'pytesseract-ocr' sometimes shows WRONG Numbers , which were extracted from an Image (URGENT)

[VV Kumar]

Hi 

I am working on an Python Project where, I supposed to extract some numbers from an Image(i.e. Invoice) , which is input to my pytesseract module and I am using image_to_string() function. 

pytesseract sometimes shows wrong Numbers as output , which were extracted from an Image. 

 

(I am not preprocessing Image, since image is in grayscale type and is in good condition)

My pytesseract code:

custom_config6 = r'--oem 3 --psm 6 outputbase digits'

print(pytesseract.image_to_string(crop_img, config=custom_config6))

Can anyone help me, how to approach further, as I'm in middle of an very Urgent assignment.

Thanks in advance

Warm Regards

VVKumar