Ciaconna: Resources for Greek OCR with Ocropus

In this forum, we will work with people who are training OCR engines for Polytonic Greek, helping them improve the quality of their engines by correcting specific texts.
Post Reply
BruceRobertson
Posts: 22
Joined: May 16th, 2014, 9:49 pm

Ciaconna: Resources for Greek OCR with Ocropus

Post by BruceRobertson »

Most of the results at http://heml.mta.ca/lace are from the Gamera-based Rigaudon engine, which requires a parallel computing environment.

However, recently, I've been using that data to train a much simpler engine, Ocropus 0.7. If you have a Unix environment and are comfortable installing software, I'd love for you to try it out by getting the github repository at https://github.com/brobertson/ciaconna. You'll need to install Ocropus first, as these are wrapper scripts and classifiers built on top of it.

Of course, if you're interested in running OCR at home, you should also look at Nick White's fantastic work on Tesseract: http://ancientgreekocr.org/.
NathanSmith
Posts: 62
Joined: June 10th, 2011, 12:38 am
Location: Portland, OR, USA
Contact:

Re: Ciaconna: Resources for Greek OCR with Ocropus

Post by NathanSmith »

BruceRobertson wrote:Most of the results at http://heml.mta.ca/lace are from the Gamera-based Rigaudon engine, which requires a parallel computing environment.
Can you share a bit about the computing environment you use?
BruceRobertson
Posts: 22
Joined: May 16th, 2014, 9:49 pm

Re: Ciaconna: Resources for Greek OCR with Ocropus

Post by BruceRobertson »

The parallelization code assumes a Sun Grid Engine scheduler and 20 - 40 cores with shared memory, since that's what I had last year. The Gamera code requires many standard Python libraries. But as I say, for many uses the Ocropus-based 'Ciaconna' library is as good or better, and it doesn't really need any parallelization or special needs beside Ocropus itself.
NathanSmith
Posts: 62
Joined: June 10th, 2011, 12:38 am
Location: Portland, OR, USA
Contact:

Re: Ciaconna: Resources for Greek OCR with Ocropus

Post by NathanSmith »

Cool, thanks for filling me in. We have some compute clusters at work, so I was curious what you were working with.
Post Reply

Return to “Polytonic Greek OCR”