[sword-devel] tesseract and Münster
Troy A. Griffitts
scribe at crosswire.org
Sat Nov 29 04:20:59 MST 2008
Had some very exciting meetings with people at ETS/SBL last week and
hope to share some information and opportunities soon.
One specific area of excitement for me personally was a meeting with
Ulrich Schmidt about the Virtual Papyri Room at Münster. I showed them
some of our tools on our labs site, boasted about all of the talented
engineers we have seeking worthwhile work, and they sounded excited to
collaborate together.
A very small project they initially suggested was a reverse engineering
of the lost data used to print (on a dot-matrix printer, mind you!) Kurt
Aland's work in Text und Textwert. I have a pdf containing scans of a
small chunk of the printout and have started a tesseract training file
for the greek font.
I would be excited to turn this project over to someone else who is keen
on this work and has the time to do an excellent job. I would like to
provide their team with the tesseract training files we produce and any
scripts we use for the work, so they can reproduce and tune the process.
Here is a sample image of 2 pages, verso, recto.
http://crosswire.org/~scribe/tundt-086.jpg
More information about the sword-devel
mailing list