
On Fri, 10 Mar 2006 09:31:18 -0500, "Joshua Hutchinson" <joshua@hutchinson.net> wrote: | |> ----- Original Message ----- |> From: "Keith J. Schultz" <schultzk@uni-trier.de> | |> Interesting is also, I had a Apple Newton and it recognized |> my handwriting with 98-99% accuracy. Yet, most OCR systems |> today will fail!! They can not be trained! I still have to find |> a system today with similar performance. So much for technological |> break throughs. |> | |Not to disagree with your main points, which I agree with, but I thought I'd point out that most major OCR packages still allowing training (I'm most familiar with FineReader), but they do tend to bury deep so that you have to hunt to find out how to do it (we have people who do it regular at DP for some of the more ... creative ... fonts we find sometimes in old texts). I use Readiris because finereader would not mount when I tried it *long* ago. The problem there is that it does not ?see? the thin strokes in "w" and "r" so no amount of training will work for those two characters. -- Dave Fawthrop <dave hyphenologist co uk> Freedom of Speech, Expression, Religion, and Democracy are the keys to Civilization, together with legal acceptance of Fundamental Human rights.