Skip to content
This repository has been archived by the owner on Dec 18, 2019. It is now read-only.

OCR orientation detection

Jerome Flesch edited this page Sep 23, 2016 · 6 revisions

Paperwork uses the orientation detection feature of Tesseract. It is quite fast and accurate.

If Tesseract is not installed, Paperwork will fallback on Cuneiform. Cuneiform doesn't provide orientation detection. So Paperwork will run Cuneiform on all 4 orientations, and score each of them. The score is established using a dictionary (Enchant/Myspell/Aspell).

Debian

# Spell checking (myspell)
$ sudo apt-get install myspell myspell-<lang>

Fedora

# Spell checking (aspell)
$ sudo yum install aspell aspell-<lang>

Ubuntu

# Spell checking (myspell)
$ sudo apt-get install myspell myspell-<lang>
Clone this wiki locally