1.环境搭建

1.1 创建conda环境

conda create --name paddlehub python=3.8
conda activate paddlehub

1.2 安装paddlepaddle和paddlehub

pip install paddlepaddle -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install paddlehub -i https://pypi.tuna.tsinghua.edu.cn/simple

1.3 安装依赖

pip install shapely pyclipper

2. 移动端模型部署

2.1 安装移动端模型

hub install chinese_ocr_db_crnn_mobile

报错1：

增加download函数

vim /home/yinsuan/miniconda3/envs/paddlehub/lib/python3.8/site-packages/aistudio_sdk/hub.py 

def download(url, target_path):os.makedirs(os.path.dirname(target_path), exist_ok=True)response = requests.get(url, stream=True)with open(target_path, 'wb') as f:for chunk in response.iter_content(chunk_size=1024):if chunk:f.write(chunk)print(f"Downloaded {url} to {target_path}")

报错2：

export PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python

安装成功

2.2 测试

将其中的np.int 修改为int即可

vim /home/yinsuan/.paddlehub/modules/chinese_ocr_db_crnn_mobile/module.py

hub run chinese_ocr_db_crnn_mobile --input_path "/PATH/TO/IMAGE"

测试成功

3. 服务部署

3.1 启动PaddleHub Serving

hub serving start -m chinese_ocr_db_crnn_mobile

3.2 发送预测请求

import requests
import json
import cv2
import base64def cv2_to_base64(image):data = cv2.imencode('.jpg', image)[1]return base64.b64encode(data.tostring()).decode('utf8')# 发送HTTP请求
data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]}
headers = {"Content-type": "application/json"}
url = "http://127.0.0.1:8866/predict/chinese_ocr_db_crnn_mobile"
r = requests.post(url=url, headers=headers, data=json.dumps(data))# 打印预测结果
print(r.json()["results"])

参考：https://www.paddlepaddle.org.cn/hubdetail?name=chinese_ocr_db_crnn_mobile&en_category=TextRecognition