It's hard to tell without actually seeing the images. In general, OCRopus is designed to handle touching characters and degraded documents; it's really all a question of what accuracy you need, how much training data you have, and how much work you're willing to put in.
This kind of adaptation should be fairly straightforward in the beta release, but OCRopus is not yet in beta. See here what still needs to happen: http://code.google.com/p/ocropus/wiki/OcropusWaves Tom On Jul 4, 8:09 pm, Madan KN <[email protected]> wrote: > Hi All, > > I have a documents printed in old manuscripts. > > 1. The words and characters does not have any pixel gap. > 2. The characters printed in the image, pixels vary from each > character > 3. The characters in the words look like hand written and running hand > writing. (Like French MT Script) > > Good News: > > - I have written a java library for fonts like French MT Script and > works with 100% accuracy. > - Planning to commit to the Open Source ASAP. > > Bad News: > > - I have old printed manuscripts which are scanned with different > angled & rotated. > - The images are not clear and miss lots of pixels in the characters > and words. > > Can OCROpus handle these kind of image conversion????????? > > Please let me know even before i explore OCROpus. > > Madan KN -- You received this message because you are subscribed to the Google Groups "ocropus" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/ocropus?hl=en.
