background
When I downloaded a scholar paper from a library database, rsc, the pdf file of paper will be added a text watermark to mark your right. However, it’s so ugly and prevents sharing. So I decided to remove it.

Solution
commercial
Solution |
Pros |
Cons |
Adobe pdf pro + pitstop |
search and replace, batch modification |
too heavy |
Foxit PhantomPDF |
search and replace |
manual modification |
LibreOffice Draw |
search and replace |
poor performance |
PDF Studio |
search and replace |
poor performance |
Open Source
pdftk, qpdf + vim, sed, awk
Operation
Remove
Most of PDF file will be compressed to reduce the file size. So we need to uncompress it, and go to the next step.
1
| pdftk xxx.pdf output uncompressed.pdf uncompress
|
Open the uncompressed file, press /watermark_string
to find the location of the object.
shift
+ v
to select multiple line and delete
them.

Finally, compress the modified file and save.
1
| pdftk uncompressed.pdf output unwatermarked.pdf compress
|

Add
1
| pdftk original.pdf multistamp watermark.pdf output final.pdf
|