[sword-devel] Neural Networks and Optical Character Recognition
greg.hellings at gmail.com
Fri Apr 25 14:24:04 MST 2008
On Fri, Apr 25, 2008 at 2:43 PM, David Haslam <d.haslam at ukonline.co.uk> wrote:
> I was thinking that any historic text that used something other than a modern
> Latin typeface, such a technique might have better chances of success.
> For example, if someone wished to apply OCR techniques to one of the earlier
> Irish Bibles that was printed using the old Irish font, such a method might
> have considerable advantages. See
> for an example of "William O'Domhnuill's (Daniel's) Translation of the New
> Testament into Irish Completed 1602, and here re-published in 1827".
> As these dates are all so ancient, any resulting digitization must surely
> become public domain. Or is my understanding flawed?
Your understanding is, I believe, slightly flawed. There is a
separate part of the copyright laws which allows for someone who made
a major effort to transform a document (I know translation is covered,
and I would imagine that paper->digital conversion is also covered)
may copyright that transformation. Hence why things like the NIV,
NRSV, NASB, JB, (indeed, even the various modern versions of the Novum
Testamentum Graecum as well as the Tenakh) may be placed under
copyright. We're left to hope that the people who make the document
transformation put them under some sort of copyleft or public domain
situation. The content is in the public domain, but the particular
copies that are made might be placed under copyright laws.
> There may be many similar such works that predate January 1, 1923..
> There have been tremendous advances in image processing techniques in recent
> years. But amateur enthusiasts choosing OCR software may not be fully aware
> of what can be done, and may merely go for the best marketed product aimed
> at modern typefaces.
Very true - it would be interesting to tackle. I might bring it up
with my adviser to see what he has to say about it.
> David H.
> Leandro DUTRA wrote:
> > 2008/4/24, David Haslam <d.haslam at ukonline.co.uk>:
> > First, how proven are the improvements over the current methods?
> > Second, how likely is that you will lead a free software implementation?
> > --
> > skype:leandro.gfc.dutra?chat Yahoo!: ymsgr:sendIM?lgcdutra
> > +55 (11) 3040 7300 r155 gTalk: xmpp:leandrod at jabber.org
> > +55 (11) 9406 7191 ICQ/AIM: aim:GoIM?screenname=61287803
> > +55 (11) 5685 2219 MSN: msnim:chat?contact=leandro at dutra.fastmail.fm
> > _______________________________________________
> > sword-devel mailing list: sword-devel at crosswire.org
> > http://www.crosswire.org/mailman/listinfo/sword-devel
> > Instructions to unsubscribe/change your settings at above page
> View this message in context: http://www.nabble.com/Neural-Networks-and-Optical-Character-Recognition-tp16850434p16904363.html
> Sent from the SWORD Dev mailing list archive at Nabble.com.
> sword-devel mailing list: sword-devel at crosswire.org
> Instructions to unsubscribe/change your settings at above page
More information about the sword-devel