先安装python3.7

1.安装编译相关工具

yum -y groupinstall “Development tools”
yum -y install zlib-devel bzip2-devel openssl-devel ncurses-devel sqlite-devel readline-devel tk-devel gdbm-devel db4-devel libpcap-devel xz-devel
yum install libffi-devel -y

2.下载安装包解压

cd /www/
wget https://www.python.org/ftp/python/3.7.0/Python-3.7.0.tar.xz
tar -xvJf Python-3.7.0.tar.xz

3.编译安装

mkdir /usr/local/python37 #创建编译安装目录
cd /www/Python-3.7.0
./configure --prefix=/www/python37/
make && make install

4.Python写入环境变量

vi /etc/profile

export PATH="/www/python37/bin/:$PATH"
写到profile文件，然后source下

source /etc/profile

做个python软链接指向Python3，效果如下图

cd /www/python37/bin
ln -s python3.7 python

验证下python环境

安装PaddlePaddle

4.安装PaddlePaddle(安装过程会比较久)(可以参考官方安装文档)

4.1 安装PaddlePaddle

您的机器安装的是CUDA9或CUDA10，请运行以下命令安装

python3 -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple

您的机器是CPU，请运行以下命令安装

python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple

更多的版本需求，请参照飞桨官网安装文档中的说明进行操作。

4.2 安装PaddleOCR whl包

pip install "paddleocr>=2.0.1"

安装成功后如下图

特别注意：如果执行图片OCR遇到这个异常，就需要把scikit-image降级scikit-image-0.14.2版本

ImportError: dlopen: cannot load any more object with static TLS
It seems that scikit-image has not been built correctly.

Your install of scikit-image appears to be broken.
Try re-installing the package following the instructions at:
https://scikit-image.org/docs/stable/install.html

scikit-image降级命令如下：

pip uninstall scikit-image
pip install scikit-image==0.14.2

5.编写Python脚本进行图片OCR

# 安装包：
# pip install paddleocr==2.0.2 -i https://mirror.baidu.com/pypi/simple

# pip install paddlepaddle==2.0.2 -i https://mirror.baidu.com/pypi/simple

# ==========================================================================

# ref:https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/doc/doc_ch/whl.md

# paddlepaddle只能安装2.0.0rc1版本,运行内存消耗稳定在3G左右;GPU显存较高的时候，安装和CUDA版本一直的paddlepaddle-gpu,运行时只需要开启GPU加速即可
# ==========================================================================

from paddleocr import PaddleOCR, draw_ocr
import cv2
import sys
if __name__=="__main__":

    ocr = PaddleOCR()

    print("-------------------------")

    arg1 = sys.argv[1]
    result = ocr.ocr(arg1)
    ret_dict={'save_path':'','data':[]}
    for item in result:
        ret_dict['data'].append({'text':item[1][0],'confidence':item[1][1],'text_box_position':item[0]})

    print("["+str(ret_dict)+"]")

6.执行代码测色

python ocr.py screen.png #screen.png为本地图片路径

正常情况得到结果如图：