Search results

If possible move the camera closer to the street. Mine are about 15' from the street so I get good shots in both directions.

To improve accuracy you'll need to decrease the camera angle.

Thanks again @MikeLud1, the usps onnx model has been spot on since I started using it with zero false alerts.

If anyone else wants this Mike converted the pt file to an onnx. https://ipcamtalk.com/threads/codeproject-ai-version-2-5.74304/post-848630

Awesome, thank you!

@MikeLud1 Hey Mike, any chance of converting the attached USPS.pt model to an onnx file? TIA

I haven’t tried that yet, I’ll give it a shot. Thanks!

Anybody have a USPS.onnx file they'd be willing to share? I've tried using the delivery.onnx file with CPAI for detecting my mail delivery, but it doesn't seem to be very reliable. It consistently identifies truck tail lights as a usps logo.

You can also use zones in BI and mask the area where your overlays are located.

Maybe it has something to do with your shiny new gaming machine (i7-14700F - 32GB DDR5 Memory) while mine is an old piece of junk (i7-4790 - 16GB Memory). :)

Updating to 5.8.9.7 causes "clips: disk can't keep up" errors. Had to roll back to 5.8.9.6.

Just as a test to see if speeds up your MQTT times. If it does then then you know AI is the issue and you can work on decreasing your processing times.

Have you tried adding your MQTT trigger to "Immediate {pre-altert) actions..." to see if that helps?

It works well with two Windows 10 installations.

I just wanted to say thanks again for the IPC-HFW5241E-Z12E-S2. I received the camera 3 days after I purchased it and it was in excellent condition, I set it up a few days ago and it works perfectly.

But did you check to see if you’re still missing plates? I don’t think the pre-trigger record time is the issue here.

You might want to try rolling back to 5.8.7.11 and see if that helps.

So I was using the following settings for my LPR camera: I sent this to Ken and this was his response: Yes with these settings, all custom models will be run. If you want only specific models, you need to specify the model name. If you use a non existent mode name, none of them will be...

Nope, it’s using all the custom models.

That's exactly what it is doing and it's obviously a bug. I would suggest emailing support.

5.8.7.8 Fixed the issue with the start/stop option.

I guess I’ve never noticed because I don’t use this option. I have the service setup in windows to start automatically.

Mine is greyed out as well.

I updated to 5.8.7.7 and AI is working. but the use custom models option field is still grayed out.

I am on the latest version and I posted this issue on the CodeProject.AI Insiders forum.

YOLOv5 .NET I shutdown yolov5.net and installed YOLOv5 6.2 and am seeing the same errors.

Bump, I now know that this not an isolated issue on my system. Any ideas on how to resolve this @MikeLud1 ?

You can take a screenshot of the trigger image and use that in CPAI Explorer or setup BI to capture the trigger image to a specific directory.

It only uses the module that you have enabled.

Using YOLOv5Net and the ipcam-dark model your squirrel #1 cropped image is identified as a cat.