I might be completely wrong, but, you might be able to upscale the image if you first skeletonize each glyph and reconstruct the a glyph from the vectorized skeleton.
I'm only putting this out as a possibility, I have no idea how to do this and i imagine it would be very slow. On Mar 30, 2011, at 0:09, Dmitri Silaev <[email protected]> wrote: > I said all this assuming you're trying to improve the accuracy. If so, > it's even enough to just upscale your original image by like 3 or 4 > times. E.g. check the results for the file I've attached. But a better > thing to do is get your source images already in a higher res. > > Warm regards, > Dmitri Silaev > > > > > > On Tue, Mar 29, 2011 at 8:06 PM, Robert P. J. Day <[email protected]> > wrote: >> On Tue, 29 Mar 2011, Dmitri Silaev wrote: >> >>> Imho you're going in the wrong direction. Thresholding won't help. >>> I've thresholded the image manually with Photoshop and Tesseract's >>> result was the same as I for the original image. >>> >>> You need to get the image in a higher res, yours is 72 DPI only, >>> that's not enough for Tesseract. In plain English it means you need >>> more pixels per letter. >> >> ok, that's reasonable, i'll just be more restrictive about which >> tiff files i accept for further processing. thanks. >> >> rday >> >> -- >> >> ======================================================================== >> Robert P. J. Day Waterloo, Ontario, CANADA >> http://crashcourse.ca >> >> Twitter: http://twitter.com/rpjday >> LinkedIn: http://ca.linkedin.com/in/rpjday >> ======================================================================== >> > > -- > 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. > > <phototest3.tif> -- 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.
