>I don't edit image in external editor. Why not use a tool such as OpenCV? You could use an operation such as erode until you have 3 contiguous blocks of black to identify paragraphs. You could then present each paragraph individually to Tesseract in 2 orientations.
On Tuesday, September 10, 2019 at 4:23:46 AM UTC-4, Jakub Fridrich wrote: > > I have next image, with multiple orientation on text, 0 and 180 degree, > without red notes. > > I need convert to text, but text on the middle block return trash of data. > I don't edit image in external editor. > How to get data for all paragraph correctly? > > I'm testing this command: > tesseract --oem 1 --psm 1 -l eng "img.jpg" "imgdata" > > > [image: DRjoxOqVAAEpDWU.jpg] > 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/dca95d4c-ec86-4e1d-bc24-5e3a761aaa2a%40googlegroups.com.
