RogerTheRabbit: There is already library software that tries to digitize the text in a scanned image, Google and library of congress use it, but its not the best at its job and needs humans to help it solve the text it can't make out or if its too blurry, thats partially what captcha text is, parts of scanned books.
So it would run into the same problems if add machine translation capabilities to it.
- Reply
So it would run into the same problems if add machine translation capabilities to it.