Previously, when ReadSoft Invoices used OmniPage to process PDF invoices, and the PDFInput INI flag was set to the default value, 1, it indirectly used the ABBYY engine for page extraction (to generate TIFF files with ,000, .001, .xxx extensions representing each page of the invoice). Consequently as the default behavior, each page was extracted and saved using ABBYY after which ReadSoft Invoices fed the newly generated files to OmniPage for the actual interpretation.

In addition, if the PDFInput flag was enabled and ABBYY failed to extract the information from a PDF invoice, OmniPage was used to extract the information as a precautionary measure.

If the flag was not enabled, OmniPage was used directly for page extraction and interpretation.

Starting with ReadSoft Invoices 6.0.3, the dependency on ABBY has been removed, and the PDFInput flag is no longer used. As a result, it will now use OmniPage directly if it is selected as the OCR engine under Profile settings. That is, moving forward, both OCR engines will work entirely independently from each other. Only the specified engine will be used for the entire extraction and interpretation process.