Configure the FineReader OCR Engine - Designer - Foundation 23.1 - Foundation 23.1 - Brainware - external

To configure the FineReader11 OCR engine, complete the following steps.

1. In the Field OCR Settings Properties dialog box, under Recognition Type, select OCR.
2. From the Available Engines list, select FineReader10 or FineReader11.
3. Below Recognition Trade Off, depending on whether performance or OCR accuracy is more important, select one of the following settings.
   • To decrease the recognition error rate accepting a higher processing time, select Most Accurate. This is the default mode.
   • To perform the recognition between Most Accurate and Fast, select Balanced.
   • To perform a fast recognition accepting less accuracy, select Fast.
4. Below Rotation, select any of the following options.
   • To automatically detect the page orientation, select Detect Orientation.
   • To automatically rotate the document so that the text is vertically aligned, select Deskew.
5. Below Enhancement, select any of the following options.
   • To remove all horizontal or vertical internal lines, select Remove Lines.
   • To removes background noise from the image for recognition, select Remove Texture.
   • To remove speckles from the documents, select Despeckle.
     Note: Speckles are a group of black pixels surrounded by white pixels or vice versa.
6. Click the Recognition tab and select any of the following options.
   • To automatically detect how the documents have been printed, select Automatic:

     Use this option unless your document input is very homogeneous and from a dot-matrix printer or a typewriter.

   • For documents printed with dot-matrix printers, select Matrix.
   • For documents created with typewriters, select Typewriter.
   • For documents with OCR-A and OCR-B fonts, select OCR-A / OCR-B.

     This is typically where high OCR character recognition rates are needed. The fonts are optimized for machine character recognition. For example, they are used on passports and other security documents.

   • For documents using MICR E13 B fonts, such as checks, select MICR E13 B.
   • To improve the recognition of text specific to sales receipts and invoices on low quality documents, where the OCR process focuses on tidying up the noises around the actual text, select Receipt.
     Note: This setting is only available in FineReader 11.
   • To recognize characters or symbols written slightly above right or left to another character, such as “™” in “Company Product™”, select Allow Superscript.
   • To recognize characters written slightly below another character, such as the “2” in the chemical formula “H2O”, select Allow Subscript.
   • To improve single digit columns recognition when using CJKT languages, select Small Digit for Asian Enh..

     This setting is only enabled if at least one of these languages is added to the list of used languages on the Language tab, such as ChinesePRC+English. For more information, see About Non-Western Languages.

7. Optional. To add a language, click the Language tab and complete the following substeps.
   1. From the Installed list, select a language and then click the arrow button to move it to the Used list.
   2. Optional. Repeat the previous step to add other languages only if you are processing multilingual documents.
   Note: The FineReader engines process the recognition with the languages in the Used list.