Use legacy engine instead of LSTM (you will need language model from https://github.com/tesseract-ocr/tessdata):
tesseract processed-res.png - --oem 0 Estimating resolution as 878 115 Dátum: štvrtok 15. júla 2021, čas: 15:18:29 UTC+2, odosielateľ: [email protected] > Hi, > > I'm n00b for tesseract and wondering how I can solve my issue. > > I'm trying to capture digits on gaming screenshots to track damages in a > game and here is an example image of the digits. > > [image: sample.png] > > Then I did `preprocess` the image (remove noises by thresholding) as > follows: > > [image: processed-res.png] > > Then I pass it to tesseract (tesseract 4.1.1-rc2-25-g9707) but it doesn't > recognize as 115 but as `ns` > > $ tesseract -l eng --psm 6 --dpi 300 -c > tessedit_char_whitelist="0123456789" processed-res.png - > > $ tesseract -l eng --psm 6 --dpi 300 processed-res.png - > nS > > Anyone can advise me how to recognize digits? I don't have a font file for > digits so wondering what else I can take.... > > 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/a90a9a92-2159-4406-b370-f7a293eefee5n%40googlegroups.com.
