Eliminating color may help. The current adaptive thresholding is not as good as it could be, but as long as you avoid low contrast combinations like red/black or green/white it should be ok. Ray.
On Fri, Feb 13, 2009 at 8:24 AM, Jordan L Dressman <[email protected]>wrote: > Thanks that is good info. > > Do you think that eliminating color would help? Such as white background > and black text. Right now it can be different backgrounds and text color. > > > ----- Original Message ----- > From: Ray Smith <[email protected]> > Date: Friday, February 13, 2009 1:36 > Subject: Re: good fonts for ocr > To: [email protected] > > > Arial isn't the best choice. If you want to optimize accuracy > > and speed, > > courier is about the best choice, but if you don't want a fixed > > pitch font, > > choose times new roman, as it will suffer less on I/l than sans- > > serif fonts. > > Courier will be useful for speed, as it is fixed pitch. If you > > go with times > > though, it would be good to increase the inter-character spacing > > just a bit > > to stop adjacent characters touching.Ray. > > > > On Thu, Feb 12, 2009 at 12:55 PM, Jordan L Dressman > > <[email protected]>wrote: > > > I'm using arial 16pt font and it is mistaking an 8 for a > > B. Is there > > > better fonts I can use that work well with ocr. I don't > > have ocr-a and > > > ocr-b font options. From the ones I have tried I have to > > increase the fonts > > > quite a bit to get more accurate. It would be great if > > there is a font that > > > works well at 10 or 12 point font. > > > > > > Also anything else that would increase the accuracy and > > speed? Would it be > > > better to only use black on white and eleminate all color? > > > > > > > > > > > > > > > > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
