Suggestions for scanner settings for pdf book

Discussion in 'How To' started by dongcopter fruitcake, Jan 12, 2014.

  1. I desire to scan an older book (in public domain) for the purposes of internet archival. I have a working knowledge of graphics and know that I need to optimize the settings on the devices for the best result. Also any info on DIY scan to pdf book compilation would be very appreciated.

    I have a household flat bed scanner, but am hoping to use a cradle book scanner. I have photoshop, etc, but I'm hoping not to have to manipulate too much (500-odd pages of shitty type).

    Thanks in advance.
  2. Anonymous Member

    In that your project involves text, do you plan for an OCR Scan?
    • Like Like x 1
  3. I wish I could, but the text is often blurred out of focus, the result of reproduced, photoscanned, photographed dox from the seventies era. I don't think OCR would work well. Besides, I think it is important that it is initially a photoscan.
  4. Cocks, I'm sorry, I meant to reply. See post above :).
  5. Anonymous Member


    I suggest making a couple or more test scans of examples of the poorest state the text is in, and try setting a variety of resolutions for the tests.

    I speculate that 72 pixels per inch is too low, so try 180 ppi.

    If 180 is too low, go to 270 ppi or 300 ppi. I think 360 ppi would be overkill and too data intensive.
    • Like Like x 1
    • Like Like x 1
  6. Ogsonofgroo Member

    Agree with Night Owl, take stuff to the best resolution in stages, just so that yer not posting/printing something at way higher res than needed. Go for optimal, but not beyond, there is no point in that. With old docs that have been scanned and posted numerous times until they are unreadable, then its worth the time to go hunt a better copy sometimes. Sometimes it might be a consideration to put it into something like Paint to maybe give it some contrast, not altering the docs, just make them more readable. Just sayin' :) Hey, hittin' some of the old archives can be a trip ya know.

    • Like Like x 1
  7. Thanks. My biggest hurdle at this point is to find a cradle book scanner with public availability in my area, old book is old and rare enough I won't hack it up to scan it.
    • Like Like x 2
  8. Asheera Member

    In my experience the best results come from using the highest possible scanner settings and saving the results to a lossless image format (eg: TIFF). ONLY THEN do you think about reducing the filesize (like with a batch command in photoshop or your PDF software). Scanning is the time-consuming part of the process, and the portion with the most problems - getting the highest quality scans you can means that, later, you can do whatever downsizing you need to do with the highest quality source available. You can repeat a batch command easily, but you cannot repeat an hour or two of hard scanning - so don't reduce the scanner settings.

    As for the next stage of turning your high-quality images into a decent-sized PDF, well that will depend on both your needs and what software you have access to. For example, if you have a fully copy of Acrobat Pro simply importing the TIFFs and using the in-built options to downgrade the images will, after a little experimentation, get you the results you want. Freeware combinations can also achieve decent results (eg: Irfanview for batch downgrade/crop and pdfMate to make the PDF) if you put time in learning how to use them.

    Generally, in my experience, there are no good freeware OCR solutions so you may want to pay a few bucks if this is needed. Omnipage is pretty good if you can afford it.

    Bottom line - turn the scanner up and only later worry about the filesize. Hi quality settings for a book would be about 10Gb of TIFF, and these can always be deleted after you complete the project.
    • Like Like x 5
  9. Thanks, Asheera. Helpful to know the TIFF thing. I was messing around with JPGS last nite and scans directly to PDF (Using NAPS 2), but perhaps TIFF would probably be a good intermediate step and helpful with the OCR. If this thing gets as far as OCR, I'll probably have some real help. I mostly want to get decent photo dox suitable for archival on the internet for this volume.
    • Like Like x 1
  10. OK, I have ended up somewhere around 150 megs. I used adobe acrobat 8 (CS2) with ocr and indexing enabled (although the poor quality of the source seemed to foul them up at times). Can this be compressed further? And can I get any suggestions on where I can stick my, I mean post it on the internets for incurable insomniacs to download for it's soporific properties?

Share This Page

Customize Theme Colors


Choose a color via Color picker or click the predefined style names!

Primary Color :

Secondary Color :
Predefined Skins