Ability to have OCR (Optical Character Recognition) of text in pdfs that do not have selectable text.
Here’s an example of our workflow for a vendor that submits pdf invoices that have text that is not selectable:
1. Type was changed from "Other" to "Invoice" by User because the OCR system could not read the word "INVOICE".
2. Supplier name was set to "[Name of vendor]" by User because the OCR system could not read the lines under "REMIT TO:".
3. Tax was set to "GST" by User because the OCR system could not read the word GST at line 114.
4. Category was set to "Shipping and Delivery" by User because the OCR system could not read the lines under "REMIT TO:".
5. Document reference was set to "292" by User because the OCR system could not read the invoice number shown on the top left corner of the document.
6. Total amount was changed from "CAD 0.00" to "CAD 4370.76" by User because the OCR system could not read the amount shown under "TOTAL CHARGES:".
7. Due date was set to "19 Apr 2025" by User because the OCR system could not read the date shown on the document.
8. Date was set to "12 Apr 2025" by User because the OCR system could not read the date shown on the document.
Thanks!
