Optical character recognition is software that converts images of printed or handwritten text into a sequence of character codes. The images used are documents captured by image scanners or photographs, landscape photographs (e.g., text on a sign in a landscape), and subtitles in images (e.g., in a TV broadcast image). Generally abbreviated as OCR.

Google Cloud Vision’s OCR performance was by far the best.


This page is auto-translated from /nishio/OCR using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.