Question

I am trying to perform recognition of a german text with fraktur typeface with ocropus but It doesn t seem to be using deu-f package.

Here are the steps I performed.

Compiled and installed tesseract and ocropus.
Downloaded http://tesseract-ocr.googlecode.com/files/tesseract-2.01.deu-f.tar.gz, unpacked it to tessdata/.

But when I call

 $ ocroscript recognize --tessLanguage=deu-f --output-mode=text image.png

the results are the same as when I call

$ ocroscript recognize --tessLanguage=eng --output-mode=text image.png

Any ideas what the problem is?

Answer 1

The problem is described in http://code.google.com/p/ocropus/issues/detail?id=87. Just need to apply the patch to ocropus and rebuild it.