sk-spell

podpora slovenčiny v Open Source programoch

How to create simple tesseract wrapper in Python   

back to tesseract-ocr-en

On github there was published Simple Tesseract Python Wrapper – jupyter notebook that give interactive guide how to use tesseract C-API in python.

There is available python module pytesseract, but it is wrapping tesseract executable so:

• it is limited to use only features available via executable
• disk operation are always involved (e.g. for saving output, for reading output)
• speed performance is not optimal in cased o bulk OCR (e.g. sub images): each time tesseract is called – initialization process (and disk operation) is started.

Simple Tesseract Python Wrapper is example how to solve (or at least improved) above mentioned limitation and give us hope that somebody give make advanced tesseract (and leptonica) python wrapper.

back to tesseract-ocr-en

© projekt sk-spell