Blue Iris and CodeProject.AI ALPR

Hi all! i'm trying to use ALPR module with my Tesla P4, but i can't. I'm on Unraid and my CPAI docker is on it. I can't even choose to enable or disable the GPU fpr this module.
Here there are some screenshot View attachment 174439
View attachment 174440
View attachment 174441

thank you so much
The current release does not support CUDA with your setup. CodeProject.AI Version 2.3 should be released sometime this week and should fix the CUDA issue you are having.

1696965147368.png
 
I need some help. I cannot seem to get Codeproject 2.2.4 or 2.3.0 License Plate Reader Module 2.7 to load up. I keep getting the error "Failed to Start" and the thing is all in red.

This is the server log error I am getting -

17:46:47:Error trying to start License Plate Reader (ALPR_adapter.py)
17:46:47:An error occurred trying to start process 'C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python37\venv\Scripts\python' with working directory 'C:\Program Files\CodeProject\AI\modules\ALPR'. The system cannot find the file specified.
17:46:47: at System.Diagnostics.Process.StartWithCreateProcess(ProcessStartInfo startInfo)
at CodeProject.AI.Server.Modules.ModuleProcessServices.StartProcess(ModuleConfig module)
17:46:47:Please check the CodeProject.AI installation completed successfully
17:46:47:ALPR has left the building

Would appreciate any idea's, suggestions, or help.


Thank you,


Rick
 
I need some help. I cannot seem to get Codeproject 2.2.4 or 2.3.0 License Plate Reader Module 2.7 to load up. I keep getting the error "Failed to Start" and the thing is all in red.

This is the server log error I am getting -

17:46:47:Error trying to start License Plate Reader (ALPR_adapter.py)
17:46:47:An error occurred trying to start process 'C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python37\venv\Scripts\python' with working directory 'C:\Program Files\CodeProject\AI\modules\ALPR'. The system cannot find the file specified.
17:46:47: at System.Diagnostics.Process.StartWithCreateProcess(ProcessStartInfo startInfo)
at CodeProject.AI.Server.Modules.ModuleProcessServices.StartProcess(ModuleConfig module)
17:46:47:please check the CodeProject.AI installation completed successfully
17:46:47:ALPR has left the building

Would appreciate any idea's, suggestions, or help.


Thank you,


Rick
It looks like you are on the beta team since you tried v2.3.0, if so install the current beta version 2.3.1-Alpha. After installing version 2.3.1-Alpha uninstall the License Plate Reader Module 2.7 then reinstall License Plate Reader Module 2.7.1
 
  • Like
Reactions: fenderman
It looks like you are on the beta team since you tried v2.3.0, if so install the current beta version 2.3.1-Alpha. After installing version 2.3.1-Alpha uninstall the License Plate Reader Module 2.7 then reinstall License Plate Reader Module 2.7.1

I rolled it back to 2.2.4 and it appears to be working. We shall see.

Thank you,

Rick
 
I rolled it back to 2.2.4 and it appears to be working. We shall see.

Thank you,

Rick


I am not having much luck with Codeproject Ai here lately. I am running 2.2.4 and the sever keeps going offline no matter what version I am running. Sometimes it'll go 5 minutes and other times it'll run for twenty minutes then drop offline. Any idea's? Thank you, Rick.
 
I am not having much luck with Codeproject Ai here lately. I am running 2.2.4 and the sever keeps going offline no matter what version I am running. Sometimes it'll go 5 minutes and other times it'll run for twenty minutes then drop offline. Any idea's? Thank you, Rick.
Post the below screenshots

1697291122772.png
One of your Cameras settings
1697291212759.png

1697291252204.png
1697291280829.png
 
Thank you for the reply Mike, Here you go.

Rick
 

Attachments

  • Codeproject PC nfo.png
    Codeproject PC nfo.png
    98.6 KB · Views: 35
  • Codeproject camera settings.png
    Codeproject camera settings.png
    56.9 KB · Views: 37
  • codeproject ai screen.png
    codeproject ai screen.png
    51.9 KB · Views: 35
  • codeproject 2.2.3.png
    codeproject 2.2.3.png
    99.7 KB · Views: 31
Newbie question, is this video based or photo based? my z12 photo snapshot often miss the plate that would be avilable on the video
 
Hey everyone! can someone explain to me how the myplates box under the AI tab works?
I'm assuming it's to put your plates in so the AI can cancel the analysis if it detects it. If this is how it's supposed to work, anyone know how to get it working?
 
Hey everyone! can someone explain to me how the myplates box under the AI tab works?
I'm assuming it's to put your plates in so the AI can cancel the analysis if it detects it. If this is how it's supposed to work, anyone know how to get it working?
Below if is from the manual.

1697339494200.png
 
  • Like
Reactions: actran
Below if is from the manual.

View attachment 175004

I did see this in the manual, thanks for sharing.

Question, do it add the plates with quotes " " or without quotes in the box? I tried both (comma delimited) and it's still finding my plates (cars parked in driveway). Reason I ask this is because next to the box it says "myplates" list:
Also, am I supposed to modify anything in the Alerts Tab or AI settings of the camera?
 
Hello everyone, glad to be here.
I have a rather newbie question, something that I am trying to understand.
Blue Iris is installed, CPAI is installed, both are working fine, they talk to each other, information is being exchanged, etc.
Now, I have been learning those tools and stumbled upon the Vision page.
Under the Vision tab, Custom Detect section, I loaded "License Plates V2" model, loaded an image, and it detects the license plate as "DayPlate" with high confidence.
1697571014542.png
Log entries for the above:
Code:
22:34:35:Object Detection (YOLOv5 6.2): Detecting using license-plate
22:34:35:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'custom' (...7b4c52) took 116ms

However, when I switch to the "Image" tab, License Plate reader, I upload the very same image, and the tool detects nothing. "No predictions returned".
1697571094429.png

Server logs show no error:

Code:
22:34:01:Object Detection (YOLOv5 6.2): Detecting using license-plate
22:34:01:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'custom' (...79c5f6) took 89ms
22:34:01:License Plate Reader: Rec'd request for License Plate Reader command 'alpr' (...44678d) took 325ms

Is this expected behavior? I'd like to make sure everything works as expected.
Further info:
1697571466699.png

Code:
Server version:   2.2.4-Beta
Operating System: Windows (Microsoft Windows 10.0.19045)
CPUs:             Intel(R) Core(TM) i7-6800K CPU @ 3.40GHz (Intel)
                  1 CPU x 6 cores. 12 logical processors (x64)
GPU:              NVIDIA GeForce GTX 1080 Ti (11 GiB) (NVIDIA) 
                  Driver: 537.58 CUDA: 12.2 (max supported: 12.2) Compute: 6.1
System RAM:       32 GiB
Target:           Windows
BuildConfig:      Release
Execution Env:    Native
Runtime Env:      Production
.NET framework:   .NET 7.0.10
Video adapter info:
  NVIDIA GeForce GTX 1080 Ti:
    Driver Version     31.0.15.3758
    Video Processor    NVIDIA GeForce GTX 1080 Ti
  NVIDIA GeForce GTX 1080 Ti:
    Driver Version     31.0.15.3758
    Video Processor    NVIDIA GeForce GTX 1080 Ti
System GPU info:
  GPU 3D Usage       16%
  GPU RAM Usage      4.2 GiB
Global Environment variables:
  CPAI_APPROOTPATH = <root>
  CPAI_PORT        = 32168
 
Hello everyone, glad to be here.
I have a rather newbie question, something that I am trying to understand.
Blue Iris is installed, CPAI is installed, both are working fine, they talk to each other, information is being exchanged, etc.
Now, I have been learning those tools and stumbled upon the Vision page.
Under the Vision tab, Custom Detect section, I loaded "License Plates V2" model, loaded an image, and it detects the license plate as "DayPlate" with high confidence.
View attachment 175239
Log entries for the above:
Code:
22:34:35:Object Detection (YOLOv5 6.2): Detecting using license-plate
22:34:35:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'custom' (...7b4c52) took 116ms

However, when I switch to the "Image" tab, License Plate reader, I upload the very same image, and the tool detects nothing. "No predictions returned".
View attachment 175240

Server logs show no error:

Code:
22:34:01:Object Detection (YOLOv5 6.2): Detecting using license-plate
22:34:01:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'custom' (...79c5f6) took 89ms
22:34:01:License Plate Reader: Rec'd request for License Plate Reader command 'alpr' (...44678d) took 325ms

Is this expected behavior? I'd like to make sure everything works as expected.
Further info:
View attachment 175241

Code:
Server version:   2.2.4-Beta
Operating System: Windows (Microsoft Windows 10.0.19045)
CPUs:             Intel(R) Core(TM) i7-6800K CPU @ 3.40GHz (Intel)
                  1 CPU x 6 cores. 12 logical processors (x64)
GPU:              NVIDIA GeForce GTX 1080 Ti (11 GiB) (NVIDIA)
                  Driver: 537.58 CUDA: 12.2 (max supported: 12.2) Compute: 6.1
System RAM:       32 GiB
Target:           Windows
BuildConfig:      Release
Execution Env:    Native
Runtime Env:      Production
.NET framework:   .NET 7.0.10
Video adapter info:
  NVIDIA GeForce GTX 1080 Ti:
    Driver Version     31.0.15.3758
    Video Processor    NVIDIA GeForce GTX 1080 Ti
  NVIDIA GeForce GTX 1080 Ti:
    Driver Version     31.0.15.3758
    Video Processor    NVIDIA GeForce GTX 1080 Ti
System GPU info:
  GPU 3D Usage       16%
  GPU RAM Usage      4.2 GiB
Global Environment variables:
  CPAI_APPROOTPATH = <root>
  CPAI_PORT        = 32168
The issues is you have CUDA 12.2 installed, the License Plate Reader only works with CUDA 11.x so the version you should use is CUDA 11.8 also the cuDNN needs to be for CUDA 11.X
There is an updated to CodeProject.AI (v2.3) that should be out any day and this version should work with CUDA 12.2