[FIXED] How can I properly detect LetsGoDigital font text?

Issue

I am on Windows 10, and I try to extract digits from this image

with the pytesseract library with language lets (cf. https://github.com/adrianlazaro8/Tesseract_sevenSegmentsLetsGoDigital or LetsGoDigital, cf. https://github.com/arturaugusto/display_ocr).

I preprocessed my image (grey, threshold and erosion) to get:

But the output of

pytesseract.image_to_string(img, lang='lets')

is empty.

Solution

You didn't set any specific page segmentation method. I'd opt for --psm 6 here:

Assume a single uniform block of text.

So, even without further pre-processing I get the proper result:

import cv2
import pytesseract

img = cv2.imread('RcVbM.jpg')

text = pytesseract.image_to_string(img, lang='lets', config='--psm 6')
print(text.replace('\n', '').replace('\f', ''))
# 004200

----------------------------------------
System information
----------------------------------------
Platform:      Windows-10-10.0.19041-SP0
Python:        3.9.1
PyCharm:       2021.1.1
OpenCV:        4.5.2
pytesseract:   5.0.0-alpha.20201127
----------------------------------------

Answered By - HansHirse