Imagemagick / Ghostscript: converting a pdf into a jpg

[Jonne]

Hi,
I'm trying to write a shellscript that converts pdf-s into jpg's, but the jpg's i get as a result are pixelated. Is there a way to make sure I get jpg's with a decent quality? The source pdf's are A3-sized, and the jpg's I get are about 1035x1474, while I need them to be almost double that size to make sure the text is readable.

this is the command I use:

convert 1.pdf -profile EuroscaleUncoated.icc -profile /usr/share/color/icc/sRGB.icm 1.jpg

these are the versions I'm using:
Imagemagick 6.2.4
Ghostscript 8.15.4
the OS is Ubuntu Feisty

Is there a way to improve quality with a switch or something?

[WW]

When converting to jpg, you can use the -quality option. The "best" quality would be -quality 100.

[Jonne]

-Quality changes the quality of the jpg compression, after rasterizing. The problem is that I want jpegs of a bigger resolution, and ghostscript/imagemagick won't allow me (even with -density and other options i tried)

[lazka]

try "-density 200x200"

edit: ok i didnt read you last post.. density is working fine here, tough.

[Jonne]

Thanks for trying anyway ;) This thread has had no comments for weeks, and I still haven't found an adequate solution. Anyway, i've tried again, and it still desn't work. I think there's something wrong with how imagemagick interfaces with Ghostscript.

If I import the pdf in the gimp (I believe the gimp also uses gs internally for pdf conversion), I can set the resolution to whatever I want, and it works. The colours are all wrong, though, because I can't use a colour profile to convert the colours as they should.

I want to do it through the command line, because it will be part of a script that converts a bunch of pdf's into jpg's automatically.

i want to replace an existing program that runs only on Windows, and uses Photoshop for the conversion, because I don't want to:
1, depend on Photoshop & Windows
2, have photoshop take all the resources and screen estate while it's running


edit: I tried with Krita, and Krita gets both the colours right, and the resolution. Is there a way I could use Krita for this, without having it spawn its gui?

[duff]

Thanks for trying anyway ;) This thread has had no comments for weeks, and I still haven't found an adequate solution. Anyway, i've tried again, and it still desn't work. I think there's something wrong with how imagemagick interfaces with Ghostscript.

Have you tried their mailing list?
http://www.imagemagick.org/mailman/listinfo/magick-users

[WW]

You can change the size of the output file with the -geometry option, e.g. -geometry 1600x1600 will ensure that the largest dimension is 1600 pixels (it will preserve the aspect ratio). I don't know if this will get you what you want, but it's worth a try. I tried it with a PDF file that contains just some text and equations, and it looked OK, but not great.

EDIT...
If I combine -geometry, -density, and -quality, it looks better. For example,


$ convert -geometry 1600x1600 -density 200x200 -quality 100 file.pdf file.jpg

Then file.jpg looks pretty good to me.

[Jonne]

I think I found it, thanks to WW. Apparently it depends on the syntax you use. If you add the switches before giving the filename of the pdf, it will not ignore -density and such. If you give the filename first (like in the example I posted), Imagemagick will first rasterize the pdf, and only apply the switches afterward, thus giving degraded quality. to recap:
bad:
convert file.pdf -density 200x200 file.jpg

good:
convert -density 200x200 file.pdf file.jpg

thanks guys :)

[itaylor]

Wow, I just wasted 45 minutes trying to convert a PDF to JPG using every imagemagick option I could find. The default resolution looked horrible. I finally found this post and the problem is solved. Thanks! I'm sure the information is there on the imagemagick website somewhere, but it sure isn't obvious.

[ridgeland]

Thanks WW
I used your command line.
Worked fine for me. I converted a one page pdf to a jpg.
I googled "Ubuntu pdf to jpg" and found this quick solution.

When I opened this jpg in the GIMP I saw that the GIMP can import pdf and then save as a jpg. Both ways worked. CLI or GUI.

[danbar]

Yes the operation was successful for me too. I used the gimp. But now that I tried changing more than one page document I saw that only the first page was changed into a jpg file. Is there a solution for longer documents ?

[Jonne]

you could use pdftk to split them first.

[danbar]

Jonne,
Thanks for your time and patience. You wrote : you could use pdftk to split them first. Could you explain more how to go about it ?

[Jonne]

basically you install pdftk (using synaptic or apt-get), then you use

pdftk mydoc.pdf burst
to split the pdf into multiple pages. After that you can run convert on the resulting single-page pdf's.

Read pdftk's manual for more info on its flags and stuff.

[danbar]

This is what the terminal gave me.....Error: Failed to open PDF file:
provapdfjpg.pdf
Done. Input errors, so no output created.

[Jonne]

i guess there's something wrong with your pdf, then :s . I've never seen pdftk fail (but that's not saying much, I rarely use it). Can evince or xpdf open it properly?

[danbar]

Yes i opened it with Evince and it's ok !

[danbar]

By the way how can I open Pdftk ?

[ridgeland]

Hi danbar,
I can get the same error if I do

pdftk nofile.pdf burst

What did you use? Full valid pathname?
You probably just need the full path name like
pdftk /home/me/myfile.pdf outfile

Also use
pdftk --help
to learn more
or
man pdftk

[k8b]

There is a muck simpler way to split multipage pdfs into a jpg:

convert -quality 100 -density 600x600 multipage.pdf single%d.jpg


The -density option defines the quality the pdf is rendered before the convert > here 600dpi. For high quality prints you can increase that number.
The %d just before the jpg suffix is for automatic numbering of the output pages 0,1,2...
The -quality option defines the compression quality of the output jpg (0 min ... 100 max)
The .jpg suffix defines the output format. You could use .png/.jpg/.pdf


On linux I recommended to use png files, because of patent issues of jpg's. If pdf output is used, the pdf will saved as an image.

Note: If you need to crop the pages add "-crop 1000x750+1450+150" between "convert" and "-quality"

[modemlamer]

I think I found it, thanks to WW. Apparently it depends on the syntax you use. If you add the switches before giving the filename of the pdf, it will not ignore -density and such. If you give the filename first (like in the example I posted), Imagemagick will first rasterize the pdf, and only apply the switches afterward, thus giving degraded quality. to recap:
bad:
convert file.pdf -density 200x200 file.jpg

good:
convert -density 200x200 file.pdf file.jpg

thanks guys :)

thx man, u saved my day

[dhjdhj]

I'm having a related problem --- I have a multipage PDF file which contains one image per page. I have tried using convert to convert that PDF into images (using the %d in the jpg filename) but I'm only getting the first page.

I have used convert in the past to convert multipage PDF files into a sequence of jpgs with no problem. Anyone know why this might now be failing ---