Does anyone know of a conversion library that can convert a PDF into
text and an image at once where the positions of the textual elements
within the image are also output?

There are many open source and commercial PDF to text, e.g., pdf2text,
pdftron, etc., and PDF to image converters, e.g., imagemagick, adobe
pdf library, but none that are able to convert to both at the same
time with the added information of where the PDF elements are located
in the resulting image.

stormrunner-ga

Just to confirm do you mean:

a) the PDF contains text, and an image layed out in seperate areas within a page,

b) the PDF contains an image, overlayed on which is some text layed
out in the same area within a page?

What sort of document type is the original? It is not a plain ASCII
text file as that has no layout information, are there any
constraints?

affordablemagic-ga

What about converting a PDF file to an Micro$oft Office equivalent,
but open-source OpenOffice (www.OpenOffice.org)? That will at least
save the pic positions.