I am attempting to zoom in on a cloned camera in order to enhance the functionality of the new license plate recognition (LPR) feature. Since the camera does not have a proper zoom feature, I am using the Area of Interest (AOI) for the cloned camera. I want to keep the master camera in full view for security purposes. Currently, I am testing the motion trigger on this zoomed camera using recorded video to see if motion triggers while passing through designated zones. However, the issue is that I am unable to properly test motion, as the recorded video seems to show the full view of the camera and not the zoomed in AOI. As a result, I am unable to test the zones that have been defined in the AOI as the motion is shifted and related to the wrong area. Am I missing something? Is there a way to test motion of an AOI'd camera? Thank you
Testing Motion with Area of Interest
- Thread starter Il_Pres
- Start date
Have you tested this day and night to confirm you can actually read a plate by digital zooming in?
That isn't the way LPR works....except on TV. You might get away with a little digital zoom, but not much.
You need optical zoom not digital zoom. You might get away with a little digital, but not much. How far away would the vehicle be from the camera?
We need higher focal length cameras to make the field of view basically be the size of a car.
Regarding plates, keep in mind that this is a camera dedicated to plates and not an overview camera also. It is as much an art as it is a science. You will need two cameras. For LPR we need to OPTICALLY zoom in tight to make the plate as large as possible. For most of us, all you see is the not much more than a vehicle in the entire frame. Now maybe in the right location during the day it might be able to see some other things, but not at night.
At night, we have to run a very fast shutter speed (1/2,000) and in B/W with IR and the image will be black. All you will see are head/tail lights and the plate. Some people can get away with color if they have enough street lights, but most of us cannot. Here is a representative sample of plates I get at night of vehicles traveling about 45MPH at 175 feet from my 2MP 5241-Z12E camera (that is all that is needed for plates):
See the LPR subforum.
The only way to test the AOI is by live video of objects going thru that AOI camera and then playing back that AOI camera. As you are seeing, you cannot use previously recorded video to test it.
That isn't the way LPR works....except on TV. You might get away with a little digital zoom, but not much.
You need optical zoom not digital zoom. You might get away with a little digital, but not much. How far away would the vehicle be from the camera?
We need higher focal length cameras to make the field of view basically be the size of a car.
Regarding plates, keep in mind that this is a camera dedicated to plates and not an overview camera also. It is as much an art as it is a science. You will need two cameras. For LPR we need to OPTICALLY zoom in tight to make the plate as large as possible. For most of us, all you see is the not much more than a vehicle in the entire frame. Now maybe in the right location during the day it might be able to see some other things, but not at night.
At night, we have to run a very fast shutter speed (1/2,000) and in B/W with IR and the image will be black. All you will see are head/tail lights and the plate. Some people can get away with color if they have enough street lights, but most of us cannot. Here is a representative sample of plates I get at night of vehicles traveling about 45MPH at 175 feet from my 2MP 5241-Z12E camera (that is all that is needed for plates):
See the LPR subforum.
The only way to test the AOI is by live video of objects going thru that AOI camera and then playing back that AOI camera. As you are seeing, you cannot use previously recorded video to test it.
Thank you very much for the comprehensive response. Appreciate your help. At the moment all is running well with a python script I've created and integrated with on online ALPR (thus needed to transfer a cropped JPG to a remote server). This is working fine, also with that camera without zooming in, but requires the remote web-service and some not so easy to maintain logic. Unfortunately, the C.P. AI License Plate object recognition is not able , with the same picture working with my script that is just looking for a car, to identify either DayPlate nor NightPlate, even if the OCR is actually getting the correct plate number... so the need to zoom in. I'd experiment more, as I'd really wouldn't like to buy a news camera for the purpose. My plates are Italian, and I can say that the model is able to identify plates with some zoom, and is able to identify plates on another camera, with less distance from the cars. Maybe a better model could be trained, but to me it seems very complicated to do.
Your last answer is actually very clear, as well.
Thank you
Your last answer is actually very clear, as well.
Thank you