Search results

  1. VideoDad

    How to capture multiple plate reads in quick succession?

    Did you add the zones and get it so it triggers when the vehicles are in the middle of the frame? With 12 frames, you've only got 1.2 seconds of footage to find an image with a plate. Plus your processing time is about 5000ms (5 seconds) so it's going queue them up and take about a minute to...
  2. VideoDad

    Full ALPR Database System for Blue Iris!

    There are 3 possible versions: Plate (one plate) Memo (possibly multiple plates) AI dump (possibly multiple plates, plus additional info for each plate). The 3rd is the preferred version and is now mentioned first on the GitHub page. At this point, the comments in the readme about PLATE and...
  3. VideoDad

    How to capture multiple plate reads in quick succession?

    I agree with Mike, rather than triggering just on motion, setup a zone on the left of the screen and the other zone for the right of the screen. Then trigger based on crossing from one to the other. Now the plate should be close to the middle of the screen (or at least fully visible) and AI...
  4. VideoDad

    Full ALPR Database System for Blue Iris!

    You seem to have an HTML encoding of ampersand &. Change that to just & and it should work. However the first version (ai_dump) is the preferred version because it gets additional information such as the position of the plate, confidence level, etc.
  5. VideoDad

    How to capture multiple plate reads in quick succession?

    Can you look at the .dat file in the viewer?
  6. VideoDad

    Full ALPR Database System for Blue Iris!

    Somewhere in the notes for the ALPR Database, it should be mentioned that having a working LPR camera (correctly zoomed and focused, and configured for day and night capture) with the correct AI settings for detecting and recognizing plates in BI is the prerequisite. I wouldn't mind like seeing...
  7. VideoDad

    Full ALPR Database System for Blue Iris!

    Could it be the extra spaces after the colon in ALERT_PATH and ALERT_CLIP?
  8. VideoDad

    Full ALPR Database System for Blue Iris!

    Did you get the latest migration.sql when doing the last update? It sounds like you are missing the most recent DB update.
  9. VideoDad

    How to capture multiple plate reads in quick succession?

    Putting anything in the To Cancel field that isn't an expected object means it will continue processing through all the frames. I was suggesting it be left blank, but maybe that's not necessary now you've lowered the frame count. The AI settings screen has the "Default object detection" turned...
  10. VideoDad

    Full ALPR Database System for Blue Iris!

    I got bitten by this too. I had two reads of the same plate in quick succession and went to remove one and it removed every read that had ever occurred. This is not very intuitive and I'm not sure why it behaves that way.
  11. VideoDad

    How to capture multiple plate reads in quick succession?

    I ran this through the standard object model. It didn't detect the pizza. ☹️
  12. VideoDad

    How to capture multiple plate reads in quick succession?

    Here are a few of my thoughts. You are using a CPU for all your detections. I'm seeing typical times in the 500ms+ range for each call, from looking at your screenshots, sometimes up towards 900ms which is close to 1 full second. You are asking for 20 frames to be tested and quicker than CPAI...
  13. VideoDad

    OpenVPN home IP address

    I have my domain (and several of its subdomains) hosted on CloudFlare. Can you tell me more how you've got yours setup? I'm looking for the best way to have a subdomain set to go to my public IP at home and also get updated automatically if the ISP changes it.
  14. VideoDad

    Dahua Inaccurate Time Related Issues

    Assuming your cameras don't have internet access (they shouldn't), you should have an internal source of network time (NTP server). That could be your NVR, a PC, or your router. In my case, the router (e.g. 192.168.1.1) is setup to be the source of NTP so that's what is configured in the NTP...
  15. VideoDad

    Full ALPR Database System for Blue Iris!

    It's the "roadmap" accessible directly from the top of the dashboard page. It's the 3rd icon that looks like a road.
  16. VideoDad

    Full ALPR Database System for Blue Iris!

    There are some known issues with converting back and forth from local to UTC. There are a couple bugs already written up related to similar problems: https://alprdatabase.featurebase.app/p/dashboard-time-distribution-graph...
  17. VideoDad

    Full ALPR Database System for Blue Iris!

    Out of curiosity, is there a way to see the statistics on number of installs, record counts, earliest read, etc.?
  18. VideoDad

    Full ALPR Database System for Blue Iris!

    Can you try running the migration manually like @hopalong did? https://ipcamtalk.com/threads/full-alpr-database-system-for-blue-iris.79082/post-917256 Also, what do you see in your ALPR DB log (second to last icon)? Are you still seeing database errors?
  19. VideoDad

    What does this mean? Are there people connecting to my Blue Iris?

    Yes, I'm just reminding people that while it may seem convenient, having your computers open via port forwarding is risky. Try putting in your city and state and you might be surprised how many ones you see nearby you.
  20. VideoDad

    Full ALPR Database System for Blue Iris!

    @svalvasori, is your install of ALPR Database on Windows or Linux? Those MD5 hashes do seem to match the latest files. Have you tried rerunning the docker commands from your ALPR folder? docker compose pull docker compose up -d
  21. VideoDad

    Which version of BI are you running with ALPR Database + CPAI?

    I've run this with a much older version of BI and the most recent. And also an older version of CPAI and the most recent. As long as you are seeing plates detected and recording as flagged clips, then it shouldn't matter which versions you are using. The call to the ALPR Database is identical...
  22. VideoDad

    How to enable OpenWeatherMaps on camera video in Blue Iris 5.9.9.32 ?

    They are built-in macros as explained in the help file:
  23. VideoDad

    Full ALPR Database System for Blue Iris!

    You have the exact same symptoms as that earlier thread. Your database schema doesn't seem to be up to date. Can you look in your ALPR folder and post the timestamps of your schema.sql, migrations.sql and docker-compose.yml?
  24. VideoDad

    Full ALPR Database System for Blue Iris!

    Did you do the rest of the steps including the docker compose statements at the end of that post?
  25. VideoDad

    Full ALPR Database System for Blue Iris!

    Sounds like a DB schema issue again. Try the solution presented in post 660 https://ipcamtalk.com/threads/full-alpr-database-system-for-blue-iris.79082/post-914757
  26. VideoDad

    How to train a custom YOLOv5 model using CodeProject.AI Training for YoloV5 6.2 module

    What is it you are trying to accomplish? That's a set of data that is ready to be used for training. But it has more than just vehicle license plates so you'd have to go through and throw out objects that you didn't want. The result would be a trained model that could detect when an image...
  27. VideoDad

    How to train a custom YOLOv5 model using CodeProject.AI Training for YoloV5 6.2 module

    Personally, I would try using Collab on a test project and see how far you can take it. I think it's better to first determine if you can make a training data set and get some results using someone else's hardware first. Or if you have a low end GPU (for example I started with a 1050Ti), see...
  28. VideoDad

    How to train a custom YOLOv5 model using CodeProject.AI Training for YoloV5 6.2 module

    That's a training data set, meaning someone had to go through each image and tag objects that were important to their training objective. In this case, the person is putting rectangles around cars, license plates, etc. but ignoring street signs, lamp posts, buildings, etc. There are some tools...
  29. VideoDad

    How to train a custom YOLOv5 model using CodeProject.AI Training for YoloV5 6.2 module

    Then I suppose you could break each image into smaller fixed images. So maybe 18 images that are each 640 × 720. You'd have to get a batch program for cropping and first run it for a crop of (0,0) to (639,719). Then for (640,0) to (1279,719), etc. Each time you move over 640 pixels in...
  30. VideoDad

    How to train a custom YOLOv5 model using CodeProject.AI Training for YoloV5 6.2 module

    Is the vehicle always in a certain region of the picture? If so there are programs that will do a batch crop to a predefined rectangle. Alternatively, if you are handy with programming, you could have each photo run through AI detection for a vehicle, that returns a rectangle for the vehicle...