A lot more work has to be done on preprocessing that image. Consider the qualities of printed text that Tesseract is designed to recognize. My advice is to always try and reduce the image to solid black text on a white background before attempting to pass it to Tesseract.
On Sun, Aug 25, 2019 at 11:41 AM Jeppe Søndergaard < [email protected]> wrote: > Hi everyone > > I have quite a challenge. I am trying to get a good reading of an > "inscription" on a tire. But so far no luck. I have preprocessed the image > and used tesseract, but no luck. I have attached the processed image I have > used for tesseract. Please let me know if you have any suggestions how to > do this, or let me know if this is fundamentally impossible with todays > technology. > > Thanks! > > > -- > 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]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/tesseract-ocr/96aa7bd7-aa9a-437f-92e3-990134894f43%40googlegroups.com > <https://groups.google.com/d/msgid/tesseract-ocr/96aa7bd7-aa9a-437f-92e3-990134894f43%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- 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]. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/CABtjQ9LatKiHHR7yHOCmXxbaNQHtgYhS-mdrQJxs0EWLhtMuNA%40mail.gmail.com.
