Can we read only a specific font size texts with tesseract ?

[Utsaww]

Hi, I am curious as many say it won't be possible to identify and read the text based on text size.

I have a condition like - read-only that text who is bigger in all the images or read the text whose font size is greater than 2px ?

is it possible in this library?

[Sicos1977]

https://github.com/Sicos1977/TesseractOCR/blob/master/TesseractOCR/Layout/Words.cs

And then use the public FontAttributes FontAttributes property.

But you still need to OCR all text because as far as I know there is not an option in Tesseract to say just get the fonts with X size.

You can iterate through the page layout like this --> https://github.com/Sicos1977/TesseractOCR#iterate-through-the-layout-of-a-page