|
(continued...)
"If
the image is on top, I can't tell how good the OCR was...
right?"
Right.
Well, you can, but it's not immediately obvious to everyone,
and it requires Adobe Acrobat to do it. Take the
Edit Menu and "Copy File to Clipboard," then
paste from the clipboard into your favorite word-processor.
You should get an idea of the OCR quality fairly quickly,
but be warned — it won't look like your document.
Document
Solutions, Inc. distinguishes three levels of cleanup
on files converted to PDF/Searchable Image:
| DSI's
PDF/Searchable Image file subtypes |
Average
Text Accuracy |
| "Ex
Machina" pronounced [ex-mach-ina]; meaning "from
the machine" |
>99%,
sometimes over 99.5% on clean originals |
| "Search
Term Correction" |
99.95%,
"stop words" omitted from correction |
| "Full
Cleanup" |
99.995%
on text 7pt or greater |
Bear
in mind that the quality of the original document scan
is CRITICAL to the success of the whole conversion enterprise!
So
who uses PDF/Searchable Image, and for what?
Click
on above link to continue...
|
