If you're getting too many errors on Courier, maybe your resolution is too low or too high? Can you post a picture (best... add it to the Flickr pool)?
Nevertheless, back to your original question, once you have the .cmodel file, you just use it as an argument to ocropus-lattices, as in the sample script: ocropus-lattices -m your.cmodel textline.png ocropus-ngraphs textline.png Note that if you want to be able to recognized mixed font text, you need to train on a variety of fonts in addition to the one you want to improve on. Tom On Thursday, October 4, 2012 6:10:43 PM UTC+2, JR wrote: > > Help in training ocropus. > > I'm trying to improve my OCRing of some text that is not coming out > well... > > I've followed the guide of uw3-500 to train ocropus on my font set > (english courier) as it was making way too many errors. Meaning, I > collected a few pages of my text, created ??????.bin.png one liners, with > .gt.txt text files.. > > It ran ok, and created me .cmodel .h5 and .tsplit files... What do I do > with theses? Do I splitly copy them into /usr/local/share/ocropus ??? > > Sorry for my ingorance... > > -- 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]. To view this discussion on the web visit https://groups.google.com/d/msg/ocropus/-/PLWeZF7km_AJ. For more options, visit https://groups.google.com/groups/opt_out.
