Tesseract returns nothing for Arabic words/letters

Question

I have installed Pytesseract and it's working perfectly on French/English text and also in numbers. But when I try to read any Arabic text/letter it doesn't return anything.

Here is the code I have used:

try:
    from PIL import Image
except ImportError:
    import Image
import pytesseract

pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files (x86)\Tesseract-OCR\tesseract.exe"

print(pytesseract.image_to_string(Image.open('maroc.jpg'), lang='ara'))

Here is the letter I'm trying to read د:

If someone was able to read it using another method please help, thanks!

Answer 1

Code :

from pytesseract import image_to_string 
from PIL import Image
import pytesseract

print(pytesseract.image_to_pdf_or_hocr('test.png', lang='ara', extension='hocr'))

Take new Arabic tessdata from here:

Answer 2

if you want to recognise arabic words download the arabic trained model from the link below then save it in the location according to your Tesseract folder

C:\Program Files\Tesseract-OCR\tessdata or C:\Program Files (x86)\Tesseract-OCR\tessdata

arabic_tesseract_trained

Answer 3

for raspberry pi 4 just download module from Eliyaz KL answer and put in this path /usr/share/tesseract-ocr/4.00/tessdata/ i don't know which operating system use i answerd in my case