Azure Document Intelligence reads European identification cards perfectly but makes a mess on driver licences?

Question

I am handling a project requiring to extract information from Italian document scans, comprising 2 types of identity cards (one old and one new template), passports and driver licences.
While the Identity Documents model captures perfectly every detail on identity cards and passports (including Tessera Sanitaria, which is sort of an ID specific for health services), driver licences seem particularly problematic with the model often not even seeing the person's name in very clear pictures.
Is there any way to improve the model output? I have info on what type of document (passport, national id, driver licence etc) beforehand and suggesting query fields such as 1, 2, 3 as field names did not solve my issue.

Answer 1

Hello @Costanza Cantalini , Thanks for using Microsoft Q&A Platform.

As per the documentation, the European driver license is a supported document type.

Could you please share the sample document with us through the private message feature here, or email it to us so that we can replicate the issue on our end?

Also, could you confirm the API version you are currently using?

If you still face issue, please raise a support ticket in the Azure portal.

I hope this helps.

Regards,

Vasavi