blob: 165059af34f61389fbaa4134e62898caa3571553 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
|
#!/bin/sh
#
# Makes a PDF with embedded text extracted by tesseract
#
# Requires imagemagick, pdftk, hocr2pdf and tesseract
#
# Also requires this tesseract configuration:
# echo 'tessedit_create_hocr 1' > /usr/local/share/tessdata/configs/hocr
for i in `ls *png`
do
a=`basename $i .png`
echo processing $a
tesseract $i $a hocr 2>/dev/null
# hocr2pdf has a habit of segfaulting, so fall back to convert
hocr2pdf -i $i -o $a.pdf < $a.html || convert $i $a.pdf
rm -f $a.html
done
pdftk *pdf cat output book.pdf
|