NUANCE OMNIPAGE PRO 9, Руководство пользователя

Страница: 105 / 114

OCR Problems
Technical Information 97
Text Does Not Get Recognized Properly
Try these solutions if any part of the original document is not converted
to text properly during OCR:
• Look at the original page image and make sure that all text areas
are enclosed by text zones. If an area is not enclosed by a zone, it
is ignored during OCR. See “Creating Zones for OCR” on page 22
for more information.
• Make sure text zones are identified correctly. Reidentify zone
types and contents, if necessary, and perform OCR on the
document again. See “Changing Zone Properties” on page 71 for
more information.
• Adjust the Brightness slider in the Scanner settings of the Options
dialog box. Lighten the setting for thick, run-together text
characters or dark backgrounds. Darken the setting for thin,
broken text characters.
• Make sure the correct main and secondary document languages
are selected in the Language settings. Only languages included in
the document should be selected. See “Language Settings” on
page 48 for more information.
• Select Use Language Analyst in the Accuracy. The Language
Analyst evaluates words and corrects likely errors during OCR.
See “Accuracy Settings” on page 46 for more information.
• Train OmniPage Pro to recognize special characters that might
normally be difficult to recognize, such as the copyright symbol
© or the registered trademark symbol ®. Do not train OmniPage
Pro for regular characters because it may interfere with
recognition. See “Training OCR for Special Characters” on page
75 for more information.
• If you use True Page as the Output Format setting, recognized text
gets put into frames (formatting boxes) in the text viewer. Some
text may be hidden from view if a frame is too small. To view the
text, place the cursor in the text frame and use the arrow keys on
your keyboard to scroll to the top, bottom, left, or right of the
frame.
• Check the glass, mirrors, and lenses on your scanner for dust,
smudges, or scratches. Clean if necessary.
OmniPage Pro only recognizes machine printed-text characters such as
typewritten or laser-printed text. However, it can retain handwritten
text, such as a signature, as a graphic. See “Do you want to retain
graphics in your document?” on page 58 for guidelines.