Pytesseract doesnt recognize simple text in image

Question

I want to recognize a image like this:

I am using the following config:

config="--psm 6 --oem 3 -c tessedit_char_whitelist=0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ,."

but when I try to convert that, I get the following:

1581

1

W

I think that the image shows really clearly what is written and think that there is a problem with pytesseract. Can you help?

it also happened for me, one time it showed me a star as a dot, while there are so many stars in the same image, don't know whats wrong... but i didn't use any configs — Ghost Ops, Sep 20 '21 at 14:47
@GhostOps Yes but with such a clean image this must be possible right? — Moritz Pfennig, Sep 20 '21 at 14:48

score 1 · Accepted Answer · answered Sep 21 '21 at 00:15

Preprocessing the image to obtain a binary image before performing OCR seems to work. You could also try to resize the image so that more details would be seen

Results

158.1
1
IT

import cv2
import pytesseract

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

# Grayscale and Otsu's threshold
image = cv2.imread('1.png')
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]

# Perform text extraction
data = pytesseract.image_to_string(thresh, lang='eng', config='--psm 6')
print(data)

cv2.imshow('thresh', thresh)
cv2.waitKey()

I resized it and now it works flawless – Moritz Pfennig Sep 22 '21 at 07:26 — Moritz Pfennig, Sep 22 '21 at 07:26

Pytesseract doesnt recognize simple text in image

1 Answers1