java实现百度云OCR文字识别 高精度OCR识别身份证信息

本文为大家分享了java实现百度云ocr识别的具体代码，高精度ocr识别身份证信息，供大家参考，具体内容如下

1.通用ocr文字识别

这种ocr只能按照识别图片中的文字，且是按照行识别返回结果，精度较低。

首先引入依赖包：

<dependency>
  <groupid>com.baidu.aip</groupid>
  <artifactid>java-sdk</artifactid>
  <version>4.6.0</version>
</dependency>

通过ocr工具类：

package util;
 
import com.baidu.aip.ocr.aipocr;
import org.json.jsonobject;
import java.util.hashmap;
 
 
public class ocrapi {
  private static final string app_id = "你的 app id";
  private static final string api_key = "xb12m5t4js2n7";
  private static final string secret_key = "9xvx9gpcsbsutz";
 
  private static aipocr getaipclient() {
    return getaipclient(api_key, secret_key);
  }
 
  public static aipocr getaipclient(string apikey, string secretkey) {
    aipocr client = new aipocr(app_id, apikey, secretkey);
    // 可选：设置网络连接参数
    client.setconnectiontimeoutinmillis(2000);
    client.setsockettimeoutinmillis(60000);
    return client;
  }
 
  public static string result(aipocr client) {
    // 传入可选参数调用接口
    hashmap<string, string> options = new hashmap<>();
    options.put("language_type", "chn_eng");
    options.put("detect_direction", "true");
    options.put("detect_language", "true");
    options.put("probability", "true");
 
    jsonobject res = client.basicgeneralurl(
        "https://lichunyu1234.oss-cn-shanghai.aliyuncs.com/1.png", options);
    return res.tostring(2);
  }
 
  public static void main(string[] args) {
    system.out.println(result(getaipclient()));
  }
}

结果如下，识别有两行信息（words即是识别的信息）：

2.高精度ocr识别身份证信息 

这种就比较高精度，且按照分类显示，返回数据更友好，高可用。

2.1 接口说明及请求参数是地址官方截图如下：

2.2 ocr身份证识别工具类

package util;
 
import com.alibaba.druid.util.base64;
import com.alibaba.fastjson.jsonobject;
import java.io.*;
import java.net.*;
import java.nio.charset.standardcharsets;
import java.util.list;
import java.util.map;
 
public class ocrutil {
  // access_token获取
  private static final string access_token_host = "https://aip.baidubce.com/oauth/2.0/token?";
  // 身份证识别请求url
  private static final string ocr_host = "https://aip.baidubce.com/rest/2.0/ocr/v1/idcard?";
  // apikey,secretkey
  private static final string api_key ="xb12m5t4js";
  private static final string secret_key = "9xvx9gpcsbsut";
 
 
  // 获取百度云ocr的授权access_token
  public static string getaccesstoken() {
    return getaccesstoken(api_key, secret_key);
  }
 
  /**
   * 获取百度云ocr的授权access_token
   * @param apikey
   * @param secretkey
   * @return
   */
  public static string getaccesstoken(string apikey, string secretkey) {
    string accesstokenurl = access_token_host
        // 1. grant_type为固定参数
        + "grant_type=client_credentials"
        // 2. 官网获取的 api key
        + "&client_id=" + apikey
        // 3. 官网获取的 secret key
        + "&client_secret=" + secretkey;
 
    try {
      url url = new url(accesstokenurl);
      // 打开和url之间的连接
      httpurlconnection connection = (httpurlconnection) url.openconnection();
      connection.setrequestmethod("get");
      connection.connect();
 
      // 获取响应头
      map<string, list<string>> map = connection.getheaderfields();
      // 遍历所有的响应头字段
      for (string key : map.keyset()) {
        system.out.println(key + "---->" + map.get(key));
      }
 
      // 定义 bufferedreader输入流来读取url的响应
      bufferedreader bufferedreader = new bufferedreader(new inputstreamreader(connection.getinputstream()));
      stringbuilder result = new stringbuilder();
      string inputline;
      while ((inputline = bufferedreader.readline()) != null) {
        result.append(inputline);
      }
      jsonobject jsonobject = jsonobject.parseobject(result.tostring());
      return jsonobject.getstring("access_token");
 
    } catch (exception e) {
      e.printstacktrace();
      system.err.print("获取access_token失败");
    }
    return null;
  }
 
  /**
   * 获取身份证识别后的数据
   * @param imageurl
   * @param idcardside
   * @return
   */
  public static string getstringidentitycard(file imageurl, string idcardside) {
    // 身份证ocr的http url+鉴权token
    string ocrurl = ocr_host+"access_token="+getaccesstoken();
    system.out.println(ocrurl);
    system.out.println("***************************************************");
    system.out.println(getaccesstoken());
    // 对图片进行base64处理
    string image = encodeimagetobase64(imageurl);
    // 请求参数
    string requestparam = "detect_direction=true&id_card_side="+idcardside+"&image="+image;
 
    try {
      // 请求ocr地址
      url url = new url(ocrurl);
      httpurlconnection connection = (httpurlconnection) url.openconnection();
      // 设置请求方法为post
      connection.setrequestmethod("post");
 
      // 设置请求头
      connection.setrequestproperty("content-type", "application/x-www-form-urlencoded");
      connection.setrequestproperty("apikey", api_key);
      connection.setdooutput(true);
      connection.getoutputstream().write(requestparam.getbytes(standardcharsets.utf_8));
      connection.connect();
 
      // 定义 bufferedreader输入流来读取url的响应
      bufferedreader bufferedreader = new bufferedreader(new inputstreamreader(connection.getinputstream(), standardcharsets.utf_8));
      stringbuilder result = new stringbuilder();
      string inputline;
      while ((inputline = bufferedreader.readline()) != null) {
        result.append(inputline);
      }
      bufferedreader.close();
      return result.tostring();
    } catch (exception e) {
      e.printstacktrace();
      system.err.println("身份证ocr识别异常");
      return null;
    }
  }
 
  /**
   * 对图片url进行base64编码处理
   * @param imageurl
   * @return
   */
  public static string encodeimagetobase64(file imageurl) {
    // 将图片文件转化为字节数组字符串，并对其进行base64编码处理
    byte[] data = null;
    try {
      inputstream inputstream = new fileinputstream(imageurl);
      data = new byte[inputstream.available()];
      inputstream.read(data);
      inputstream.close();
 
      // 对字节数组base64编码
      return urlencoder.encode(base64.bytearraytobase64(data), "utf-8");
    } catch (exception e) {
      e.printstacktrace();
      return null;
    }
 
  }
 
  /**
   * 提取ocr识别身份证有效信息
   * @param
   * @return
   */
  public static map<string, string> getidcardinfo(multipartfile image, int idcardside) {
    string value = getstringidentitycard(image, idcardside);
    string side;
    if (idcardside == 1) {
      side = "正面";
    }else {
      side = "背面";
    }
    map<string, string> map = new hashmap<>();
    jsonobject jsonobject = jsonobject.parseobject(value);
    jsonobject words_result = jsonobject.getjsonobject("words_result");
    if (words_result == null || words_result.isempty()) {
      throw new myexception("请提供身份证"+side+"图片");
    }
    for (string key : words_result.keyset()) {
      jsonobject result = words_result.getjsonobject(key);
      string info = result.getstring("words");
      switch (key) {
        case "姓名":
          map.put("name", info);
          break;
        case "性别":
          map.put("sex", info);
          break;
        case "民族":
          map.put("nation", info);
          break;
        case "出生":
          map.put("birthday", info);
          break;
        case "住址":
          map.put("address", info);
          break;
        case "公民身份号码":
          map.put("idnumber", info);
          break;
        case "签发机关":
          map.put("issuedorganization", info);
          break;
        case "签发日期":
          map.put("issuedat", info);
          break;
        case "失效日期":
          map.put("expiredat", info);
          break;
      }
    }
    return map;
 
  }
 
}

官方返回示例：

对于身份证识别有个大坑：

1.有的base64编码后有头部“base64：”要去掉，阿里巴巴的base64可以正常使用。

2.ocr识别官方只说明图片要base64编码，但是实际上还是要再urlencode再编码一次才可以。

以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持。