|
|
| Author |
Message |
qshadow
Joined: 29 Jan 2010
Posts: 3
|
 Posted: Fri Jan 29, 2010 2:28 pm Post subject: Sharpen text of an old scanned magazine |
 |
|
|
|
|
|
|
 |
pixel8or
Joined: 31 Oct 2009
Posts: 142
Location: Ireland
|
| Posted: Sat Jan 30, 2010 1:29 pm Post subject: |
|
|
I used Levels in Text1 and I used UnSharp Mask with the following settings in Text2, Amount -> 109 and Radius -> 20. |
|
|
|
|
|
qshadow
Joined: 29 Jan 2010
Posts: 3
|
| Posted: Sat Jan 30, 2010 2:24 pm Post subject: |
|
|
| pixel8or wrote: | | I used Levels in Text1 and I used UnSharp Mask with the following settings in Text2, Amount -> 109 and Radius -> 20. |
Wow!
the difference is Astounding!
Thanks pixel8or! |
|
|
|
|
|
qshadow
Joined: 29 Jan 2010
Posts: 3
|
| Posted: Sun Jan 31, 2010 6:06 am Post subject: |
|
|
By the way,
i have hundreds of pages that need this fix,
any idea how can i do it automatically?
maybe some batch command or utility?
Thanks,
Qshadow. |
|
|
|
|
|
magkelly
Joined: 04 May 2010
Posts: 7
|
| Posted: Tue May 04, 2010 1:46 pm Post subject: |
|
|
In all honesty just fixing each scanned page in a large document would take you forever and it still wouldn't be ideal for reading. You're far better off making a pdf file of all the page images from the session and running that through a good OCR software like Omnipage. You might have to go back and make some spelling corrections where the OCR misses something but you'd likely get much cleaner type and a far more readable document out of doing it that way. I'd do that or maybe use voice recognition software. It would save you a lot of time.
How many hundreds of pages are you talking? Are you scanning a digital back up for the periodical in question here per chance or maybe planning for digital distribution? If so, if you are doing lots of issues OCR or talk type is definitely the way to go. Just correcting scans would take you ages for one magazine.
I'm not seeing any illustrations to work with here, but you could always scan those separately, clean them up, and set it back up with the text in a desktop publishing application if you needed to. But getting clear text sans OCR/voice typing will be time really consuming otherwise.
I used to do some scanning for a publishing house that was converting a lot of books over to e-book format? Doing it sans OCR/VT software it could take us weeks to do one 400 page book using individual scans and often the results were iffy. Using the software I could get one done in a week usually, sometimes less depending upon book length and the end result was always better.
It's tedious work no matter how you look at it, but it can be made a whole lot easier with the right software. |
|
|
|
|
|
|
