to read a license plate, you will need\u2026 image(s) of the license plate of the vehicle in proper quality.<\/strong><\/p>\n

And that’s pretty much it. But how you can achieve this? First determine how many images should be taken of each vehicle. Is one image enough? Or do you prefer running a validation process requiring 2 or even more images of each car?<\/p>\n

And that’s pretty much it. But how you can achieve this? First determine how many images should be taken of each vehicle. Is one image enough? Or do you prefer running a validation process requiring 2 or even more images of each car?<\/p>\n

Of course you need to be able to read the license plates in all these images so
\n– the images must contain the complete license plate
\n– the license plate must be clear, sharp and readable<\/p>\n

The first requirement is related to camera<\/strong> positioning<\/strong>. The second is related with the shutter speed of the<\/strong> camera<\/strong>. For now, assume the camera has suitable shutter speed. Note that both requirements are related only with the camera.<\/p>\n

Let\u2019s see a quick calculation:
\nAssume there is a camera which covers a distance (s), and takes images at \u201cf\u201d FPS. So the license plate of the vehicle travels \u201cs\u201d meters while it is still in the camera’s FOV (field of view). What is the maximum speed for a vehicle in order to have at least n images containing its license plate?<\/p>\n

What you know:
\n“n” is the number of images taken of the vehicles.
\n“f” is the number of images taken by the camera in a second.
\n“s” is the distance covered by the camera (length of the FOV).
\n“v” is the maximum speed that you look for.<\/p>\n

The time elapsed between the capture of two consequent images is 1\/f seconds, the maximum distance the vehicle may travel during this time is s\/n meters.<\/p>\n

So the maximum speed is v = (s\/n)\/(1\/f) m\/s = 3.6*s*f\/n km\/h.<\/p>\n

Calculating with usual values for a 1.3MP camera:
\ns = 10m
\nf = 30 FPS
\nn = 1
\nv = 1080 km\/h
\nEven if n=3 (ensures good input for result validation), v=360 km\/h which is still fairly good.<\/p>\n

\"\"<\/p>\n

I think the result speaks for itself and you can ask again: do you really need cameras that take 100 FPS or even more for your LPR system.<\/p>\n

\n

Don\u2019t forget to sign up for our newsletter!
\nYou will receive first hand reviews on Asura services and LPR related news and articles!<\/p>\n

<\/p>\n