[neogenix]

Yes, for sure. You could probably train a specific vision model for this style of image parsing and run it on device for free. However, this was much easer to make :-)

[bonoboTP]

Ask Claude to code it up, then it's also easy.