NUANCE PAGEKEEPER PRO 3, Руководство пользователя

Страница: 103 / 118

Performing Background OCR on Images
Technical Information 97
Improving OCR Accuracy
High OCR accuracy enhances your results when searching for images in
PageKeeper. The following are ways to increase OCR accuracy.
Select Settings for Higher Accuracy
Select a grayscale or color scan mode in PageKeeper’s Scanner settings
if you are scanning poor-quality pages, pages with very small type, or
pages with text on colored or shaded backgrounds. If you scan in
grayscale, select a high resolution as well. Grayscale and color settings
are only available with supported grayscale and color scanners.
If you do not have a grayscale or color scanner, experiment with the
Brightness setting in PageKeeper’s Scanner settings to get a good scan.
Lighten the setting for thick, run-together text characters or dark
backgrounds. Darken the setting for thin, broken text characters.
Use High-Quality Images
• Try to scan original documents rather than photocopies. Typeset,
high-quality printed page images return the best OCR accuracy.
With lesser-quality pages, OCR accuracy may be poorer.
• Ask senders to select Fine or Best Mode when they send you a
fax.
• Page images should be free of notes, lines, or doodles. Any
character distorted by a mark may be unrecognizable.
Rather than making annotations on your paper documents, make
text, line and scribble annotations in PageKeeper View. See
“Annotating Documents” on page 62 for more information.
• Text in page images should be reasonably clean and crisp.
Characters should be separated from each other and not blotched
together or overlapping.
• Text in page images should not be underlined. It is difficult to
recognize underlined text because the underline changes the
shape of descenders on the letters q, g, y, p, and j.