https://github.com/amirabbasasadi/persianocr

Simple word-level OCR program for the Persian language based on Recurrent Neural Networks using Pytorch and OpenCV
https://github.com/amirabbasasadi/persianocr

image-processing machine-learning ocr ocr-recognition opencv python pytorch

Last synced: about 1 year ago
JSON representation

Simple word-level OCR program for the Persian language based on Recurrent Neural Networks using Pytorch and OpenCV

• Host: GitHub
• URL: https://github.com/amirabbasasadi/persianocr
• Owner: amirabbasasadi
• Created: 2020-07-25T15:35:14.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2020-07-26T17:08:22.000Z (almost 6 years ago)
• Last Synced: 2023-03-08T10:03:28.089Z (over 3 years ago)
• Topics: image-processing, machine-learning, ocr, ocr-recognition, opencv, python, pytorch
• Language: Jupyter Notebook
• Homepage:
• Size: 723 KB
• Stars: 15
• Watchers: 2
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Simple Persian Word-Level OCR using RNN in Pytorch and OpenCV

This notebook uses [Shotor dataset](https://github.com/amirabbasasadi/Shotor), a synthetic dataset for word-level OCR.   

![OCR Result](https://raw.githubusercontent.com/amirabbasasadi/PersianOCR/master/OCR_result.png)

## References

This paper helped me a lot, however my architecture is not same

- https://arxiv.org/abs/1805.09441

- [Pytorch Tutorial on RNNs](https://pytorch.org/tutorials/intermediate/char_rnn_classification_tutorial.html)  

For word segmentation using dilation see this:

- https://stackoverflow.com/a/10970473/4334320

The text of the image which I used to show the final result is a translation of this book:

- The Theory That Would Not Die, Sharon McGrayne