Is it possible to have an API for recognize() which would just perform recognition and return the output text with confidence? The pytesseract.image_to_string() call only gives the recognized text.

This repository contains OCR project convert from image to text using Pytesseract and flask. And deploy to google cloud ...