智能文本提取

注意：在学习使用不同函数之前，建议先阅读请求描述，了解基本的PDF处理流程。使用不同函数时，可以在上传文件时设置各自的特殊参数。其他基本步骤一致。

智能文本提取：

java

{    
  "lang": "auto",  
}

所需参数

lang：支持的类型和定义

auto - 自动分类语言
english - 英语
chinese - 简体中文
chinese_tra - 繁体中文
korean - 韩语
japanese - 日语
latin - 拉丁语
devanagari - 梵文字母

示例

授权

您需要将认证响应中的和替换为从控制台获取的 publicKey 和 secretKey，并使用 accessToken。

curljava

curl

curl --location --request POST 'https://api-server.compdf.com/server/v1/oauth/token' \
--header 'Content-Type: application/json' \
--data-raw '{
    "publicKey": "publicKey",
    "secretKey": "secretKey"
}'

java

 import java.io.*;
 import okhttp3.*;
 public class main {
   public static void main(String []args) throws IOException{
     OkHttpClient client = new OkHttpClient().newBuilder()
       .build();
     MediaType mediaType = MediaType.parse("text/plain");
     RequestBody body = RequestBody.create(mediaType, "{\n    \"publicKey\": \"{{public_key}}\",\n    \"secretKey\": \"{{secret_key}}\"\n}");
     Request request = new Request.Builder()
       .url("https://api-server.compdf.com/server/v1/oauth/token")
       .method("POST", body)
       .build();
     Response response = client.newCall(request).execute();
   }
 }

创建任务

您需要将替换为上一步获取的 accessToken，将 ***替换为所需的界面和任务错误消息语言类型。请求成功后，您将从响应中收到 taskId。

curljava

curl

curl --location --request GET 'https://api-server.compdf.com/server/v1/task/documentAI/ocr' \
--header 'Authorization: Bearer accessToken'

java

 import java.io.*;
 import okhttp3.*;
 public class main {
   public static void main(String []args) throws IOException{
     OkHttpClient client = new OkHttpClient().newBuilder()
       .build();
     MediaType mediaType = MediaType.parse("text/plain");
     RequestBody body = RequestBody.create(mediaType, "");
     Request request = new Request.Builder()
       .url("https://api-server.compdf.com/server/v1/task/documentAI/ocr?language={{language}}")
       .method("GET", body)
       .addHeader("Authorization", "Bearer {{accessToken}}")
       .build();
     Response response = client.newCall(request).execute();
   }
 }

上传文件

将替换为您要转换的文件，将替换为您上一步获取到的 taskId，将替换为您需要的界面错误信息语言类型，将替换为您第一步获取到的 accessToken。

-支持图像格式: jpg,jpeg,png,bmp

curljava

curl

curl --location --request POST 'https://api-server.compdf.com/server/v1/file/upload' \
--header 'Authorization: Bearer accessToken' \
--form 'file=@"test.png"' \
--form 'taskId="taskId"' \
--form 'password=""' \
--form 'parameter="{    \"lang\": \"auto\"    }"' \

java

 import java.io.*;
 import okhttp3.*;
 public class main {
   public static void main(String []args) throws IOException{
     OkHttpClient client = new OkHttpClient().newBuilder()
       .build();
     MediaType mediaType = MediaType.parse("text/plain");
     RequestBody body = new MultipartBody.Builder().setType(MultipartBody.FORM)
       .addFormDataPart("file","{{file}}",
                        RequestBody.create(MediaType.parse("application/octet-stream"),
                                           new File("<file>")))
       .addFormDataPart("taskId","{{taskId}}")
       .addFormDataPart("language","{{language}}")
       .addFormDataPart("parameter","{    \"lang\": \"auto\"    }")
       .build();
     Request request = new Request.Builder()
       .url("https://api-server.compdf.com/server/v1/file/upload")
       .method("POST", body)
       .addHeader("Authorization", "Bearer {{accessToken}}")
       .build();
     Response response = client.newCall(request).execute();
   }
 }

执行任务

将替换为 创建任务 步骤中获取的 taskId，将替换为第一步获取的 access_token，将替换为所需的界面错误信息语言类型。

curljava

curl

curl --location -g --request GET 'https://api-server.compdf.com/server/v1/execute/start?taskId=taskId' \
--header 'Authorization: Bearer accessToken'

java

 import java.io.*;
 import okhttp3.*;
 public class main {
  public static void main(String []args) throws IOException{
    OkHttpClient client = new OkHttpClient().newBuilder()
      .build();
    MediaType mediaType = MediaType.parse("text/plain");
    RequestBody body = RequestBody.create(mediaType, "");
    Request request = new Request.Builder()
      .url("https://api-server.compdf.com/server/v1/execute/start?taskId={{taskId}}&language={{language}}")
      .method("GET", body)
      .addHeader("Authorization", "Bearer {{accessToken}}")
      .build();
    Response response = client.newCall(request).execute();
  }
 }

获取任务信息

将替换为 创建任务 步骤中获取的 taskId，将替换为第一步中获取的 access_token。

curljava

curl

curl --location -g --request GET 'https://api-server.compdf.com/server/v1/task/taskInfo?taskId=taskId' \
--header 'Authorization: Bearer accessToken'

java

 import java.io.*;
 import okhttp3.*;
 public class main {
   public static void main(String []args) throws IOException{
     OkHttpClient client = new OkHttpClient().newBuilder()
       .build();
     MediaType mediaType = MediaType.parse("text/plain");
     RequestBody body = RequestBody.create(mediaType, "");
     Request request = new Request.Builder()
       .url("https://api-server.compdf.com/server/v1/task/taskInfo?taskId={{taskId}}")
       .method("GET", body)
       .addHeader("Authorization", "Bearer {{accessToken}}")
       .build();
     Response response = client.newCall(request).execute();
   }
 }

结果

文件类型	说明
.JSON	OCR 识别结果

内容

参数	说明
cost	OCR 识别时间
boxes	输入图片所有检测到的物体框位置
text	OCR 识别内容
rec_scores	OCR 文本识别分数，分数越高，结果越可信

java

{
        "cost": 149,
        "boxes": [
            [
                150,
                71,
                198,
                71,
                198,
                110,
                150,
                110
            ],
            [
                74,
                117,
                274,
                120,
                273,
                166,
                73,
                163
            ],
            [
                99,
                179,
                249,
                182,
                249,
                208,
                99,
                205
            ],
            [
                65,
                203,
                276,
                205,
                276,
                230,
                65,
                228
            ]
        ],
        "text": [
            "EPPING",
            "Twinned with",
            "Eppingen,Germany"
        ],
        "rec_scores": [
            0.46275457739830017,
            0.9971449971199036,
            0.9649983048439026,
            0.9587073922157288
        ]
}

智能文本提取 ​

智能文本提取