这个爱酱不太聪明

使用docker安装PaddleOCR

2024年09月29日4 min read
PaddleOCR

1. 安装必要的系统工具

sudo yum install -y yum-utils device-mapper-persitent-data lvm2
sudo yum install -y epel-release

2. 添加阿里云镜像仓库

sudo yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo

3.更新并安装Docker-CE

sudo yum makecache fast 
sudo yum -y install docker-ce

4. 开启Docker服务

sudo service docker start

5. 安装DockerCompose

sudo curl -L "https://github.com/docker/compose/releases/latest/download/docker-compose-Linux-x86_64" -o /usr/local/bin/docker-compose

6. 授予执行权限

sudo chmod +x /usr/local/bin/docker-compose

7. 创建ocr-cpu目录

cd /home
mkdir ocr-cpu
cd ocr-cpu

8. 编写Dockerfile

# 使用最新版 Python  
FROM python:slim  
  
# 更新 apt-get 并安装必要的依赖  
RUN apt-get update && \  
    apt-get install -y git libgomp1 libgl1-mesa-glx libglib2.0-0 && \  
    rm -rf /var/lib/apt/lists/*  
  
# 更新 pip 并安装 setuptools  
RUN pip install --upgrade pip setuptools -i https://mirrors.aliyun.com/pypi/simple  
  
# 安装 paddlepaddle 和 paddlehub（使用最新版）  
RUN pip install paddlepaddle paddlehub -i https://mirrors.aliyun.com/pypi/simple  
  
# **关键修改**：降级 protobuf 到兼容版本 3.20.x，以避免 TypeError 错误  
RUN pip install protobuf==3.20.* -i https://mirrors.aliyun.com/pypi/simple  
  
# 克隆 PaddleOCR 仓库  
RUN git clone https://gitee.com/paddlepaddle/PaddleOCR.git /PaddleOCR  
WORKDIR /PaddleOCR  
  
# 直接安装依赖  
RUN pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple  
  
# **关键修改**：设置环境变量以避免 protobuf 错误  
ENV PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python  
  
# 下载并解压模型  
ADD https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar /PaddleOCR/inference/  
RUN tar xf /PaddleOCR/inference/ch_PP-OCRv3_det_infer.tar -C /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_rec_infer.tar -C /PaddleOCR/inference/  
  
ADD https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar /PaddleOCR/inference/  
RUN tar xf /PaddleOCR/inference/ch_ppocr_mobile_v2.0_cls_infer.tar -C /PaddleOCR/inference/  
  
# 安装 OCR 模块  
RUN hub install deploy/hubserving/ocr_system/  
  
# 暴露端口  
EXPOSE 8866  
  
# 启动 PaddleHub 服务  
CMD ["hub", "serving", "start", "-m", "ocr_system"]

9. 创建 docker-compose.yml 文件

version: '3'  # 正确的版本声明  
  
services:  
  ocr-cpu:  
    image: ocr-cpu  # 注意这里的冒号应为连字符  
    restart: always  
    hostname: ocr-cpu  
    container_name: ocr-cpu  
    ports:  
      - "8866:8866"  # 使用引号包裹端口映射

10. 运行Docker构建命令，docker-compose命令部署

docker build --network=host -t ocr-cpu .
docker-compose up -d

11. 查看的容器

image.png

12. 编写后端测试代码

package main  
  
import (  
    "bytes"  
    "encoding/base64"    "encoding/json"    "fmt"    "io/ioutil"    "net/http"    "os")  
  
func main() {  
    // URL 和图片路径设置  
    url := "http://192.168.31.203:8866/predict/ocr_system"  
    imagePath := `C:\Users\doudou\Desktop\微信截图_20240929171217.png`  
  
    // 读取图片  
    imgFile, err := os.Open(imagePath)  
    if err != nil {  
       fmt.Println("Error opening image file:", err)  
       return  
    }  
    defer imgFile.Close()  
  
    imgData, err := ioutil.ReadAll(imgFile)  
    if err != nil {  
       fmt.Println("Error reading image file:", err)  
       return  
    }  
  
    // Base64 编码  
    imgBase64 := base64.StdEncoding.EncodeToString(imgData)  
    data := map[string]interface{}{  
       "images": []string{imgBase64},  
    }  
    jsonData, err := json.Marshal(data)  
    if err != nil {  
       fmt.Println("Error marshaling JSON:", err)  
       return  
    }  
  
    // 发送请求  
    req, err := http.NewRequest("POST", url, bytes.NewBuffer(jsonData))  
    if err != nil {  
       fmt.Println("Error creating request:", err)  
       return  
    }  
    req.Header.Set("Content-Type", "application/json")  
  
    client := &http.Client{}  
    response, err := client.Do(req)  
    if err != nil {  
       fmt.Println("Error sending request:", err)  
       return  
    }  
    defer response.Body.Close()  
  
    body, err := ioutil.ReadAll(response.Body)  
    if err != nil {  
       fmt.Println("Error reading response body:", err)  
       return  
    }  
  
    // 解析 JSON 响应  
    var result map[string]interface{}  
    if err := json.Unmarshal(body, &result); err != nil {  
       fmt.Println("Error unmarshaling JSON:", err)  
       return  
    }  
  
    // 检查状态  
    if result["status"] == "000" {  
       if results, ok := result["results"].([]interface{}); ok {  
          for _, res := range results {  
             if resList, ok := res.([]interface{}); ok {  
                for _, item := range resList {  
                   if itemMap, ok := item.(map[string]interface{}); ok {  
                      text := itemMap["text"].(string)  
                      confidence := itemMap["confidence"].(float64)  
                      fmt.Printf("Text: %s, Confidence: %.2f\n", text, confidence)  
                   }  
                }  
             }  
          }  
       }  
    } else {  
       fmt.Println("Error in response:", result["msg"])  
    }  
}

12. 测试结果

微信截图_20240929171217.png

image.png