Optical-Character-Recognition-OCR-for-Telugu
This repository contains code for training and using an OCR system for Telugu.
ocr.py is the file for using pretrained model for ocr.
Training folder contains code for training the models and also the weights and architechture of pretrained models.
The Images folder contain the images for ocr with the output is in .html format (for displying Telugu text)
The Encoding folder contains the utf codes of Telugu.
The pre-trained model required GPU to run.
PS: The above work is an implementation of the papaer, K. Chandra Prakash, Y. M. Srikar, G. Trishal, S. Mandal and S. S. Channappayya, "Optical Character Recognition (OCR) for Telugu: Database, Algorithm and Application"