lifloveyou / tr

Free Offline OCR 离线的文本识别SDK

tr - Offline OCR

Tested under Python3 with Ubuntu 16.04 and Ubuntu 18.04

一款针对扫描文档的离线文本识别SDK，核心代码全部采用C++开发，并提供Python接口

Requirements

pip install numpy Pillow opencv-python

Install

• 安装方法一

git clone https://github.com/myhub/tr.git
cd ./tr
sudo python setup.py install

• 安装方法二

pip install git+https://github.com/myhub/tr.git@master

• 测试是否安装成功

python test.py
python test_angle.py

How To Use?

import tr

# detect text lines, return list of (x, y, width, height)
print(tr.detect("imgs/web.png"))

# detect text lines with angle, return list of (cx, cy, width, height, angle)
print(tr.detect("imgs/id_card.jpeg", tr.FLAG_ROTATED_RECT))

# recognize text line, return (text, confidence)
print(tr.recognize("imgs/line.png"))

# detect and recognize text lines, return list of ((x, y, width, height), text, confidence)
print(tr.run("imgs/name_card.jpg"))

# detect and recognize text lines with angle, return list of ((cx, cy, width, height, angle), text, confidence)
print(tr.run_angle("imgs/id_card.jpeg"))

TODO

• 检测带角度的文本框
• 识别部分支持带角度的文本框
• 优化识别部分代码，当前文本识别准确率不够高
• 支持Windows/Mac OS等操作系统
• 支持表格检测
• 识别英文空格
• 支持GPU/手机端

效果展示