python3: Recognize image text

1. Install Python dependencies: pytesseract and PIL

# 安装pytesseract
pip install pytesseract

# 安装PIL
pip install PIL
#此处如果电脑是window64位系统不能直接使用pip安装，解决方法:
1.先下载：https://www.lfd.uci.edu/~gohlke/pythonlibs/ 根据Python版本下载对应的版本，我的是python3.5所以下载:
 Pillow-5.0.0-cp35-cp35m-win_amd64.whl
2.安装wheel依赖，因为安装Pillow-5.0.0-cp35-cp35m-win_amd64.whl需要wheel依赖
pip install wheel
3.切换到Pillow-5.0.0-cp35-cp35m-win_amd64.whl的下载目录，执行安装名称
pip intall Pillow-5.0.0-cp35-cp35m-win_amd64.whl
到此PIL安装完毕

# 安装识别引擎tesseract-ocr
1.下载安装包tesseract-ocr安装包和中文语言包
下载好后直接点击.exe文件安装即可，默认不支持中文，如果想要支持英文需要把中文包复制到tesseract-ocr/tessdata的目录下

At this point, you can use pictures to recognize text, but only support English

Test image (test2.png):

from PIL import Image
import pytesseract

#测试的图片资源
text = pytesseract.image_to_string(Image.open('test2.png'))
print(text)

result:

If you need to identify Chinese, you need to configure the library path

Open the python installation directory, find and edit D:\python\Lib\site-packages\pytesseract\pytesseract.py

#tesseract_cmd = 'tesseract'
tesseract_cmd = 'C:/Program Files (x86)/Tesseract-OCR/tesseract.exe'

After saving, it can support Chinese recognition

Test image:

result:

It can be seen that the Chinese recognition is still at the level of children!