Hi, OCRopus really isn't ready for a comparison; the character shape and language models that are shipping with the software are totally outdated relative to the codebase. We have mainly just been shipping them with the alpha versions for rough testing and as a basis for alignment and training. Any actual performance benchmark involving OCRopus at this point would be meaningless.
The software can use a neural network for character recognition. It can also use shape-based matching, nearest neighbor classification, and HMMs (not released yet). The software does work for handwriting recognition and appears to give state-of-the-art results when used with the right kind of preprocessing (it originally was a handwriting recognition system). Tom -- 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.
