PDF Structure recogntion with PDFNet

[Support]

Q:

We are trying to build an ontology based search engine. We are running into a problem where we need to identify some sections like tables headers, headings, footers etc. from a pdf document. Is there a way to accomplish this using pdftron?

--------

A:

PDFNet could be used to implement structure recognition (or to extract existing structure, if available).

For some background on structure recognition please see:

   http://blog.pdftron.com/2014/03/02/table-extraction-and-pdf-to-xml-with-pdfgenie/

PDFGenie technology described in the article is also available as a PDFNet SDK add-on (in 'pdftron.PDF.Convert.ToHtml()' when SetReflow() option is set ).