this technique has worked very well for me most of the time (using a kindle). if it works, footnotes are removed, images are kept, no page numbers, etc.

1) convert pdf to .doc using convertfiles.com. this is the important part and doesn't work properly on other sites or on calibre. the conversion tool on this site is great at keeping the format of the pdf when converting to a doc file and at separating the headers and footers from the main text
2) download and install for free libreoffice or run the portable version of libreoffice.
3) download and install the free writer2epub addon for libreoffice
4) open the converted doc file in libreoffice writer and save as an .odt file. this can take time and use up lots of cpu power. the doc file should appear well formatted like the original pdf at this point. if not, the conversion process hasn't worked out well with that particular pdf.
5) click the writer2epub button and set any metadata required. if there is no cover image found automatically, click the no cover button
6) wait until the progress bar has ended, which can take a long time depending on size, and then the epub should be saved in the directory the original file was opened from.
7) use calibre or a conversion site to convert the epub to mobi

[webhill]

Thanks for the tips

FWIW, Atlantis saves doc, docx, txt, & rtf directly to epub and works quite well. There's also a freeware version of it, I believe.

[marooned2]

Thanks for the tips.
Trying it out on a book which I converted to .doc with Adobe Pro but lots of spelling/formatting mistakes!

ABBYY Fine reader 12 is good for converting PDF files - but the better the PDF file the easier the conversion and you really do need to spellcheck and proof.

ABBYY PDF Transformer is also good, but there is no subsitute for a good read-through proof and spellcheck.

But it is fun and with a scanner you are able to produce books for which no digital version existed - and then you have saved another one from being lost.

But it is fun and with a scanner you are able to produce books for which no digital version existed - and then you have saved another one from being lost.

Amen to that, jim3692!

My workflow:

1) Cropping with Briss to remove headers/footers.
2) Convert with Abbyy.
3) Post-processing and saving to epub with Word + ePUB tools add-in.
4) Final editing with Sigil.

I like Abbyy because of the control. E.g. if there's an image+text, sometimes you don't wanna ocr it but make the whole thing a .jpg
Tried convertfiles.com, it is pretty good! Might be using it for text-only pdfs.
Also experimenting with Scan Tailor. It can do page splitting, deskewing etc.

Links:
http://briss.sourceforge.net
https://toxaris.nl/en
http://scantailor.org

[Anonyma'am]

I don't do this often, but the last time I did I used http://pdf2docx.com/ to convert PDF to docx. Then used Word to edit the docx. Transfer the docx directly to Kindle by right-clicking and choosing Send to Kindle. Or copy docx to Calibre and convert to azw/azw3.

I often use the below website for ebook conversions.
Convert to AZW
Convert to ePub
Convert to FB2
Convert to LIT
Convert to LRF
Convert to MOBI
Convert to PDB
Convert to PDF
Convert to TCR

***website:
https://ebook.online-convert.com/convert-to-epub

[eohippus]

I save all my ebook downloads in my Google Drive. Cloud Converter works really well for me extracting files and converting pdf/mobi/epub files.

[Kandibar]

I also like using FineReader to convert from PDF scans or originals. You can then copy the text directly into Sigil once you've set up some generic templates i.e., I have some template files set up to save from doing all the HTML formatting from scratch. Depending on the quality of the original PDF, some corrections may be in order. Once I fix the formatting in FR I do the spell check and corrections after the text has been transferred to Sigil. Also, once you get the PDF in FR, save as HTML file to extract the graphics. I'm not saying it's the best way, but it certainly works for me.

Regards