Linux安装使用paddleOCR

一、安装paddle

1. 安装python3.6

yum install python3

2. 升级pip

python -m pip install -U pip

3. 安装paddle

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

4. 验证安装是否成功

python3
import paddle
paddle.utils.run_check()
//出现PaddlePaddle is installed successfully!则安装成功

二、安装paddleOCR

1. 安装git

yum install git

2. 使用git下载paddleOCR

git clone https://gitee.com/PaddlePaddle/PaddleOCR.git

3. 切换路径并安装依赖

cd PaddleOCR
pip install -r requirements.txt

4. 错误处理（无则跳过）

python-Levenshtein安装错误

error: command 'gcc' failed with exit status 1错误

统一解决方法：

sudo yum install python3-devel

5. 安装inference model(参考官网)

!mkdir inference && cd inference
# 下载中英文检测模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar && tar xf ch_ppocr_mobile_v2.0_det_infer.tar
# 下载中英文方向分类器模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar && tar xf ch_ppocr_mobile_v2.0_cls_infer.tar
# 下载中英文识别模型并解压
wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar && tar xf ch_ppocr_mobile_v2.0_rec_infer.tar

6. 快速运行

python3 tools/infer/predict_system.py --image_dir="./doc/imgs/11.jpg" --det_model_dir="./inference/ch_ppocr_mobile_v2.0_det_infer/"  --rec_model_dir="./inference/ch_ppocr_mobile_v2.0_rec_infer/"