My answer can add to Sven's. Although close edges can be indeed a problem, and Tesseract feels better when a bigger background area surrounds the target text, probably you'd want to pay attention to the lower edge of the "Sheoldred One.jpg" image. There's a thin dark line along the edge - that's the main difference between your images. This line can confuse Tesseract and be the reason of different recognition results. Passing as clean as possible images to Tesseract would let you achieve better recognition.
HTH Warm regards, Dmitri Silaev www.CustomOCR.com On Tue, Aug 30, 2011 at 7:08 PM, Rick Appleton <[email protected]> wrote: > Hello all, > > I'm fairly new to Tesseract, so please forgive me if this is something > that I can easily fix with a specific setting. > > I have two images which are extremely similar, yet give very different > results. > > http://www.daedalus-development.net/ricka/Sheoldnd%20Whispering%20One.jpg > http://www.daedalus-development.net/ricka/Sheoldred%20One.jpg > > The first image results in: 'Sheoldnd. Whispering One' > The second image results in: 'Sheoldred. One' > > The correct result should be: 'Sheoldred, Whispering One' > > The results in the first image are acceptable, and close enough for me > to work with. However, the results from the second image are > unacceptable to me. I appreciate that it has correctly detected the > words it has found, but the fact that the middle word is missing > entirely gives me lots of problems. > > Is this normal behaviour, or can I tweak Tesseract into giving me some > kind of result for the middle word? > > Kind regards, > Rick > > -- > 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 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
