Acrobat has jumbled my pdfs!
February 9, 2010 5:39 PM Subscribe
I need to read a series of pdfs tonight - the trouble is, the resolution on them is messed up so as to be nearly illegible. Reading all of these will surely make my eyeballs shrivel up & fall out. With only access to Acrobat Reader, is there anything I can do to increase the readability of these pdfs - a way to change from color to b&w, increase the sharpness - anything?
If it helps, these were scanned at 200 dpi and show all the signs of being normal color pdfs - for some reason, the scanner these came from has suddenly turned text into pixelated nonsense. So bonus points if anyone can tell me why our scanner might have suddenly gone on strike?
My eyes & headache thank you in advance!
If it helps, these were scanned at 200 dpi and show all the signs of being normal color pdfs - for some reason, the scanner these came from has suddenly turned text into pixelated nonsense. So bonus points if anyone can tell me why our scanner might have suddenly gone on strike?
My eyes & headache thank you in advance!
Response by poster: emailed, chudmonkey - thanks!
posted by AthenaPolias at 5:54 PM on February 9, 2010
posted by AthenaPolias at 5:54 PM on February 9, 2010
You could try inverting the colors...sometimes that makes things easier on the eyes. But no guarantees. Here's a process for doing that sort of thing for your entire screen in Windows (I can't confirm these as I'm on linux right now).
But yeah, it would be easier to make suggestions if you could post a screenshot somewhere or something.
posted by Salvor Hardin at 5:56 PM on February 9, 2010
But yeah, it would be easier to make suggestions if you could post a screenshot somewhere or something.
posted by Salvor Hardin at 5:56 PM on February 9, 2010
Ah, well, maybe chudmonkey will take care of it for you :)
posted by Salvor Hardin at 5:56 PM on February 9, 2010
posted by Salvor Hardin at 5:56 PM on February 9, 2010
Best answer: Without seeing them, it's hard to say what sorts of repair they'd qualify for, but it sounds pretty bad. I was going to suggest one of the online optical character recognition utilities, so you could get some actual text to paste somewhere else (OCR isn't perfect, but can often be good enough), but if it's so bad you're having trouble reading it, then such a utility might have a hard time recognizing characters.
Another thing to google would be "clean up pdf" and look at the various things you find there in terms of free demos and trialware. For example, I just randomly found REpair, an add-on for pdfREnew. I haven't used either but they appear to have a free trial. Looks like there are tools such as darkening the text, etc.
As for why your scanner didn't do right, you might have a look at your compression settings. If somebody fiddled with those on the previous use, trying to get a really small file size, it's possible you have something that has been squooshed farther than it wants to be squooshed.
posted by Askr at 5:57 PM on February 9, 2010
Another thing to google would be "clean up pdf" and look at the various things you find there in terms of free demos and trialware. For example, I just randomly found REpair, an add-on for pdfREnew. I haven't used either but they appear to have a free trial. Looks like there are tools such as darkening the text, etc.
As for why your scanner didn't do right, you might have a look at your compression settings. If somebody fiddled with those on the previous use, trying to get a really small file size, it's possible you have something that has been squooshed farther than it wants to be squooshed.
posted by Askr at 5:57 PM on February 9, 2010
I haven't yet received the files, AthenaPolias - if the total size of the attachment(s) is over 5 megs, could you break them up or send me an attachment-free e-mail to make other arrangements...
posted by chudmonkey at 6:05 PM on February 9, 2010
posted by chudmonkey at 6:05 PM on February 9, 2010
Best answer: Having looked at one of the PDFs, I would say that the scanning was not done at 200 dpi, or something else happened during the creation of the PDF to substantially alter the resolution. It looks more like 20 DPI to me... I tried to run the document through some freeware OS X OCR software and got nothing usable.
Here's a look at what we're talking about.
posted by chudmonkey at 6:46 PM on February 9, 2010
Here's a look at what we're talking about.
posted by chudmonkey at 6:46 PM on February 9, 2010
Would printing them help at all? I've had to read pixelated PDFs before (although not as bad as what you apparently have) and printing improved them, IIRC.
posted by MadamM at 7:00 PM on February 9, 2010
posted by MadamM at 7:00 PM on February 9, 2010
2nding printing them out. Much less glare. You can even use a glass on them.
posted by Ironmouth at 7:29 PM on February 9, 2010
posted by Ironmouth at 7:29 PM on February 9, 2010
What happened to them is very aggressive, lossy JPEG compression. These won't get any better -- you can't make up data where it's been lost, and the artifacts are pretty severe.
posted by squid patrol at 7:56 PM on February 9, 2010
posted by squid patrol at 7:56 PM on February 9, 2010
I tried importing the jpeg Chudmonkey posted into iPhoto. I then chose edit and selected "effects" which has an option to "boost color"; I let it boost to the maximum choice (9), and it did improve it a bit. So, you could try this, though I don't know if the effort would be worth it, since you'd have to save all the pdf files as jpegs first, and probably go through the "color boost" edit page by page.
posted by gubenuj at 8:42 PM on February 9, 2010
posted by gubenuj at 8:42 PM on February 9, 2010
You could futz around with Levels in Photoshop; maybe boost the resolution, apply a blur, and then mess with the levels sliders until you got something legible. But that'd be one page at a time, and a lot of importing and exporting. I don't know of a way to do that in Acrobat, sadly.
posted by bink at 9:56 PM on February 9, 2010
posted by bink at 9:56 PM on February 9, 2010
If this were on a TV show instead of IRL, some quirky artsy character would just wiggle her fingers on a graphical keypad of some type (I'll just enhance the optical resolution and compress the image onto the flux capacitor...) and it would be perfectly legible in 5 seconds, including the fingerprints of the person who printed it and the make/model/serial number of the printer.
So - maybe send it to Hollywood and see if they can do anything with it ??
I need to quit watching so much TV
posted by CathyG at 8:54 AM on February 10, 2010
So - maybe send it to Hollywood and see if they can do anything with it ??
I need to quit watching so much TV
posted by CathyG at 8:54 AM on February 10, 2010
These look like they were scanned to .jpg and over-compressed, before making the .pdf. Using Chudmonkey's image and Gimp, I raised the contrast and applied an unsharp mask, which took out most of the .jpg fringing. It didn't fix the text, but it was somewhat more readable with the .jpg noise gone.
posted by TruncatedTiller at 5:30 PM on February 10, 2010
posted by TruncatedTiller at 5:30 PM on February 10, 2010
Response by poster: Thanks, everyone. I ended up waitingt until my library re-opened after multiple snow cancellations to read them in hard copy. As for the jpeg compression ideas - the files weren't processed that way, oddly enough. I just now tried to scan something else with that same scanner, playing with all of the settings - grayscale, b&w bmp, color, varying resolutions, etc - and nothing changed it. The scanner claims it is producing 200dpi pdfs, though changing the resolution to 400 dpi didn't change what the output looks like. Something is seriously effed up here...
Anyhow, thanks again, very much.
posted by AthenaPolias at 6:34 PM on February 10, 2010
Anyhow, thanks again, very much.
posted by AthenaPolias at 6:34 PM on February 10, 2010
This thread is closed to new comments.
posted by chudmonkey at 5:48 PM on February 9, 2010