Blue Iris and CodeProject.AI ALPR

hxhoang said:
Then dashboard shows LPR status with GPU(CUDA), but it seems LPR does not actually use GPU: No change in inference time.
Furthermore, nvidia-smi does not show process PID of LPR, only YOLO.
Click to expand...

Thanks for the info. Good observation on the CPU vs GPU times for ALPR. Looking a my ALPR times, they are under 50ms with CPU. Switching to the GPU for YOLO has significantly improved over CPU plus now the CPU isn't pegged the whole time.

Any idea on how to completely remove modules? I can seem to uninstall Face and YOLO NET.
 
hopalong said:
Thanks for the info. Good observation on the CPU vs GPU times for ALPR. Looking a my ALPR times, they are under 50ms with CPU. Switching to the GPU for YOLO has significantly improved over CPU plus now the CPU isn't pegged the whole time.

Any idea on how to completely remove modules? I can seem to uninstall Face and YOLO NET.
Click to expand...
No luck for me to uninstall Face and YOLO Net too. Have to set AutoStart to false in modulesettings.json.
I used to have CPAI and BI on Windows 11 VM on the same promox box (i9-13900HX), but inference speed of YOLO5.62 was slower.
I moved CPAI to LXC container with frigate, and YOLO5v6.2 time got below 10ms vs ~100ms with windows VM.
ALPR is still around 300-500ms with BI, and ~100ms with CodeProject.AI Explorer on both setup.

Your ALPR times of 50ms is quite good, even with CPU. Not sure why my ALPR time is still quite high for the HW of my box.
Plan to try CPAI and BI on bare metal once I have time for it.
 
  • Like
Reactions: hopalong
hxhoang said:
No luck for me to uninstall Face and YOLO Net too. Have to set AutoStart to false in modulesettings.json.
I used to have CPAI and BI on Windows 11 VM on the same promox box (i9-13900HX), but inference speed of YOLO5.62 was slower.
I moved CPAI to LXC container with frigate, and YOLO5v6.2 time got below 10ms vs ~100ms with windows VM.
ALPR is still around 300-500ms with BI, and ~100ms with CodeProject.AI Explorer on both setup.

Your ALPR times of 50ms is quite good, even with CPU. Not sure why my ALPR time is still quite high for the HW of my box.
Plan to try CPAI and BI on bare metal once I have time for it.
Click to expand...

It's more of a maintenance, keeping things clean type of thing. Not super critical. I'm staying at YOLOv5 6.2 - 1.9.1 as the 1.9.2 update breaks GPU. Thanks again for all your feedback.

Here are my unRaid machine specs for reference:
Supermicro X9SCM-F
Intel Xeon CPU E3-1230 V2 @ 3.30GHz
16 GB DDR3 Single-bit ECC
Nvidia Tesla P4 8GB
Coral USB

This machine is just for long term data storage for BI and running things like CP.AI and Frigate.
 
  • Like
Reactions: hxhoang
I'm getting a few that show nothing found when the same .dat file shows otherwise. Also, before the memo would show Plate:xxxxx but now it just lists the plate number making them impossible to search for.
Edit: most hits are showing nothing found now

1722802986358.png
1722803005438.png
 
Last edited:
Evidently this had to do with object detection as the icons in the AI analysis show the double blue square. I never had a problem with this before, but disabling it fixed it. This was very effective when it worked to capture front and rear plates but I can increase post frames to compensate
1722959734477.png
1722959790264.png
 
  • Like
Reactions: hopalong
Ok, I am trying to get CP.AI and LPR to work with BI. I’m having some trouble. Here are some screen shots of how things are setup right now. Any suggestions to be able to save these plates? Thank you!
 

Attachments

  • IMG_0142.jpeg
    IMG_0142.jpeg
    3.9 MB · Views: 15
  • IMG_0143.jpeg
    IMG_0143.jpeg
    4 MB · Views: 18
  • IMG_0144.jpeg
    IMG_0144.jpeg
    7.6 MB · Views: 18
I lost my server from dead C:drive and have built a new one. I ignorantly installed the newest versions of everything, but couldn't get ALPR to see a plate. :rolleyes: It was working fantastic on the old server but I can't remember what versions I had. What are the current versions of Blue Iris, Codeproject.ai and CUDA / cudnn that play nice for ALPR?

BTW : R5 7600 , GTX1050 and Windows10
 
Last edited:
Scrumblero said:
I lost my server from dead C:drive and have built a new one. I ignorantly installed the newest versions of everything, but couldn't get ALPR to see a plate. :rolleyes: It was working fantastic on the old server but I can't remember what versions I had. What are the current versions of Blue Iris, Codeproject.ai and CUDA / cudnn that play nice for ALPR?

BTW : R5 7600 , GTX1050 and Windows10
Click to expand...

For your GPU install CUDA 11.8 first and then run install_cuDNN.bat

1722993161440.png
 
Hi- Picked up a new machine after my old one started randomly shutting itself down. Was working on trying to get CodeProject.AI installed to help with motion events, and ultimately ALPR.

Ran into a snag right out of the gate...can't seem to get CodeProject.AI installed on the local machine. I tried it as a docker container on my Unraid rig, but couldn't get that to go either.

Here are the specs for the machine that I'm trying to get it installed on, according to the CodeProject.AI server tab:

Server version: 2.6.5
System: Windows
Operating System: Windows (Microsoft Windows 11 version 10.0.22631)
CPUs: Intel(R) Core(TM) i5-9600 CPU @ 3.10GHz (Intel)
1 CPU x 6 cores. 6 logical processors (x64)
GPU (Primary): (NVIDIA), CUDA: (up to: ), Compute: , cuDNN:
System RAM: 16 GiB
Platform: Windows
BuildConfig: Release
Execution Env: Native
Runtime Env: Production
Runtimes installed:
.NET runtime: 7.0.10
.NET SDK: Not found
Default Python: Not found
Go: Not found
NodeJS: Not found
Rust: Not found
Video adapter info:
Intel(R) UHD Graphics 630:
Driver Version 30.0.100.9864
Video Processor Intel(R) UHD Graphics Family
System GPU info:
GPU 3D Usage 0%
GPU RAM Usage 0
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168

Screenshot of the Status Tab attached as well as portion of server log with error message. Thanks for any pointers you can provide.
 

Attachments

  • server-status.png
    server-status.png
    133.8 KB · Views: 10
  • server-error.png
    server-error.png
    125.5 KB · Views: 10
You must log in or register to reply here.
Top Bottom