Any easy way to OCR images?

Mostly used only Times New Roman font and a small rectangle (220x86). The text is homogeneous, the same size, no strange characters, AZaz09, eight different colors (one per line).

Is there an easy way to do this?

+4
source share
1 answer

The best Tesseract open source OCR. He is training, so you can set the time for a new novel, although he already comes with training for a similar text. You can use it with python using os.popen()

+1
source

Source: https://habr.com/ru/post/1333368/


All Articles