There are multiple (paid) solutions available:
An alternative would be to integrate yourself with Google Vision or Azure cognitive service.
This demo includes a example of extracting a VIN number from an image via the Google Vision API.
There is or was a module or widget for OCR. I made a POC that recognized texts. Also, there is an alternative in the making which will be available in a couple of days. So you might want to put your development on a hold for a bit and focus on the rest of the application first. I will keep you posted.
Then new Mendix module has a notify button for when it is going to be released. Do note that this is a paid option: https://marketplace.mendix.com/link/component/118392