The only external dependency is pdftk
The script cleanpdf.py
takes as inputs the source pdf, the destination pdf, and a list of text strings identifying watermarks to remove from the pdf
Classical watermarks used by many publishers indicate the name of the institution, the IP address, the date and time of download, and standard patterns text such as 'For personal use only'...
Example:
./cleanpdf.py input.pdf output.pdf 'My university' 'Downloaded from' 'Access provided' 'For personal use only' '129.000.000.001'