Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Chinese #8

Open
wants to merge 15 commits into
base: master
Choose a base branch
from
Open

Chinese #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
be5d847
Update config.py
wushilian Jan 26, 2018
cdf2087
Update model.py
wushilian Jan 26, 2018
4be9e27
Update train.py
wushilian Jan 26, 2018
c6e142f
Add files via upload
wushilian Jan 26, 2018
68d9454
Add files via upload
wushilian Jan 26, 2018
2cdc68e
Update README.md
wushilian Jan 26, 2018
23e73f2
Update train.py
wushilian Jan 26, 2018
552201d
Update README.md
wushilian Jan 26, 2018
e8da2ea
Update train.py
wushilian May 29, 2018
a15ae4d
Update train.py
wushilian Jun 3, 2018
0bf3a64
Update config.py
wushilian Jun 19, 2018
245a091
Update train.py
wushilian Aug 12, 2018
5e95bee
Update infer.py
wushilian Aug 12, 2018
667101a
Update model.py
wushilian Mar 26, 2019
c01df08
Update model.py
wushilian Jun 6, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 13 additions & 10 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,11 +1,14 @@
# CRNN_Attention_OCR
CRNN with attention to do OCR，this is just a toy code to show how use seq2seq with attention in OCR
## network model
CRNN is base CNN,and BiLSTM with 256 hidden_units is encode network ,GRU with 256 hidden_units is decode network

## how to use
put your image in 'train' dir,and image name should be like "xx_label_xx.jpg",Parameters are set in config.py,<br>and then just run the train.py

## Dependency Library
# 用CRNN+seq2seq+attention识别中文
中文类别为5990类，类别可以在char_std_5990找到，中文样本的合成参考的是caffe_ocr项目，
## 网络结构
CNN用的是CRNN中的结构，一层双向lstm做编码器，一层GRU做解码器
## 如何训练
训练需要2个txt文件（train.txt，test.txt）保存图片的名字以及label,<br>
可以在这里下载样本[(caffe_ocr)百度网盘](https://pan.baidu.com/s/1dFda6R3#list/path=%2F)<br>
在config.py里修改路径，运行train.py
## 依赖
TensorFlow >=1.2<br>
opencv
opencv<br>
## 引用
[caffe_ocr](https://github.com/senlinuc/caffe_ocr)<br>
[attention-ocr-toy-example](https://github.com/ray075hl/attention-ocr-toy-example)
Loading