Blue Iris and CodeProject.AI ALPR

First I suggest read the BI manual. I was confused too and did a lot of reading to understand. Also, BI has so many options it can be confusing at first (it was for me). Open the Trigger tab and then click the Help button. It will bring you directly to the correct section of the manual.

This is my understanding and my not be 100% accurate.

Trigger just means something happened. It could be motion configured/sensed by BI or could be an ONVIF event from a camera (which could just be motion too). You configure the "how" it knows something happened.

Alert is what to do once something was triggered (what to do when something happened). The basics (old way) was nothing additional and you just had to look to see what happened. The modern way is to use AI. However, notice the option to Confirm with AI is optional. This is technically a separate/additional/optional configuration.

AI is also configurable. You are telling it what to look for. By default, if it doesn't find it, it will cancel the alert. But that is an option that can be changed so you can keep it on the list and look at the analysis yourself. Looking for License plates (LP) is not the same as looking for vehicles (cars, trucks, etc). I agree technically if you found a plate then you would think you found a vehicle but technically they are not mutually inclusive. That's the tricky part. Also, I'm not 100% certain that it isn't a bug that it is not marking the vehicle because I don't use ONVIF with LPR.

From my understanding the LPR models are trained on LP only and not vehicle so it doesn't care where it finds them.

One more thing. If the camera has AI and can detect Humans and Vehicles then you can use that to confirm and don't need the BI/CPAI. That is what I do for some cameras. Confirming the vehicle will add the icon as my previous post mentioned. You do have to play with the ONVIF to do determine which if you want.

Hopefully that helps. I'm no expert though. :)
Thanks for your reply, I actually did read the BI manual and searched for every sentence that has the word vehicle, didn't come up with much. As you see in this, BI with AI found the license plate, but didn't make it down as vehicle? Blue Iris and CodeProject.AI ALPR

Plus, on the latest Nov 19 update (I think), BI actually removed the "Mark as Vehicle" option on the each of the camera settings, so I don't know what's going on anymore. I'm back on 5.9.7.9.
 
Thanks for your reply, I actually did read the BI manual and searched for every sentence that has the word vehicle, didn't come up with much. As you see in this, BI with AI found the license plate, but didn't make it down as vehicle? Blue Iris and CodeProject.AI ALPR

Plus, on the latest Nov 19 update (I think), BI actually removed the "Mark as Vehicle" option on the each of the camera settings, so I don't know what's going on anymore. I'm back on 5.9.7.9.
Did you read the "update" notes in the BI help file for each version?
 
Problem with alpr, i upgraded from 2.6.5 did the uninstall and manual deleted the two codeproject folders. After Installed ver 2.9.3 and yolo 5.6.2 and alpr, I keep getting no paddle installed. I even manual installed with cmd same results.
System specs
Server version: 2.9.3
System: Windows
Operating System: Windows (Microsoft Windows 11 version 10.0.22631)
CPUs: 11th Gen Intel(R) Core(TM) i5-11500 @ 2.70GHz (Intel)
1 CPU x 6 cores. 12 logical processors (x64)
GPU (Primary): NVIDIA GeForce RTX 3060 (12 GiB) (NVIDIA)
Driver: 566.14, CUDA: 11.8.89 (up to: 12.7), Compute: 8.6, cuDNN: 8.5
System RAM: 16 GiB
Platform: Windows
BuildConfig: Release
Execution Env: Native
Runtime Env: Production
Runtimes installed:
.NET runtime: 9.0.0
.NET SDK: 8.0.204
Default Python: Not found
Go: Not found
NodeJS: Not found
Rust: Not found
Video adapter info:
NVIDIA GeForce RTX 3060:
Driver Version 32.0.15.6614
Video Processor NVIDIA GeForce RTX 3060
System GPU info:
GPU 3D Usage 6%
GPU RAM Usage 10 GiB
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168


23:44:59:ALPR_adapter.py: Traceback (most recent call last):
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR_adapter.py", line 11, in
23:44:59:ALPR_adapter.py: from ALPR import init_detect_platenumber, detect_platenumber
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR.py", line 17, in
23:44:59:ALPR_adapter.py: from paddleocr import PaddleOCR
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\init.py", line 14, in
23:44:59:ALPR_adapter.py: from .paddleocr import *
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\paddleocr.py", line 21, in
23:44:59:ALPR_adapter.py: import paddle
23:44:59:ALPR_adapter.py: ModuleNotFoundError: No module named 'paddle'
23:49:27:ALPR_adapter.py: Traceback (most recent call last):
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR_adapter.py", line 11, in
23:49:27:ALPR_adapter.py: from ALPR import init_detect_platenumber, detect_platenumber
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR.py", line 17, in
23:49:27:ALPR_adapter.py: from paddleocr import PaddleOCR
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\init.py", line 14, in
23:49:27:ALPR_adapter.py: from .paddleocr import *
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\paddleocr.py", line 21, in
23:49:27:ALPR_adapter.py: import paddle
23:49:27:ALPR_adapter.py: ModuleNotFoundError: No module named 'paddle'
Logging level
warning
 
Problem with alpr, i upgraded from 2.6.5 did the uninstall and manual deleted the two codeproject folders. After Installed ver 2.9.3 and yolo 5.6.2 and alpr, I keep getting no paddle installed. I even manual installed with cmd same results.
System specs
Server version: 2.9.3
System: Windows
Operating System: Windows (Microsoft Windows 11 version 10.0.22631)
CPUs: 11th Gen Intel(R) Core(TM) i5-11500 @ 2.70GHz (Intel)
1 CPU x 6 cores. 12 logical processors (x64)
GPU (Primary): NVIDIA GeForce RTX 3060 (12 GiB) (NVIDIA)
Driver: 566.14, CUDA: 11.8.89 (up to: 12.7), Compute: 8.6, cuDNN: 8.5
System RAM: 16 GiB
Platform: Windows
BuildConfig: Release
Execution Env: Native
Runtime Env: Production
Runtimes installed:
.NET runtime: 9.0.0
.NET SDK: 8.0.204
Default Python: Not found
Go: Not found
NodeJS: Not found
Rust: Not found
Video adapter info:
NVIDIA GeForce RTX 3060:
Driver Version 32.0.15.6614
Video Processor NVIDIA GeForce RTX 3060
System GPU info:
GPU 3D Usage 6%
GPU RAM Usage 10 GiB
Global Environment variables:
CPAI_APPROOTPATH = <root>
CPAI_PORT = 32168


23:44:59:ALPR_adapter.py: Traceback (most recent call last):
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR_adapter.py", line 11, in
23:44:59:ALPR_adapter.py: from ALPR import init_detect_platenumber, detect_platenumber
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR.py", line 17, in
23:44:59:ALPR_adapter.py: from paddleocr import PaddleOCR
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\init.py", line 14, in
23:44:59:ALPR_adapter.py: from .paddleocr import *
23:44:59:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\paddleocr.py", line 21, in
23:44:59:ALPR_adapter.py: import paddle
23:44:59:ALPR_adapter.py: ModuleNotFoundError: No module named 'paddle'
23:49:27:ALPR_adapter.py: Traceback (most recent call last):
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR_adapter.py", line 11, in
23:49:27:ALPR_adapter.py: from ALPR import init_detect_platenumber, detect_platenumber
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR.py", line 17, in
23:49:27:ALPR_adapter.py: from paddleocr import PaddleOCR
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\init.py", line 14, in
23:49:27:ALPR_adapter.py: from .paddleocr import *
23:49:27:ALPR_adapter.py: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddleocr\paddleocr.py", line 21, in
23:49:27:ALPR_adapter.py: import paddle
23:49:27:ALPR_adapter.py: ModuleNotFoundError: No module named 'paddle'
Logging level
warning
Try uninstalling the module then reinstall

1732660408556.png
 
Try the below steps.

Stop the ALPR module
1732673148610.png

Delete C:\Program Files\CodeProject\AI\modules\ALPR\bin
1732673244913.png

Replace requirements.windows.cuda11_8.txt with the attached requirements.windows.cuda11_8.txtfile
1732673643441.png

In a command prompt in folder C:\Program Files\CodeProject\AI\modules\ALPR run ..\..\setup.bat
1732673831284.png

After setup.bat finishes restart the ALPR module and hopefully it should be working
1732673973048.png
 

Attachments

How do I avoid this ALPR error?

ai.png

It is taking the first good license plate image and using this, even though it is incomplete and lesser confidence than others.

I have already maxed out my IVS rule all the way over to the right going one direction.

ai1.png


Is there a way to put a delay before AI starts evaluation? Is there a way for AI to use the best image?

ai3.png
 
Last edited:
Dumb question unrelated to any specific feature or configuration/setting: Does anyone have a more recent version of the Blue Iris / CodeProject.ai integration or FAQ documents posted on the CodeProject.ai Github webpages? Those are for BI v5.5.x.y and BI v5.7.x.y and we're up to v5.9.x.y these days, with significant differences in many of the options/settings/configuration. Just wondering before I spend a lot of time reading older stuff that's out-of-date. Thank you!
 
So it's snowing outside, and my LPR camera alerts are going nuts and constantly filling my drive with jpgs. What should I be adjusting so that it's not alerting for excessive snow?

1733362089786.png

1733362063415.png