There are libraries you can use to get content from PDF as from a document. But it is not as straightforward as it looks. You will not be able to read encrypted and obfuscated PDF files. Even more, there is no guarantee that the PDF content you are looking for is actually text and not raster image or vector graphics. On the other hand, even if you extract the text somehow, it will be hard to match your ROI.
So, if you don't use OCR, you will end up with a solution that it is working only in specific situations, but no general one for sure.
There are some OCR engines you could use for free, like
Tesseract[
^]. But as it has
no native PDF support[
^], you will need some pre-processing.
So I suggest you look for a good but not expensive commertial solution, like this one:
http://www.abbyy.com/ocr_sdk_windows/[
^] (Abbyy is really great).
On the other hand you could try
Adobe PDF IFilter[
^] with C# (
Using IFilter in C#[
^]).
In newer windows version there is an
OCR engine[
^] which could be used, but I have no further knowlede about it's capabilities.
Submit an article or tip