Nick, Nope I think that's what I'm going to do. Thanks for you help.
Cheers Shaun Farrell On Tue, Mar 25, 2014 at 12:22 PM, Nick White <[email protected]>wrote: > Hi Shaun, > > Why not just use an arbitrary font name? Like mylabelfont33, or > whatever? Tesseract doesn't do anything interesting with the font > name, it's just a label. > > Or am I missing your question? > > Nick > > On Mon, Mar 24, 2014 at 06:14:25AM -0700, Shaun Farrell wrote: > > I'm working on a prototype to be able to OCR Beverage labels and pull the > > description off them. The problem that I have is that the fonts can be > all > > different and I may or may not know the font. I want to be able to > script this > > as much as possible. Is there a way to train Tesseract in a way that > you don't > > need to know the name of the font? Can I supply an image to train it > myself > > without the font name? I have attached a couple of examples. One Idea > that have > > it to automatically crop out the description text so that the OCR > doesn't have > > to figure out where the text is. > > > > > > [Revolver][CigarCity] > > > > > > The first image (Revolver Brewing) does a pretty good job when I crop > out the > > right had side description: > > > > A full-flavored bock finished with > > Northern Brewer and Saphir hops. > > Brewed with an abundance of > > Munich and caramel malts for a > > hearty biscuit and toffee choracter. > > > > The second image (Cigar City) not so much. I cropped out the middle > > description and this is what I got: > > > > WMNF 88.5Fm IS 3 > > I1s'rener-supporreo > > communrru l'aDi0 s1'a11on > > TH3'l' cetesrares Cl.IlT|.Il'al > > DiVel’SiT9 am: is commmeb > > T0 GQUHIH9. Peace ano > > GCOn0miC JUSTICE. WMNF in > > Tampa Has Been Sel'VinG > > THE communrru since 1979, > > ano is Cel9Bl‘aTil1G THE > > 33]‘ D H|1|1|'Vel'Sal‘9 OF THe > > WMNF Tl‘0PiCal Hearwave. > > > > T0 Learn more asour WMNF, > > GO TO lUl‘I1I1F.0l' G. > > > > T|"0PiCal Heatwave WH9aT > > ate IS an American WHGHT > > Ale. Generousw HOPPGD > > UJi'I' H Kouaru HOPS Fl'0I'n > > New zealano. THE KOHHTU > > HOPS Pl‘0ViDe 3 very > > Tl‘0PiCal FLaV0f mar F1’ > > perrecns WIT H THi$ > > summer ate. > > > > I know this is because its not sure of the font. > > > > Most common fonts work pretty well... But does anyone have any > suggestions on > > how one might go about this? > > > > Cheers! > > > > -- > > -- > > You received this message because you are subscribed to the Google > > Groups "tesseract-ocr" 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/tesseract-ocr?hl=en > > > > --- > > You received this message because you are subscribed to the Google Groups > > "tesseract-ocr" group. > > To unsubscribe from this group and stop receiving emails from it, send > an email > > to [email protected]. > > For more options, visit https://groups.google.com/d/optout. > > -- > -- > You received this message because you are subscribed to the Google > Groups "tesseract-ocr" 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/tesseract-ocr?hl=en > > --- > You received this message because you are subscribed to the Google Groups > "tesseract-ocr" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/d/optout. > -- -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" 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/tesseract-ocr?hl=en --- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
