New CodeProject.AI License Plate Recognition (YOLO11) Module

[MikeLud1]

Server version: 2.9.5
System: Windows
Operating System: Windows (Windows 11 24H2)
CPUs: Intel(R) Core(TM) i9-14900HX (Intel)
1 CPU x 24 cores. 32 logical processors (x64)
GPU (Primary): NVIDIA GeForce RTX 4070 Laptop GPU (8 GiB) (NVIDIA)
Driver: 576.02, CUDA: 12.4.99 (up to: 12.9), Compute: 8.9, cuDNN: 8.9
System RAM: 96 GiB
Platform: Windows
BuildConfig: Release
Execution Env: Native
Runtime Env: Production
Runtimes installed:
.NET runtime: 9.0.0
.NET SDK: Not found
Default Python: Not found
Go: Not found
NodeJS: Not found
Rust: Not found
Video adapter info:
NVIDIA GeForce RTX 4070 Laptop GPU:
Driver Version 32.0.15.7602
Video Processor NVIDIA GeForce RTX 4070 Laptop GPU
Intel(R) UHD Graphics:
Driver Version 32.0.101.6987
Video Processor Intel(R) RaptorLake-S Mobile Graphics Controller
System GPU info:
GPU 3D Usage 6%
GPU RAM Usage 3.2 GiB
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168

Ideally both modules should be using your NVIDIA GeForce RTX 4070 Laptop GPU - GPU(DirectML) and this should work. I have both a RTX 3060 and RTX 5090 and they both work with .NET and the ALPR module using GPU(DirectML). Both of these PC have AMD CPUs that do not have an integrated GPU.
What I am seeing is that most of the issues people are having is when there are two GPS. I am going see if I can fix this over the weekend.

 

[MikeLud1]

For anyone having issues with v1.0.2 running please run the below in a command prompt and post the results

wmic path win32_videocontroller get Name

 

[Vettester]

For anyone having issues with v1.0.2 running please run the below in a command prompt and post the results

As you know I was having issues, but a reinstallation resolved them.

 

[MikeLud1]

As you know I was having issues, but a reinstallation resolved them.

View attachment 230255

Can you also post a screenshot of your GPU info from the task manager.

 

[MikeLud1]

For anyone having issues with v1.0.2 running please post both screenshots

Post in thread 'New CodeProject.AI License Plate Recognition (YOLO11) Module'

Thursday at 8:46 PM

For anyone having issues with v1.0.2 running please run the below in a command prompt and post the results

wmic path win32_videocontroller get Name

• MikeLud1

Post in thread 'New CodeProject.AI License Plate Recognition (YOLO11) Module'

Thursday at 10:33 PM

As you know I was having issues, but a reinstallation resolved them.

View attachment 230255

Can you also post a screenshot of your GPU info from the task manager.

• MikeLud1

 

[Vettester]

Can you also post a screenshot of your GPU info from the task manager.

 

[pepperfr]

For anyone having issues with v1.0.2 running please run the below in a command prompt and post the results

wmic path win32_videocontroller get Name

View attachment 230252

Sorry for my slow response, but here are the results:

 

[pepperfr]

Can you also post a screenshot of your GPU info from the task manager.

View attachment 230257

Here my GPU info:

 

[Skinny1]

For anyone having issues with v1.0.2 running please run the below in a command prompt and post the results

wmic path win32_videocontroller get Name

View attachment 230252

 

[Skinny1]

 

[PeteJ]

While I wait for the Linux version, has anyone tried the new state and vehicle description features?

 

[MikeLud1]

While I wait for the Linux version, has anyone tried the new state and vehicle description features?

This is not finished yet, I want to get the main ALPR working

 

[MikeLud1]

Another feature I am working on is vehicle speed. Speed looks to be calculating correctly, I need to do a drive by at set speeds to confirm.

 

[PeteJ]

Mike, do you have an approx time frame for the Linux release? I'll be on a 11th gen Intel with iGPU only so I wont be able to provide any feedback for CUDA, etc.

 

[MikeLud1]

Mike, do you have an approx time frame for the Linux release?

Not yet, I am still ironing out Windows. If you do not have a Nvidia GPU on Linux it will be using your CPU, DirectML only works in windows

 

[Vettester]

Another feature I am working on is vehicle speed.

This would be an awesome addition! I'm currently running Frigate on one camera specifically for speed.

 

[adventuredphoto]

Everything been solid for me. Works much faster than the default module. Thank you Mike

 

[ipcamwalt]

Hey Mike,

I was running your new yolo11 v1 all week long and thought I would give the update to 1.0.2 a go. I updated to it by stopping it, uninstalling it, shutting down CPAI, deleting the modules folder for Onyx11, restarted CPAI, enabled do not use download cache, installed v1.0.2, started it up and got these errors:





I only connect to this machine remotely, it is headless, but it is detecting the remote display adapters.
That being said, I don't have an E: currently on this machine, or any machine connected to this, so that might be a path error in the module?

v1.0.0 worked fine using this exact setup, I have tried disconnecting the remote desktop connection and just using the web gui to start the module but it still fails the same.

It says it has started successfully but since updating to 1.0.2 it has identified 0 objects and I haven't seen it be passed anything.

Are we still passing DayPlate and NightPlate?

 

[MikeLud1]

Hey Mike,

I was running your new yolo11 v1 all week long and thought I would give the update to 1.0.2 a go. I updated to it by stopping it, uninstalling it, shutting down CPAI, deleting the modules folder for Onyx11, restarted CPAI, enabled do not use download cache, installed v1.0.2, started it up and got these errors:
View attachment 230342
View attachment 230343

I only connect to this machine remotely, it is headless, but it is detecting the remote display adapters.
That being said, I don't have an E: currently on this machine, or any machine connected to this, so that might be a path error in the module?

v1.0.0 worked fine using this exact setup, I have tried disconnecting the remote desktop connection and just using the web gui to start the module but it still fails the same.

I think I know what is happening post the below screenshot

 

[ipcamwalt]

Looks like my driver is old, I think that was a CUDA thing to get old packages installed? Should I grab the latest drivers?