I was looking around for some documentation on how gocr works. I've been
interested in OCR for many years and was curious if there was an explanation
available. A scan of the readme and source files did not find anything of
FYI - tonight I had a need for a bit of OCR but unfortunately gocr did not
quite work as about half the text is in italics. It was ok on the plain
Times Roman text but parsed the italics pretty badly and was worse in that
the main text was English while the italic text was Japanese using romanji.