Adobe Acrobat XI Pro - problems converting TIF G4 to PDF

[tilmaen]

Hi,

i scanned a very very nice score (surprise ) and put a lot of effort in cleaning out the pages and cropping the pages to the exact boundries of the printed stuff on the page (page number in the upper left/right corner, as well as plate number on bottom -> page size is roughly the same, for every crop).

My problem is the following: 5 or so years ago when i was still using windows xp - i would just select all the tiff files hit print and it would create a pdf through a ghostscript printer extremely fast (freepdfXP). those pdfs were clean, all the same page size (A4) and most important: the same size as the output images.
Now under windows 7 when i hit print, the adjustments for each page take seemingly forever, also when the output is handed to freepdfxp the resulting pdfs are just very very bad. in some cases it's not b&w any more, it looks more like a terrible jpg compressed image (pdf export). In other cases it is actually black and white, however there are many many white crosses embedded into black areas (screenshot attached) (tif 600 dpi export).

i tried copping the images in adobe professional and then saving it (after importing the raw tiffs from the scan in to adobe acrobat), however, the resulting pdf is 1,5 times the size as the original files, and the images again are not purely black and white. they look really good, but not tiff g4, so not good for me.

does anyone here have an idea as to how i could create a simple pdf with one pagesize for the whole document without any quality loss?

greetings
tilmaen

[tilmaen]

i created a clean pdf of the score on an old windows xp laptop, however, that shouldnt be the perfect solution. there must be a better way!?!

[Choralia]

does anyone here have an idea as to how i could create a simple pdf with one pagesize for the whole document without any quality loss?

I use the PDFill free PDF Tools, and I'm pretty happy with them.

Max

[tilmaen]

thx so much! you just made my day!

i just dont get why, apparently, it is too hard for adobe to make a decent software that would do the simplest tasks.

would you let me know what your workflow is?

- 1 i usually scan 600 dpi jpgs directly on my usb drive using a brother MFC-6490CW multifunctional thingy (nice scanner unit with A3 size glass, also the auto-document feed has come handy for scanning orchestral musich for our orchestra that was already on copy paper pages). I scan Jpg, because the firmware only supports 300dpi b&w scanning to medium (USB) - for some reason. i don't like starting every scan manually on my computer (after turning the pages on the scanner).
- 2 i convert the jpgs to tiffs (G4) and crop them to a reasonable size with irfanview bach processing (i usually scan the entire scaner area)
- 3 after that i manually correct tilted pages (fine rotation) and crop the pages to the exact margins of the printed stuff (almost no white edges)
- 4 finally i would use the software you just recommended to convert the tifs (differnet sizes due to cropping) into one pdf with a 0,5 inch edge. Page format is A4 (what do you use, A4, letter, or something entirely different?)

result is (now finally!) a nice looking pdf file. (so far i tried cropping in adobe and saving it there - the crop featuer in adobe professional is rather handy, but when saving it to a proper format doesnt work, thats not an option :/)

thx again! tilmaen

[Choralia]

My process is a little bit different because I begin with a 300 dpi, 256 level grayscale scan. I always use TIFF format to prevent JPEG compression artifacts, and cut pages from the binding so that they stay flat on the scan bed. The reason is that in most cases I have to perform automatic optical recognition of music, and a 300 dpi grayscale scan allows me to adjust the black/white detection threshold to the value that provides the best accuracy for the automatic optical recognition process following the scan.

Then, for publication on IMSLP, I convert resolution to 600 dpi, 256 level grayscale. I rotate, crop, cleanup, and finally select a black/white detection threshold that is best to remove shadows and so on. Image files are then saved as monochrome G4 TIFF, and assembled into a pdf file using PDFill tools. That's it, more or less.

Max

[tilmaen]

Wow! You put a LOT of effort in your scans!

however: if you scan 300dpi greyscale and later adjust to 600 dpi B&W you don't really gain quality information, right? Also you can not take away the binding from stuff you got at your University Library. I actually thought about doing it to printed music that i own, but decided against it. (i could have used the auto document feed a lot more, then, which has the huge advantage that i don't have to sit there all day and the pages are never "tilted"

anyways, thx for your input! very interesting!

greetz
tilmaen

[Choralia]

if you scan 300dpi greyscale and later adjust to 600 dpi B&W you don't really gain quality information, right?

Correct: you just improve the efficiency of TIFF compression. As an example, the original scan of "Messa di Gloria" by Puccini available on IMSLP was made by me using 300 dpi, 256 level grayscale, TIFF LZW compression, and the file size is typically around 1 MB per page. The same scan after conversion to 600 dpi, monochrome, TIFF G4 compression, is less than 250 KB per page. So, you get a reduction of the file size by a factor of 4 and more. Quite significant.

Also you can not take away the binding from stuff you got at your University Library.

I graduated 27 years ago . At that time I knew nothing about music (and still I don't know much - I just manage servers used by musicians, like this one), and there was no one music score at the library of my faculty (electronic engineering) .

More seriously, I use to receive work orders to prepare training aids for choirs (see my website), and the donations that I receive cover the budget required to purchase my own copies and then destroy them to make a good scan

Max

[tilmaen]

Wow, checked out your site! lots of work and nice results! i will recommend it to various choirs conductors. In fact, we'll play Mendelssohns Eliah in 2 weeks (Lohr, Germany) and my guess is you have that piece in your repertoire already ^^

have you ever thought about one of these:

http://www.youtube.com/watch?v=7MYvoJjV4VA (commercial, expensive, turns pages by itself)
or
http://www.youtube.com/watch?v=a4-qMc2QSOw (DIY, open hardware, cheaply self made and also fun to make - only downside: limited to "book size" but i guess one could easily adjust the sizes of that thing to allow for scanning large Scores (like conductors scores and so on - that thing does 800-1000 Pages per hour, if handled properly)
i've seen a DIY-scanner that turns pages automatically, but can't find it right now :/

greetz
tilmaen

[Choralia]

have you ever thought about one of these:

Honestly speaking, I didn't. Scanning time is not very important for me, as other tasks are much more time consuming in the process to prepare choral training aids. On average, the whole process takes around 30 minutes per score page. Scanning time, without any automation (but after having removed pages from binding), takes less than one minute per page, so it rather marginally affects the overall production rate.

Max

[tilmaen]

i checked their forums and it seems they have developed software that automatically rotates the pages so they are exactly rectangular, you can set the filters for the type of paper and print-paint once and it will apply the filters automatically and most importantly, you scan 2 pages per "scan" (photo).

i'll keep checking their forums, maybe one day the get even further in automating the scan process and i may find some time to build one of these nice machines

[mildldie]

http://www.pdfill.com/pdf_tools_free.html works fine.