The "llgen" tool will do that for you. However, the classifiers generated from TrueType fonts are generally not that good because they lack the statistical variation found in real documents. "llgen" can generate additional, artificial character degradations, which may help.
However, we haven't been optimizing recognition for the case where little training data is available. That kind of functionality obviously would be useful at some point. Tom On Wednesday, May 30, 2012 5:15:29 PM UTC+2, Lucas Fermino wrote: > > Thanks! > > I'm trying generate a new model using truetype fonts. The first results > weren't so good. I believe that I used few samples in the training. > > Em quinta-feira, 26 de abril de 2012 17h40min55s UTC-3, Lucas Fermino > escreveu: >> >> I'm trying to create a new characters model, but I don't know which >> are the steps of this process. Someone could explain me this process >> in details? > > -- You received this message because you are subscribed to the Google Groups "ocropus" group. To view this discussion on the web visit https://groups.google.com/d/msg/ocropus/-/wdjEPREl_P4J. 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.
