summaryrefslogtreecommitdiff
path: root/extras/mkocrpdf-cuneform.sh
diff options
context:
space:
mode:
Diffstat (limited to 'extras/mkocrpdf-cuneform.sh')
-rw-r--r--extras/mkocrpdf-cuneform.sh21
1 files changed, 21 insertions, 0 deletions
diff --git a/extras/mkocrpdf-cuneform.sh b/extras/mkocrpdf-cuneform.sh
new file mode 100644
index 0000000..94e10eb
--- /dev/null
+++ b/extras/mkocrpdf-cuneform.sh
@@ -0,0 +1,21 @@
+#!/bin/sh
+#
+# Makes a pdf with embedded text as extracted by cuneiform
+#
+# Requires imagemagick, pdftk, hocr2pdf and cuneiform
+
+for i in `ls *png`
+do
+ a=`basename $i .png`
+ echo processing $a
+
+ convert $i $a.bmp
+ cuneiform -f hocr -o $a.html $a.bmp
+ rm -f $a.bmp
+
+ # 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