extract the pdf information

Dear sir,
I have a set of pdf files (300 files). I want to write a searching program that: allow user to search the desire pdf files by typing some criteria as: text, layout, image, ...Example: search all matching pdf files that contain the string "multimedia", or search all matching pdf files that contain the string "portal" and the size of this file is larger 100 pages, or search all matching pdf files that contain the images with a certain histogram.
The main issue that I encounter is: I donot know how to extract the text, image, layout of a pdf files. Would you please give me your instruction on how to build the such program.
Please mail to me at: [email protected]
Thank you in million.
hoang.

You can find the specifications for the .pdf format on http://www.wotsit.org. I had a look, and it is a very large (500 page) document; happy reading :sick: