This repository aims to provide a collection of mini-projects for a beginner learning Optical Character Recognition (OCR) using Deep Learning. Idea is to start with very simple concept for OCR and then progress toward complex ones.
Currently following concepts have been added to repository:
- Approach1: Training a custom CNN end-to-end for random string recognition
- Approach2: Reading paragraph from a screenshot. Using OpenCV to segment out the characters and a custom CNN to classify charaters.
I hope in future I get to do more mini-projects. If you would like to contribute then contact me at [email protected]