I have a multi-page pdf file containing two images on each each page. I would like to extract each image into a separate file (preferably but not necessarily in jpeg format). This is a one-off task so I am looking for a simple straight-forward method using free software. Is this possible? If so how can this be achieved please?
This can be done using Acrobat Pro. There's a free trial version of this application that you can use for 7 days.
The term “extract” in this case seems to be a misnomer.
In other words, it may be named incorrectly.
I think many users will find that, upon extracting images from a PDF, the images are not the same as the original images which went into the PDF. This has been the case with all my testing in Adobe Acrobat 7.0 Professional where there are 3 different ways to get images out of the PDF, before we even factor-in the various options for file type, but none of these methods produces an image which looks as good as the original.
I have the originals sitting in a web page resources folder for comparison. There is loss of image quality no matter how I “extract”. Therefore, I think a better name for it might be “translate” or “recreate”.
However, I am very interested in being proven wrong here, so please do educate me. If there's a way to really “extract” the original images, I want to know.
What does you mean with "original images" ?
My original image started as a PNG at 72 pixels/inch before entering the PDF.
If I “extract” it from the PDF as a PNG at 72 ppi, the resulting image looks horrible.
If I extract it as a PNG at 96 ppi, it also looks horrible.
At 150 ppi it looks reasonable, but is blurry, not crisp like the original, and the file size of course is much larger (as a percentage).
I have tested all output methods using Adobe Acrobat 7.0 Professional (actually 7.1.4, fully updated as far as I can tell), using all file types and resolutions up to 300 ppi. The only thing that looks as good as the original is .pict format, and it looks crisp at 72 ppi.
However, that format has mostly been killed-off, so… what's the solution here?
Is there a solution?
Oh, sorry, I just remembered that I didn't test it with JPEG2000. That was the only one I avoided.
I know you don't like to hear it, but it's impossible to discuss issues that you're having with Acrobat 7. If you could share a sample file with us we could test it in a more current version of Acrobat and see what the results are. If they are the same then we could discuss it. If the issue was fixed since your version then the problem is yours to deal with...
Images have arrived.
Don't worry, I think the info in the box is not personally identifying, otherwise they wouldn't have put it up on the internet. This is the original (72 dpi png) taken from the web page:
Notice how it actually looks good.
The following were all taken back out of the PDF by Acrobat 7 Professional after it was “printed”.
1. (72 dpi png)
Notice how this one is not only smaller, but, when scaled-up to the same size, looks much worse than all the others.
Don't ask me why it's this particular resolution. I didn't specify it.
2. (119.01 dpi png)
Pay attention to the:
T in Assistant.
CCO in Account.
R in dsandras.
T in later.
And, the descenders on the whole last line.
And, the third one.
3. (150 dpi png)
Notice how it's really blurry.
The 300 dpi .png version looked blurry too. The .tiff and .jpeg versions it made looked the same as the 2nd version up there with deformed letters. Only the .pict version looked as good as the original.
Edit: The .tiff version looks a little worse than the .png files above. It has more noticeable deformation of the letters. The .jpeg has the added penalty of visible artifacts where there are fields of color, and I always set the quality slider to just under maximum for .jpeg files.
Bonus image number 4, the PICT.
I thought it was perfect, but I spoke too soon.
Just noticed that there are 2 vertical lines, looks like 1 or 2 pixels in width, running down the image where the data is just missing. It can be seen only when flipping back and forth between the original and this image. Actually, the 4th dot in the password field gives it away.
Also, there are 2 black ears on the top corners where the PICT doesn't handle the transparency, or alpha channel(?), correctly.
4. (72 dpi pict)
Notice that I had to convert this image to PNG because the graphics professionals at Adobe haven't enabled PICT file uploads to their website! : D
Thank you. That was one of the 3 ways which I tried while liberating the images from the PDF, as noted above. Anyway, the more help, the merrier.