Why is the text not recognized from this image?

Question

The text isn't properly recognized from the following image, although it's working for other images:

from PIL import Image
from pytesseract import pytesseract

pytesseract.tesseract_cmd = (r'C:\Users\SPT\AppData\Local\Programs\Tesseract-OCR\tesseract.exe')

img = Image.open(r'C:\Users\SPT\Desktop\Bot\bot\savedimage.png')

x = pytesseract.image_to_string(img)
print(x)

Answer 1

Keep Improving the quality of the output nearby, when using tesseract.

Your input is some three-channel RGB image. First try, even before binarization, would be to convert to single-channel grayscale:

from PIL import Image
from pytesseract import pytesseract

img = Image.open('b8JEG.png').convert('L')

x = pytesseract.image_to_string(img)
print(repr(x))
# '40\n\n'

As you can see, that already helped.

Alternatively, have a look at the page segmentation mode section. Mode 6 seems reasonable here:

from PIL import Image
from pytesseract import pytesseract

img = Image.open('b8JEG.png')

x = pytesseract.image_to_string(img, config='-psm 6')
print(repr(x))
# '40\n\n'

That also does the job.

----------------------------------------
System information
----------------------------------------
Platform:      Windows-10-10.0.16299-SP0
Python:        3.9.1
PyCharm:       2021.1.1
Pillow:        8.2.0
pytesseract:   4.00.00alpha
----------------------------------------