Ricoh Aficio FX10 Operator's Manual page 304

This section explains how to achieve the best recognition accuracy
possible. See the online Help for more detailed information.
OCR accuracy depends on the quality of the scanned image. You'll have
very few errors if you scan sharp, laser-printed documents. Scanning
faxes, dot matrix documents, and poorly printed pages may produce
unacceptable results.
Consider the following when you scan a page:
• The print should be reasonably clean and crisp.
• The document should be free of notes, lines, or doodles; anything
that is not a printed character will slow OmniPage Limited
Edition considerably.
• The document font must be non-stylized; for example, the Zapf
Chancery font is too fancy for OmniPage Limited Edition.
• It's hard to recognize underlined text accurately; the underline
changes the shape of descenders on the letters q, g, y, p, and j.
• The scanner glass must be clean.
If you scan a foreign-language document and you get many OCR errors,
check that the document's language is selected as the Recognition
Language in the Tools menu/Language dialog.
If there is no listing for the language, upgrade to OmniPage Pro, which
recognizes 13 language-character sets.
Improving OCR Accuracy
Scanning Tips - 16