text-detection-ctpn 图片文字识别 mac环境 cpu版安装

1. 图像文字位置检测  git上下载源码  https://github.com/eragonruan/text-detection-ctpn

   因为默认是gpu版本的，修改为不用gpu有几个坑

   先参考 https://github.com/eragonruan/text-detection-ctpn/issues/43

   把需要注释的几个地方注释掉

   然后修改setup.py, 用下面这段代码整个替换掉

from Cython.Build import cythonize
import numpy as np
from distutils.core import setup
from distutils.extension import Extension

try:
    numpy_include = np.get_include()
except AttributeError:
    numpy_include = np.get_numpy_include()

ext_modules = [
    Extension(
        'bbox',
        sources=['bbox.c'],
        include_dirs = [numpy_include]
    ),
    Extension(
        'cython_nms',
        sources=['cython_nms.c'],
        include_dirs = [numpy_include]
    )
]
setup(
    ext_modules=ext_modules
)

   运行：

 python setup.py build_ext --include-dirs=/anaconda2/envs/chinese-ocr/lib/python2.7/site-packages/numpy/core/include

 include-dirs 换成自己的numpy地址，将build完之后的.so文件copy到utils文件夹下

 运行demo需要下载 https://github.com/eragonruan/text-detection-ctpn/releases checkpoint文件，就是用已经训练好的模型，解压  到text-detection-ctpn的主目录，然后到主目录运行python ./ctpn/demo.py，即可将demo下的图片进行标注

2.文字识别

 Git地址：https://github.com/xiaofengShi/CHINESE-OCR/blob/master/demo.py