How useful is this to use with OCR in an image

[atv2016]

I've found several times now where it recognises text in an image (sometimes camera overlay) and I was wondering if It will recognise larger text (like PRIME) on amazon prime vans, or if it really looks for a license plate.

It would be super useful to have automations run on that if that would be the case. Rather then training a frigate+ model, I would just use platerecognizer to recognise text. It's a really useful use case, as the text will always be parked in front of my street and more importantly, it will be in large enough letters to be easily recognised.