中文版说明请见中文README。

cnocr

A python package for Chinese OCR with available trained models. So it can be used directly after installed.

The accuracy of the current crnn model is about 98.7%.

The project originates from our own (爱因互动 Ein+) internal needs. Thanks for the internal supports.

Changes

Most of the codes are adapted from crnn-mxnet-chinese-text-recognition. Much thanks to the author.

Some changes are:

use raw MXNet CTC Loss instead of WarpCTC Loss. No more complicated installation.
public pre-trained model for anyone. No more a-few-days training.
add online predict function and script. Easy to use.

Installation

pip install cnocr

Please use Python3 (3.4, 3.5, 3.6 should work). Python2 is not tested.

Usage

Predict

from cnocr import CnOcr
ocr = CnOcr()
res = ocr.ocr('examples/multi-line_cn1.png')
print("Predicted Chars:", res)

When you run the previous codes, the model files will be downloaded automatically from Dropbox to ~/.cnocr. The zip file will be extracted and you can find the resulting model files in ~/.cnocr/models by default. In case the automatic download can't perform well, you can download the zip file manually from Baidu NetDisk with extraction code pg26, and put the zip file to ~/.cnocr. The code will do else.

Try the predict command for examples/multi-line_cn1.png:

examples/multi-line_cn1.png

python scripts/cnocr_predict.py --file examples/multi-line_cn1.png

You will get:

Predicted Chars: [['网', '络', '支', '付', '并', '无', '本', '质', '的', '区', '别', '，', '因', '为'], ['每', '一', '个', '手', '机', '号', '码', '和', '邮', '件', '地', '址', '背', '后'], ['都', '会', '对', '应', '着', '一', '个', '账', '户', '一', '一', '这', '个', '账'], ['户', '可', '以', '是', '信', '用', '卡', '账', '户', '、', '借', '记', '卡', '账'], ['户', '，', '也', '包', '括', '邮', '局', '汇', '款', '、', '手', '机', '代'], ['收', '、', '电', '话', '代', '收', '、', '预', '付', '费', '卡', '和', '点', '卡'], ['等', '多', '种', '形', '式', '。']]

Predict for Single-line-characters Image

If you know your image includes only one single line characters, you can use function Cnocr.ocr_for_single_line() instead of Cnocr.ocr(). Cnocr.ocr_for_single_line() should be more efficient.

from cnocr import CnOcr
ocr = CnOcr()
res = ocr.ocr_for_single_line('examples/rand_cn1.png')
print("Predicted Chars:", res)

With file examples/multi-line_cn1.png：

examples/rand_cn1.png

you will get:

Predicted Chars: ['笠', '淡', '嘿', '骅', '谧', '鼎', '皋', '姚', '歼', '蠢', '驼', '耳', '胬', '挝', '涯', '狗', '蒽', '子', '犷']

(No NECESSARY) Train

You can use the package without any train. But if you really really want to train your own models, follow this:

python scripts/cnocr_train.py --cpu 2 --num_proc 4 --loss ctc --dataset cn_ocr

Future Work

support multi-line-characters recognition
Support space recognition
Bugfixes
Add Tests
Maybe use no symbol to rewrite the model
Try other models such as DenseNet, ResNet

Metadata

Release files for cnocr 0.2.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for cnocr 0.2.0
File	Size	Uploaded
cnocr-0.2.0.tar.gz	19.4 kB	Apr 7, 2019	Details

Built distribution (wheel)

Table of built distributions (wheels) for cnocr 0.2.0
File	Interpreter	ABI	Platform	Reset
cnocr-0.2.0-py3-none-any.whl 35.6 kB Apr 7, 2019	Python 3	none	any	Details

Total release size: 55.0 kB

Release files / cnocr-0.2.0.tar.gz

Download URL	cnocr-0.2.0.tar.gz
Size	19.4 kB
Tags	Source
SHA-256 checksum How to use checksums	`7f736b33a29cf7ccfd72e1c27e49fc68bfd9e2129480c08218f4f815a0b20f14`
BLAKE2b-256 checksum How to use checksums	`1ad72156d29de187f00ec27551c8d6bd798f2b1c4e001f82a21e0c8fc2b1c489`
Upload date	Apr 7, 2019
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/1.13.0 pkginfo/1.5.0.1 requests/2.9.2 setuptools/40.6.3 requests-toolbelt/0.9.1 tqdm/4.23.3 CPython/3.6.5`

Release files / cnocr-0.2.0-py3-none-any.whl

Download URL	cnocr-0.2.0-py3-none-any.whl
Size	35.6 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`dbc50f9c3bf5c594a666bd09ebd99f01d3d2d2dbc48011bc611c53fa13d205e4`
BLAKE2b-256 checksum How to use checksums	`49b7a9d383d87e892721683042bcb18128eae7435bc5e379f7f13c3d1591c64e`
Upload date	Apr 7, 2019
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/1.13.0 pkginfo/1.5.0.1 requests/2.9.2 setuptools/40.6.3 requests-toolbelt/0.9.1 tqdm/4.23.3 CPython/3.6.5`