OCR reading

The OCR program converts information from images to text which can then be read by the text reader.

Of course, no OCR tool is perfect, so after conversion, the text needs to be checked and certain errors have to be manually corrected. Still, using an OCR tool is much faster than manually copying all text.

An OCR reader, i.e. a system for invoice recognition, is also built into PANTHEON. It is a self-learning software which uses complex algorithms of artificial intelligence for learning and increasingly improves data recognition.

The OCR reading system in PANTHEON can read various different data, usually all standard, i.e. mandatory, content on the invoice, such as the sender tax ID, sender title, net amount, value date etc.

This can be observed on the example of a received invoice below. The set of data that can be recognized by the system also depends on the PDF quality of the invoice.

Document conversion is charged by packages of processed invoices, regardless of the number of pages of an individual invoice. When you spend all the invoices in a package, you can simply order a new package.

Standard service users in PANTHEON can select between two methods of PDF parsing:

• Detailed by item line – Reads each individual document line.
• Summarized lines – Reads amounts joined by tax rates.