Problem with OCR of image with persian text?
by shams from LinuxQuestions.org on (#51ZTQ)
I am running manjaro linux, i want to OCR some images with persian text, i search the google and read a lot since few days, i installed the tesseract and gImageReader, added the fas.traineddata best and the ara.traineddata best to the tessdata and the tesseract got it, then installed the hunspell and add the fa.aff and fa.dic to it and it is working, but with all these the tesseract and the gImageReader cannot put the persian characters in the good readable word order they put the characters in miss order, any suggestion please?
