Reduce size of pdf file with ocr and tables

Question

I have a scanned file, where text has been recognized already, yet it weights a whooping 80 mbytes. During the scanning process I did a silly thing such as scanning in grayscale. Now the pdf has a huge swarm of dash and stroke marks in it. I believe that the size comes from the fact that all those stains.

Simple conversion with doesn't seem to be the solution here

ps -dPDFSETTINGS=/screen    

This results in fonts getting jagged and unpleasant to read. Pretty much every other question about reducing size is solved by downsampling involving one or the other conversion that hurts text more than anything else.

Is there a way to discard that background information and preserve the actual text?

Answer 1

If you would like to have only the text, without much formatting, you could try this command:

less yourfile.pdf >> output.txt

If the document did not have the "swarm of dash and stroke marks" in the original scan, I normally have great results with this site to reduce PDF size.