docker部署PaddleOCR

用 docker 部署 PaddleOCR 是因为 PaddleOCR 以源码安装的方式比较繁杂，要注意比较多的细节，而且很多环境往往是没有外网的，因此Docker就是一个很好的解决方案，它将开放所需要的环境都封装在镜像中了，方便部署使用。

1. PaddleOCR 镜像制作

Dockerfile文件

# Version: 2.0.0
FROM paddlepaddle/paddle:2.6.1

# PaddleOCR base on Python3.7
RUN pip install --no-cache-dir --upgrade pip -i https://mirror.baidu.com/pypi/simple

RUN pip install --no-cache-dir paddlehub --upgrade -i https://mirror.baidu.com/pypi/simple

RUN pip uninstall -y astroid

RUN pip install astroid==2.12.2

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

WORKDIR /PaddleOCR

RUN pip install --no-cache-dir -r requirements.txt -i https://mirror.baidu.com/pypi/simple

RUN mkdir -p /PaddleOCR/inference/

# Download orc detect model(light version). if you want to change normal version, you can change ch_ppocr_mobile_v2.0_det_infer to ch_ppocr_server_v2.0_det_infer, also remember change det_model_dir in deploy/hubserving/ocr_system/params.py）
ADD https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar /PaddleOCR/inference/
ADD https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar /PaddleOCR/inference/
ADD https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_infer.tar /PaddleOCR/inference/


RUN tar xf /PaddleOCR/inference/ch_PP-OCRv3_det_infer.tar -C /PaddleOCR/inference/
RUN tar xf /PaddleOCR/inference/ch_ppocr_mobile_v2.0_cls_infer.tar -C /PaddleOCR/inference/
RUN tar xf /PaddleOCR/inference/ch_PP-OCRv3_rec_infer.tar -C /PaddleOCR/inference/

RUN pip install protobuf==3.20.0 -i https://mirror.baidu.com/pypi/simple

EXPOSE 8866

CMD ["/bin/bash","-c","hub install deploy/hubserving/ocr_system/ && hub serving start -m ocr_system"]

创建好Dockerfile文件后，执行如下命令即可自动构建镜像，要预留足够的存储空间，构建完成后大概6G多，整个构建过程根据网速定，我花了差不多1.5小时才构建完。

docker build -t paddle-ocr:2.6.1 .

2. 运行

docker run -dp 8866:8866 --name ocr paddle-ocr:2.6.1

当然也可以用 docker-compose 管理

version: '3'
services:
  ocr:
    image: paddle-ocr:2.6.1
    restart: always
    container_name: ocr
    ports:
      - 8866:8866

3. 测试调用

package com.aaron;

import java.io.IOException;
import java.io.InputStream;
import java.nio.charset.Charset;
import java.nio.file.Files;
import java.nio.file.Paths;
import java.text.ParseException;
import org.apache.http.HttpResponse;
import org.apache.http.HttpStatus;
import org.apache.http.client.HttpClient;
import org.apache.http.client.methods.HttpPost;
import org.apache.http.entity.StringEntity;
import org.apache.http.impl.client.DefaultHttpClient;
import org.apache.http.util.EntityUtils;

import com.alibaba.fastjson.JSONArray;
import com.alibaba.fastjson.JSONObject;

import sun.misc.BASE64Encoder;
public class InvoiceOcr {

    //接口地址
    private static String apiURL = "http://192.168.0.101:8866/predict/ocr_system";
    private HttpClient httpClient = null;
    private HttpPost method = null;
    private long startTime = 0L;
    private long endTime = 0L;
    private int status = 0;
    
    public InvoiceOcr(String url) {

        if (url != null) {
            this.apiURL = url;
        }
        if (apiURL != null) {
            httpClient = new DefaultHttpClient();
            method = new HttpPost(apiURL);

        }
    }

    public static void main(String[] args) throws ParseException {
        InvoiceOcr ac = new InvoiceOcr(apiURL);
        JSONArray arry = new JSONArray();
        JSONObject param = new JSONObject();
        arry.add(imageToBase64("E:\\第16页-0.png"));
        param.put("images",arry);
        String result = ac.post(param.toJSONString());
        System.out.println(result);
    }

    /**
     * 调用 API
     */
    public String post(String parameters) {
        String body = null;
        if (method != null & parameters != null && !"".equals(parameters.trim())) {
            try {

                // 建立一个NameValuePair数组，用于存储欲传送的参数
                method.addHeader("Content-type","application/json");
                method.setHeader("Accept", "application/json");
                method.setEntity(new StringEntity(parameters, Charset.forName("UTF-8")));
                startTime = System.currentTimeMillis();

                HttpResponse response = httpClient.execute(method);

                endTime = System.currentTimeMillis();
                int statusCode = response.getStatusLine().getStatusCode();

                System.out.println("statusCode:" + statusCode);
                System.out.println("调用API 花费时间(单位：毫秒)：" + (endTime - startTime));
                if (statusCode != HttpStatus.SC_OK) {
                    System.out.println("Method failed:" + response.getStatusLine());
                    status = 1;
                }
                body = EntityUtils.toString(response.getEntity(),"utf-8");
            } catch (IOException e) {
                // 网络错误
                status = 3;
                e.printStackTrace();
            } finally {
                System.out.println("调用接口状态：" + status);
            }
        }
        return body;
    }

    public static String imageToBase64(String path) {
        byte[] data = null;
        // 读取图片字节数组
        try {
            InputStream in = Files.newInputStream(Paths.get(path));
            data = new byte[in.available()];
            in.read(data);
            in.close();
        } catch (IOException e) {
            e.printStackTrace();
        }
        // 对字节数组Base64编码
        BASE64Encoder encoder = new BASE64Encoder();
        return encoder.encode(data);// 返回Base64编码过的字节数组字符串
    }
    /**
     * 0.成功 1.执行方法失败 2.协议错误 3.网络错误
     *
     * @return the status
     */
    public int getStatus() {
        return status;
    }

    /**
     * @param status
     *            the status to set
     */
    public void setStatus(int status) {
        this.status = status;
    }

    /**
     * @return the startTime
     */
    public long getStartTime() {
        return startTime;
    }

    /**
     * @return the endTime
     */
    public long getEndTime() {
        return endTime;
    }
}

4. 遇到的问题

1. 报 protobuf 包依赖冲突，后在Dockerfile 文件里后面加了这行，安装个低版本的 protobuf 解决

RUN pip install protobuf==3.20.0 -i https://mirror.baidu.com/pypi/simple

2. PaddlePaddle出现“非法指令”或“illegal instruction”

原因：PaddlePaddle使用avx SIMD指令提高cpu执行效率，因此错误的使用二进制发行版可能会导致这种错误，请先判断你的电脑是否支持AVX指令集，再选择性的安装支持AVX指令集的PaddlePaddle还是不支持AVX指令集的PaddlePaddle，或者使用Docker镜像来安装最新版本的PaddlePaddle，Docker镜像中的PaddlePaddle默认支持是支持AVX指令集的，可以提高cpu的执行效率。在启动的时候会检查系统是否支持PaddlePaddle初始化时所需要的资源，从AVX指令集开始检查。
解决：遇到此问题可以先用 lscpu 命令查看docker宿主机是否支持AVX指令集，我是换台服务器解决。

3. 容器运行后，调用的时候报了 module ‘numpy’ has no attribute ‘int’. 异常

AttributeError: module 'numpy' has no attribute 'int'.
`np.int` was a deprecated alias for the builtin `int`. To avoid this error in existing code, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.
The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at:

原因：np.int 在 NumPy 1.20 中已弃用，在 NumPy 1.24 中已删除。
解决：将容器中的 /PaddleOCR/deploy/hubserving/ocr_system/module.py
/np.ini 文件中的 np.int 更改为np.int_

具体操作如下：

# 进入容器内
docker exec -it ocr /bin/bash
# 修改文件
vim /PaddleOCR/deploy/hubserving/ocr_system/module.py
/np.ini

# 如下位置
            for dno in range(dt_num):
                text, score = rec_res[dno]
                rec_res_final.append({
                    'text': text,
                    'confidence': float(score),
                    'text_region': dt_boxes[dno].astype(np.int_).tolist()
                })
            all_results.append(rec_res_final)
        return all_results

# 修改后重启容器即可解决
docker restart ocr