How long is your system taking to process the 50 images you're sending to CPAI?
As Actran pointed out to me, if the AI is taking too long to process, you may be losing frames in the processing pipeline. The BI manual actually covers this; I bumped up my pre-trigger record time and found things...
Worth a shot I guess. :) I've also considered 24 hr time and a smaller font size. I think I'll still set up my tracking app to pull all inferences not just the first one. Then I can clearly see if I'm getting odd patterns.
Does everyone remove the timestamp from their LPR camera?
I recently noticed that the car with plate "PM" was passing by quite a lot and finally figured out what was up:
I did notice that the license plate model in YOLO v8 doesn't seem to capture the timestamps.
For now I'll just disable...
I'm going to cross-post this from my other thread.
I've been unable to use the license plate CP.AI with the GPU since it just stops for no reason. Works GREAT with the CPU only.
So I reinstalled the CUDA 11.8 drivers (I guess I hadn't done that and was running 12.X). GPU inference now...
I realized I didn't have the CUDA 11.8 drivers/toolkit installed. So I did that; and the GPU does get used:
However, the AI sure is hallucinating..
{
"api":"alpr",
"found":{
"success":true,
"processMs":882,
"inferenceMs":840...
I did some more investigation and got it dialed in much better. Removing object detection so I wasn't doing the license-plates model AS well as ALPR seemed to help. I also noticed that my camera was set to H.265. Setting it to H.264 resulted in stable frame grabs with nothing missing.
I...
AH HA!!! THANK YOU! No it wasn't clear to me at that the CP.AI response times would have any impact on the ability to grab a plate # before the plate is out of view.
However, reading the manual I do see the following very relevant writeup and diagram:
My mental model for this was that as...
Hello, thanks for the response.
As for the slow CP.AI response time, I do have an issue in my particular configuration right now; YOLO object detection leverages the GPU just fine but license plate detection crashes when the GPU is used. So for the moment LPR is CPU only.
The plates do get...
I'm using CodeProject LPR and I'm having some issues reliably capturing plates. The illustration below shows what I think is the problem.
Here's an analysis of one of the AI DAT files:
The car moves from right to left; as you can see, at T=0msec, the bumper of the car is just barely in frame...
Not yet. I will have to find that and give it a shot. I wonder if the lack of wdr for plates is a known issue.
Mind you the default shutter speed is way too slow and the cars are smears. I really had to quicken the shutter speed to get things to work.