On Tuesday, 12 May 2015 04:31:24 UTC+2, smwikipedia smwikipedia wrote: > > I am trying to use Tesseract to do OCR for some screenshots. >
I guess you will need to pre-proced/improve quality of images (see Wiki or VietOCR has screenshot mode) > But the characters on screen can be of multiple fonts. > This should not be problem. But image examples would help better words ;-) > I see that in the latest 3.03 version, the training tool `text2image` can > easily generate training tif/box pair from training text and font files. > Why do you think you need to do training? Common experience is that image pre-processing is needed more that training. For some font (e.g. 7 segments) /cases it can be better to use other tools... > > If I want to support multiple fonts, do I need to make multiple > *.traineddata files and switch between them at runtime? > no. -- 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 http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/768948db-2a20-4c10-aefa-7604fd5d6e02%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
