Recognition: Characters (OCR/ICR)

 Audience

This article is meant for PSIcapture Administrators.

Overview

The recognition engines used for Optical Character Recognition and Intelligent Character Recognition throughout the workflow are configured here. Select the engine and associated settings that best suit your business needs.

NOTE: For ICR to function, please also select a country name (not just a language) in the "Recognition languages and countries" list. Without a country selected, ICR will fail.

Select Default Recognition Language

Supported languages are listed for selection.

NOTE: Different Languages/Regions are available for the OmniPage and RecoStar engines respectively.

Engine Settings

• Engine - There are two available engines:
  • OmniPage - The newest generation of PSIcapture 7.9+ uses the OmniPage engine.
  • RecoStar (Deprecated) - PSIcapture 7.8 and previous used the RecoStar engine.
• Enable image auto-rotation - Rotates the image to the orientation of the rest of the document.
• Remove shading from images before running recognition - Certain OCR operations can be slow if the image being processed contains shaded regions. Enable this option to remove the shading and potentially speed up OCR processing.

Word Merging Options

The value entered here represents the spacing between characters and how much space the engine will consider separate characters to be part of the same word.

Example: If the user scanned the word "W   O   R   D", the OCR engine would detect that as 4 words. Increase the tolerance and it will detect it as "WORD".

NOTE: This setting only affects OCR Assisted Indexing and Zone OCR Anchoring.