Recognition of PDF invoices in edoc document reader

edoc document reader recognizes invoice properties of PDF invoices (e.g. company name or vendor name). Azure Document Intelligence, a cloud-based AI service based on machine learning, is used for recognition. PDF invoices are therefore recognized by third-party software.

The predefined Invoice model provided by Microsoft is used when processing invoices. The standard Invoice model is specially designed for extracting typical invoice elements.

To find out which languages are supported when recognizing invoices, see: Language and locale support for prebuilt models - Document Intelligence - Azure AI services

Master data matching after document recognition

After the document properties have been recognized by Azure Document Intelligence, edoc master data matching service compares certain recognized values with existing master data, e.g. the company name or the vendor name.

If a clear matching is possible, properties can be added and updated on the basis of the master data, e.g:

Company ID and company name.
Vendor ID and vendor name.
Further information on the vendor such as tax number, VAT ID, account number, address and bank details of the vendor.

For more information on how the master data matching service works, see: Master data matching mechanisms (edoc master data matching service)

When edoc document reader is started, the schema Dokumenterkennung | Eingangsrechnung (document recognition | incoming invoice) is created in edoc documents . The schema contains all attributes (properties) of the schema Eingangsrechnung | aktiv (incoming invoice | active) including any recognized QR codes or barcodes.

Recognition of header data in invoices

The following header data of an invoice can be recognized by the Azure Document Intelligence invoice model:

Name	Description
Mandant-Name	Specifies the full name of the company.
Rechnungsnummer extern	Specifies the invoice number that is specified on the invoice
Rechnungsdatum	Specifies the date on which the invoice was issued.
Mandant \| Adresse \| Straße	Specifies the street and house number in the company's address.
Mandant \| Adresse \| Stadt	Specifies the city in the company's address.
Mandant \| Adresse \| Postleitzahl	Specifies the zip code in the company's address.
Mandant \| Adresse \| Land/Region	Specifies the country/region in the company's address.
Lieferanten-Name	Specifies the full name of the vendor.
Lieferanten \| Bank \| Iban	Specifies an IBAN of an account of the vendor.
Lieferanten \| Bank \| IBANs	Specifies the IBANs of the vendor's accounts.
Lieferanten \| Bank \| Bic	Indicates the BIC (Bank Identifier Code) of an account of the vendor.
Lieferanten \| Bank \| BICs	Specifies the BICs (Bank Identifier Code) of the vendor's accounts.
Lieferanten \| Adresse \| Street	Specifies the street and house number in the vendor's address.
Lieferanten \| Adresse \| City	Specifies the city in the vendor's address.
Lieferanten \| Adresse \| Postcode	Specifies the zip code in the vendor's address.
Lieferanten \| Adresse \| Country	Specifies the country/region in the vendor's address.
Lieferanten \| Umsatzsteuer-ID	Specifies the VAT identification number of the vendor.
Purchase order numbers	Specifies the purchase order numbers indicated on the invoice.
Brutto-Betrag	Specifies the gross amount of the invoice.
Netto gesamt	Specifies the invoice total as the net amount of all invoice items.
Umsatzsteuer-Betrag 1	Specifies the VAT amount for tax rate 1.
Währungs-Text	Specifies the currency in which the invoice amounts were entered.
Zahlungsbedingung	Specifies the description of the payment term.
Skontobetrag	Specifies the amount of the cash discount granted.
Fälligkeit	Specifies the due date of the payment.

Recognition of line item data in PDF invoices

The following line item data of an invoice can be recognized by the Azure Document Intelligence invoice model:

Name	Description
Position \| Buchungstext	Specifies the description in the invoice lines.
Position \| Bestell-Nr.	Specifies the purchase order number in the invoice lines.
Position \| Menge	Specifies the quantity in the invoice line.
Position \| Einzelpreis	Indicates the net price per unit in the invoice lines.
Position \| Netto	Specifies the net amount in the invoice lines.
Position \| MwSt. %	Specifies the VAT percentage rate in the invoice lines.
Position \| MwSt.	Specifies the VAT amount in the invoice lines.

Recognition of barcodes and QR codes in PDF invoices

With edoc document reader, barcodes and QR codes on documents can be recognized automatically.

To find out which QR codes and barcodes are captured, see: Add-on capabilities - Document Intelligence - Azure AI services