Due to the weird internal structure, I suppose that calibre-conversation will fail on this particular PDF. AFAIK calibre tries to extract the text form the PDF and is not using any OCR-technics.
I think I remember doreenjoy was recommending an ABBYY product:
FineReader or PDF-Transformer. Both can treat the PDF as picture and use OCR on "reading" the text, not trying to extract it.
|