search

xingjian-f / DeepLearning-OCR

Build an Optical Character Recognition service using deep learning method
Apache License 2.0
54 stars 35 forks source link

using english #6

Open by-tech opened 6 years ago

by-tech commented 6 years ago

Hi

is it possible to add new language easily ?

Regards

xingjian-f commented 6 years ago

Sure, once you prepared the training dataset(in English, in German, in Japanese and so on), you can train your model to recognize the characters in the image.

by-tech commented 6 years ago

there's no readme. Can you explain please the list of step to your programm ?

Also what accuracy (error) you have reached ?

Regards

xingjian-f commented 6 years ago

I can get 99% accuracy for most types of verification code.

So this is a powerful and dangerous tool, it's shared only for research purpose.

I will write a readme when the next generation of verification code comes out.

For now, you can read the code from the train.py.