# I want \includegraphics{} to crop to the bounding box of an included PDF

Apparently, it's possible to use `gs` to automatically find the bounding box of a PDF. See https://stackoverflow.com/questions/10962235/how-to-find-blank-page-in-pdf-file/11274720#11274720.

What I would like to do is to have a way of including a bunch of images such as the bounding box is scaled to the box in which I want the image to be shown.

Any thoughts on how to do that?

• Just to be clear: Your included image has some surrounding white space that you want to automatically trim and insert into your document constrained to some fixed width and height? Why not just pre-process the (all) image(s) to have tight bounding boxes before you include them? – Werner Jul 5 '13 at 1:52

Instead of doing this using `\includegraphics{}`, I would suggest a workaround to achieve the desired goal. If you are running on Linux/CygWin you can do the following in command line prompt:

``````for i in *.pdf
do
oname=`basename \$i .pdf`-old.pdf
mv \$i \$oname
pdfcrop --margins 0 \$oname \$i
done
``````

After doing this, all the PDFs that were already in the directory will be renamed to `*-old.pdf` and the newly generated PDFs will be cropped tightly without white margin. Then you can use these PDFs in the TeX file without any trim option in `\includegraphics{}`.

• This method is as per Werner's comment. – Jagath Jul 5 '13 at 2:47
• Wow. That's great. I didn't know about `pdfcrop`. Thanks. – vy32 Jul 6 '13 at 1:08
• Ps - as an aside, `pdfgcrop` is a perl script that runs GhostScript – vy32 Jul 6 '13 at 3:39
• pdfcrop does not get the bounding box right, always, for me. Inkscape (command line/script) is still the only sure way I have found to correctly find the bounding box in an automated way, especially for text axis labels, and for whitespace beyond them. It's ungainly, though. – CPBL Jul 6 '16 at 15:38
• I tried `pdfcrop`, it turned a 1MB PDF file into an 8MB PDF file. – Alexey Nov 25 '17 at 11:02