OCR

This project refers to the handwritten recognition with CNN and RNN, decode with CTC.

Dataset

[Model1 - word_model.png] Train on WORD unit of dataset.
[Model2 - line_model.png] Train on LINE unit of dataset.
Split data: https://www.openslr.org/56/ https://drive.google.com/drive/u/0/folders/1xqek060j9_zVWGY80m33qSCrZf3scqUy

Test on IAM dataset:

Model	Test Unit	CER(%)	WER(%)
WORD	WORD	17.20	37.78
WORD	LINE	?????	?????
LINE	LINE	?????	?????

[Google colab]

$ python3 Train.py

$ python3 Prediction.py

$ python3 EvaluationWord.py
$ python3 EvaluationLine.py

Name		Name	Last commit message	Last commit date
Latest commit History 43 Commits
Demo		Demo
Resource		Resource
__pycache__		__pycache__
.gitignore		.gitignore
CRNN_Model.py		CRNN_Model.py
EvaluationLine.py		EvaluationLine.py
EvaluationWord.py		EvaluationWord.py
ImageGenerator.py		ImageGenerator.py
Parameter.py		Parameter.py
Prediction.py		Prediction.py
Preprocessor.py		Preprocessor.py
README.md		README.md
Spell.py		Spell.py
Train.py		Train.py
Utils.py		Utils.py
WordSegmentation.py		WordSegmentation.py