6.0
apostrophes are treated as one character – the straight apostrophe. Thus,
you will never see right and left apostrophes in recognized text, even if you
attempt to train FineReader into recognizing them.
2. The way in which certain characters are recognized depends on their envi
ronment.
How to edit a user pattern
You may wish to edit a new pattern before you start using it, as an incorrectly trained pat
tern will result in recognition quality being adversely affected.
The pattern should only contain whole characters or ligatures. Characters with cut edges
and incorrectly labeled characters should be removed from the pattern.
To edit a user pattern:
1. Select the Pattern Editor item in the Tools menu. The Pattern Editor
dialog will open.
2. Select the relevant pattern and click the
Edit button in the dialog. The User
Pattern
dialog will open.
3. Select a character and click the
Properties button to edit the character
caption and set the correct typeface: italic, bold, subscript or superscript.
Click on the
Delete button to remove any incorrectly trained characters
from the batch.
User languages and language groups
In addition to the builtin languages and language groups, you may also create new lan
guages and language groups (made up of languages supported by FineReader) and use them
for recognition.
You may want to create a new language if you need:
1. To use a user dictionary.
z For example, you want to recognize an English text containing many
abbreviations. You therefore create an abbreviation dictionary, create a
new language and linkup the dictionary to the language. You then cre
ate a new language group consisting of English (using the application
dictionary) and your new language (containing the abbreviations dic
tionary), and use this language group to recognize your texts.
2. To recognize documents of a specialized nature, for example:
z supermarket productline lists containing only product codes. Product
codes are usually made up of numbers and a few letters. Consequently,
59
Chapter 6. Recognition