Last active
May 14, 2021 09:39
-
-
Save trufanov-nok/50e8019b363b7cbfb03dbf6054b984b6 to your computer and use it in GitHub Desktop.
Add searchable and selectable text layer in multipage djvu with help of Teseract
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/perl | |
# based on | |
# https://en.wikisource.org/wiki/Help:DjVu_files/OCR_with_Tesseract | |
# requirements: | |
# sudo apt install python2 | |
# sudo rm /usr/bin/python | |
# sudo ln /usr/bin/python2 /usr/bin/python | |
# sudo pip2 install lxml python-djvulibre | |
$lang = "eng+rus"; | |
if ( $#ARGV < 0 ) { | |
print "give a DJVU file as 1st argument \n" ; | |
exit 0 ; | |
} | |
$inputdjvu = $ARGV[0] ; | |
$imagetmp = "/tmp/temp.tif" ; | |
$djvutmp = "/tmp/outdjvu" ; | |
print "processing of $inputdjvu\n" ; | |
# calculate the number of pages | |
$nbpages = `djvused "$inputdjvu" -e 'n'` ; | |
print "number of pages: $nbpages" ; | |
for ($i=1 ; $i <= $nbpages ; $i++) { # | |
print "OCR de la page $i\n" ; | |
# page extraction as an image | |
`ddjvu -format=tiff -mode=black -page="$i" "$inputdjvu" $imagetmp` ; | |
`tesseract $imagetmp /tmp/outocr$i -l $lang hocr`; # use hOCR format | |
print "OCR done\n" ; | |
`hocr2djvused /tmp/outocr$i.hocr > $djvutmp`; # create djvused script from hOCR | |
`sed -i $djvutmp -e '1 s/.*/select $i/'`; # replace select 1 with right page number | |
# writing the text in the DJVU file | |
`djvused "$inputdjvu" -f $djvutmp -s `; | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment