Sometimes spaces between words are ignored when tesseract is used to recognize Sinhala text.
- The traineddata from tesseract does not have a spacing problem, even though there ware changes in tesseract since it was uploaded. - The spacing problem occurs regardless of whether I start the training from scratch or bootstrap with the traineddata from tesseract. - The spacing problem gets worse with more training. - Adding more space between the words during training does not make a difference. - Adding double space between the words during recognition solves the problem. - The spacing problem is not consistent, i.e. in the recognition of a text only some of the inter-word spaces are ignored (could not figure out any logic as to when it happens). I have attached a screenshot, comparing a sample of input and output text. Words missing spaces are underlined. <https://lh3.googleusercontent.com/-T6hAiA4VclA/Wv1HEKkrioI/AAAAAAAAIN4/hZors3-ZJq01n24E3_c_JFzhws90X-x9gCLcBGAs/s1600/Screenshot_20180517_143558.png> -- 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 post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/dfba845a-abe4-48fa-b834-7c64faf54f13%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
