Learn OCR pre-processing: denoising, deskewing, binarization, and more in Python

Explore the commonly overlooked pre-processing steps that help make Optical Character Recognition (OCR) models work properly in practice.

This repository contains code, a walkthrough notebook (ocr_preprocessing_walkthrough.ipynb), and streamlit demo app for playing around with common ocr pre-processing steps, and seeing their resulting effects on ocr quality.

All processing - from the various pre-processing steps to the ocr itself (here using the popular / classic tesseract model - are performed locally.

Installation instructions

To create a handy tool for your own memes pull the repo and install the requirements file

pip install -r requirements.txt

Starting the streamlit app

Start the streamlit app by pasting the following in your terminal

python -m streamlit run ocr/app.py

Ocr your own images

Note: you can drag and drop any desired image directly into the streamlit app, and play around with how pre-processing steps effect the final ocr output.