Add Indic numerals and missing punctuation to Arabic

tesseract-ocr / langdata

Source training data for Tesseract for lots of languages

Apache License 2.0

837 stars 888 forks source link

Open mustafa0x opened 6 years ago

mustafa0x commented 6 years ago

Previously: #71 and https://github.com/tesseract-ocr/tessdata_best/issues/11 (also contains a pertinent discussion on how well the different traineddata deal with these characters).

• Indic numerals: (٠ ١ ٢ ٣ ٤ ٥ ٦ ٧ ٨ ٩) • Punctuation: (؛, ،, ﴿﴾) • Also, a ligature very commonly found in Arabic texts: ﷺ

If I can do this myself please simply point me the way.

CC @Shreeshrii

Shreeshrii commented 5 years ago

Please see https://github.com/tesseract-ocr/tesseract/issues/2263#issuecomment-466675793 and test if the traineddata files linked there add all the required characters.

wewark commented 4 years ago

Is this fixed? I've tried the latest version and it didn't detect any Indic numerals.

ShroukMansour commented 4 years ago

@wewark you have to use Arabic.traineddata file. It recognizes arabic, English letters and Arabic-Indic and Arabic numbers

AhmedElsayedTaha commented 3 years ago

@ShroukMansour I use ara.traindata and texts not accuracy also numbers have no accuracy . Is there a solution for this ?