PDF Image Extraction

CaMelRyder has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: PDF Image Extraction by McDarren (Abbot) on Jul 16, 2006 at 05:03 UTC
I've not worked with PDF's myself, but I believe that PDF::Extract is the recommended module for extracting data from them. Although it doesn't appear to have a method to extract images - just pages. CAM::PDF::Renderer::Images looks like it might do what you want. Or maybe CAM::PDF::Content Hope this helps, Darren :)	[reply]
Re: PDF Image Extraction by Khen1950fx (Canon) on Jul 16, 2006 at 05:07 UTC
If you have Adobe, then go to: http://ict.cas.psu.edu/training/howto/util/pdf_extract_graphics_v6.htm It explains how to extract an image or images. If you have xpdf, then see the man pages for pdfimages and xpdfrc. Pdfimages will extract images, then I would recommend PDF::Create to rebundle your images. Darren's suggestion that you use CAM::PDF works too.	[reply]
Re: PDF Image Extraction by marto (Cardinal) on Jul 16, 2006 at 09:59 UTC
CaMelRyder, To second some of what McDarren has already advised, CAM::PDF comes with a script called extractallimages.pl which will 'Save copies of all PDF images to a directory'. Hope this helps. Martin	[reply]
Re^2: PDF Image Extraction by CaMelRyder (Pilgrim) on Jul 18, 2006 at 04:38 UTC
Thanks for the posts guys, the CAM::PDF seems to be exactly what I need. However, I'm having some pretty extreme difficulties in getting it to install. I managed to get all of the dependencies installed except for one trouble maker - Text::PDF::File. I install it over and over again and it installs successfully with no errors and no failed tests, yet everytime I try to get CAM::PDF installed it complains about Text::PDF::File not being there and attempts to install it again. Probably a versioning problem So I let and it bombs out during the testing phase because it can't find Text::PDF::FlateDecode. This module has completely alluded me. I can't find it anywhere. It doesn't come up in the depedency list either. While I've got your attention, perhaps I should elaborate a little on what my project is. I have about 100 pdfs which are technical journals on welding specifications. Welders pay alot of money for people to turn the specifications for a job into a set of legal documents so as to cover there butts if something goes wrong. This is how I view the project thus far and intend to tackle each step as a different problem: Step 1: -------- First have to extract the journals from the pdfs(which are stored as images). Step 2: ------- I will have to OCR the images in order to turn them into character data. Step 3: ------- Use a form of statistical analysis to turn the character data into a searchable database, which will much easier to custom taylor to the needs of the user. It will be long road and I really appreciate you're help at these early steps. ¥peace from CaMelRyder¥	[reply]
Re^3: PDF Image Extraction by CaMelRyder (Pilgrim) on Jul 25, 2006 at 20:19 UTC
I found a simple solution. I simply wrote a script to drive convert which is part of the ImageMagick package. It has the capability of pulling all of the images from a pdf and storing them in any format that I want. Now it's on to the OCR step. I think I'll do that one in the Java because of how easy it is to draw on the screen. ¥peace from CaMelRyder¥	[reply]