Add Arabic-Indic numerals to Arabic

[Shreeshrii]

Please see https://github.com/tesseract-ocr/tesseract/issues/858

include both 0-9 and ( ٠ ١ ٢ ٣ ٤ ٥ ٦ ٧ ٨ ٩) for Arabic.

[Shreeshrii]

• 0x30 to 0x39
• U+0660 through U+0669

[Testya]

Please add the Arabic comma too, (،) U+060C.

[aboelmor]

Any idea of when the eastern-arabic numerals will be added to the language packs?

[theraysmith]

Added to my local copy for next round of training. Then I will push updated langdata as well.

[Shreeshrii]

@theraysmith

I hope you have seen other comments regarding using only persian number range for persian and arabic range for Arabic.

[theraysmith]

Yes, I hope the experts also see my question about the Arabic languages not mentioned by those issues (kur_ara, pus, uig).

On Mon, Aug 7, 2017 at 6:38 PM, Shreeshrii notifications@github.com wrote:

@theraysmith https://github.com/theraysmith

I hope you have seen other comments regarding using only persian number range for persian and arabic range for Arabic.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/tesseract-ocr/langdata/issues/71#issuecomment-320825657, or mute the thread https://github.com/notifications/unsubscribe-auth/AL056VBYghCZM2R0sOYmSVFtcDC26cPeks5sV7whgaJpZM4NN0lN .

-- Ray.

[AbdelsalamHaa]

Hi guys , I'm using tesseract 4 I'm using ara.traineddata to extract the text from the image. it's working well for the letters but numbers is not good at all . From the comment above there should be some other traineddata for only numbers . any body can guide me where to find it .

thank a lot

[amitdo]

It seems that Ray didn't push the data to our side (langdata_lstm and best/fast repos).

[amitdo]

This issue should be re-opened.