OCR options

Selecting the right OCR options is important if you want fast and accurate results. When deciding which options you want to use, you should consider not only the type and complexity of your document, but also how you intend to use the results. The following groups of options are available:

You can find the OCR options on the Recognition Languages tab of the Options dialog box (click Tools > Options... to open this dialog box).

ABBYY FineReader automatically recognizes any pages you add to an OCR project. The currently selected options will be used for OCR. You can turn off automatic analysis and OCR of newly added images on theImage Processing tab of the Options dialog box (click Tools> Options... to open this dialog box).

If you change the OCR options after a document has been recognized, run the OCR process again to recognize the document with the new options.

PDF recognition modes

These settings are intended for PDF documents with text layers and pictures. Such PDF documents are usually created from documents in editable formats. Other types of PDF documents, such as searchable PDFs and image-only PDFs, are always processed in Use OCR mode, which is the default mode for these types of PDF. No additional settings are required to process such PDFs.

Three recognition modes are available:

  • Automatically Choose between OCR and Text from PDF
    The program will examine the text layer and use the existing text layer if it contains good-quality text. Otherwise, OCR will be used to create a new text layer.
  • Use OCR
    OCR will be used to create a new text layer. This mode takes more time but is more suitable for documents with poor-quality text layers.
  • Use Only Text from PDF
    This is the default mode for PDF documents with text layers. The program will use the original text layer without running OCR.

balance between OCR speed and accuracy

ABBYY FineReader 14 allows you to perform:

  • Thorough recognition
    In this mode, ABBYY FineReader analyzes and recognizes both simple documents and documents with complex layouts, even those with text printed on a colored background and documents with complex tables (including tables with white grid lines and tables with color cells).
    Thorough recognition requires more time but delivers better quality.
  • Fast recognition
    This mode is recommended for processing large documents with simple layouts and good quality images.

Print type

Specifies the type of device that was used for printing the document. See also: Document features to consider prior to OCR.

Detect structural elements

Select the structural elements you want the program to detect: headers and footers, footnotes, tables of contents, and lists. The selected elements will be clickable when the document is saved.

Barcodes

If your document contains barcodes and you wish them to be converted into strings of letters and digits rather than saved as pictures, select Convert barcodes into strings. This feature is disabled by default.

Patterns and languages

Recognition with training is used to recognize the following types of text:

  • Texts with special symbols (e.g. mathematical symbols)
  • Text with decorative elements
  • Large volumes of text from low-quality images (over 100 pages)

Training is disabled by default. Enable Use training to recognize new characters and ligatures to train ABBYY FineReader while performing OCR.

You can use either built-in or your own patterns for recognition.

You can save and load settings for user patterns and languages. See also: OCR project.

Fonts

Here you can select the fonts to be used when saving recognized text.

To select fonts:

  1. Click the Select Fonts... button.
  2. Select the desired fonts and click OK.

11/2/2018 4:19:17 PM


Please leave your feedback about this article