Il giorno giovedì 17 ottobre 2013 19:58:35 UTC+2, Nick White ha scritto: > > On Thu, Oct 17, 2013 at 07:20:13PM +0200, zdenko podobny wrote: > > there is no 100% accuracy - in any OCR. > > Zdenko's correct, I'm afraid, even for self-generated images where > you can produce "perfect" specimens. Depending on what you're doing, > it may make more sense to use something like a QR code to do your > stuff, which has error correction and whatnot embedded. > > Otherwise, if you know the sorts of data you'll be encountering, you > can add replace / pattern rules, but if it's base64-encoded text > then that isn't really an option. >
damn, there goes my hacking plan. While I could substitute base64 with something else it would still not be possible to apply any lingusitic pattern or error correction. While what tesseract can do is pretty impressive, I am still shocked to learn that it is not possible to achieve 100% reliability having full control of the fonts and all the rest. thank you for your feedback - much appreciated. Marco -- -- 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/groups/opt_out.
