AI setup help for ALPR
Posted: Mon Dec 23, 2024 5:52 pm
Hi folks,
I am trying to setup my Dahula IPC HFW 5241E as an ALPR camera. I have watched an excellent video on youtube and read the detailed writeup by the same author here https://www.youtube.com/watch?v=jqc1ukrkMmI and https://www.thesmarthomehookup.com/free ... lkthrough/ .
I have a few problems, but one at a time. So i have Blue Iris and code project setup and "working" in that on my other camera it tells me when it sees a person, car, animal etc, just like it should. On my ALPR camera, i am mostly just interested in reading plates, so have just Dayplate, Nightplate and ALPR turned on. At least i think that's what's going on. The analysis shows vehicles with Dayplate and Nightplate, but not the actual number. I have attached some pictures of the AI setup, code project and a typical output.
I'm using BI version 5.9.9.17, and its running on a windows 11 VM with 12Gb ram allocated, 8 cores 16 threads, under Unraid, on a EPYC 7K62 48 core 96 thread CPU with 128Gb ram. CPU load in the VM is typically 10-15% with 11 cameras. CodeProject AI 2.9.5 is running in a docker on unraid with a dedicated Nvidia P8 GPU. What am i doing wrong?
Thanks, Des
I am trying to setup my Dahula IPC HFW 5241E as an ALPR camera. I have watched an excellent video on youtube and read the detailed writeup by the same author here https://www.youtube.com/watch?v=jqc1ukrkMmI and https://www.thesmarthomehookup.com/free ... lkthrough/ .
I have a few problems, but one at a time. So i have Blue Iris and code project setup and "working" in that on my other camera it tells me when it sees a person, car, animal etc, just like it should. On my ALPR camera, i am mostly just interested in reading plates, so have just Dayplate, Nightplate and ALPR turned on. At least i think that's what's going on. The analysis shows vehicles with Dayplate and Nightplate, but not the actual number. I have attached some pictures of the AI setup, code project and a typical output.
I'm using BI version 5.9.9.17, and its running on a windows 11 VM with 12Gb ram allocated, 8 cores 16 threads, under Unraid, on a EPYC 7K62 48 core 96 thread CPU with 128Gb ram. CPU load in the VM is typically 10-15% with 11 cameras. CodeProject AI 2.9.5 is running in a docker on unraid with a dedicated Nvidia P8 GPU. What am i doing wrong?
Thanks, Des