Has anyone here used opencv to identify blocks of text in photographs? (it is not about OCR)

For example, in an app called Brainly, the app prompts users to take a picture of an exercise and then crop the question area. After this step, they use OCR to extract the text from the image. (screenshots below)

What Iā€™m questioning is whether anyone else here has used OpenCV to do the same job that the Brainly app asked users to do: identify and cut blocks of text.

If so, could you share what your project was like? How accurate did it work?

Thanks :slight_smile: