What is needed?
It's needed to detect the orientation and rotation of the text in the image (selected region) and normalize the image before performing OCR.
Also, it would be great to clean up the image before OCR too.
Related materials
https://pyimagesearch.com/2022/01/31/correcting-text-orientation-with-tesseract-and-python/