java通过百度AI开发平台提取身份证图片中的文字信息
Posted 五柳先生柳三变
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了java通过百度AI开发平台提取身份证图片中的文字信息相关的知识，希望对你有一定的参考价值。
废话不多说，直接上代码。。。
IdCardDemo.java
1 package com.wulss.baidubce; 2 3 import java.io.BufferedReader; 4 import java.io.InputStreamReader; 5 import java.net.HttpURLConnection; 6 import java.net.URL; 7 import java.net.URLEncoder; 8 import java.util.Map; 9 10 import com.wulss.utils.Base64Util; 11 import com.wulss.utils.FileUtil; 12 import com.wulss.utils.HttpUtil; 13 14 /** 15 * 16 * @Descript TODO (身份证图片识别 案例) 17 * @author yeting 18 * @date 2019年4月18日 19 * 20 */ 21 public class IdCardDemo { 22 23 private static final String URL_IDCARD = "https://aip.baidubce.com/rest/2.0/ocr/v1/idcard";//身份证识别地址 24 private static final String URL_ACCESSTOKEN = "https://aip.baidubce.com/oauth/2.0/token?"; // 百度AI开发平台 获取token的地址 25 private static final String API_KEY = "afdH343CAt342YFT7F"; // 百度AI开发平台 获取的 API Key 更新为你注册的 26 private static final String SECRET_KEY = "js45sdfqRFF65gOd667sd1R7sdr"; // 百度AI开发平台 获取的 Secret Key 更新为你注册的 27 28 /** 29 * 获取API访问token 30 * 该token有一定的有效期，需要自行管理，当失效时需重新获取. 31 * @param ak - 百度云官网获取的 API Key 32 * @param sk - 百度云官网获取的 Securet Key 33 * @return assess_token 示例： 34 * "24.460da4889caad24cccdb1fea17221975.2592000.1491995545.282335-1234567" 35 */ 36 public static String getAccessToken() { 37 String getAccessTokenUrl = URL_ACCESSTOKEN 38 + "grant_type = client_credentials" // 1. grant_type为固定参数 39 + "&client_id = " + API_KEY // 2. 官网获取的 API Key 40 + "&client_secret = " + SECRET_KEY; // 3. 官网获取的 Secret Key 41 String accessToken = ""; 42 try { 43 URL realUrl = new URL(getAccessTokenUrl); 44 45 // 打开和URL之间的连接 46 HttpURLConnection connection = (HttpURLConnection) realUrl.openConnection(); 47 connection.setRequestMethod("GET"); 48 connection.connect(); 49 50 // 获取所有响应头字段 51 // Map<String, List<String>> map = connection.getHeaderFields(); 52 // 遍历所有的响应头字段 53 // for (String key : map.keySet()) { 54 // System.err.println(key + "--->" + map.get(key)); 55 // } 56 57 // 定义 BufferedReader输入流来读取URL的响应 58 BufferedReader in = new BufferedReader(new InputStreamReader(connection.getInputStream())); 59 String result = ""; 60 String line; 61 while ((line = in.readLine()) != null) { 62 result += line; 63 } 64 65 System.err.println("result:" + result); 66 67 org.json.JSONObject jsonObject = new org.json.JSONObject(result); 68 accessToken = jsonObject.getString("access_token"); 69 } catch (Exception e) { 70 System.err.printf("获取token失败！"); 71 e.printStackTrace(System.err); 72 } 73 return accessToken; 74 } 75 76 /** 77 * 身份证识别请求 78 * @param side 识别身份证正面 front;识别身份证背面 back; 79 * @param filePath 图片路径 80 * @param accessToken 线上环境有过期时间， 客户端可自行缓存，过期后重新获取。 81 * @return 返回身份证号码 82 */ 83 public static String requestIdCard(String side,String filePath,String accessToken) { 84 String result = ""; 85 86 try { 87 //1.请求获取结果 88 String requestParams = "id_card_side = " + side 89 + "&" + URLEncoder.encode("image", "UTF-8") 90 + "=" + URLEncoder.encode(Base64Util.encode(FileUtil.readFileByBytes(filePath)), "UTF-8"); 91 92 result = HttpUtil.post(URL_IDCARD, accessToken, requestParams);//返回json格式的结果 93 System.out.println(result); 94 95 // 请求返回结果eg: 96 // String result = 97 // "[{"log_id": 3812339812321238679, "words_result_num": 6,"direction": 2, "image_status": "normal", 98 // "words_result": { 99 // "住址":{"location": {"width": 123, "top": 123, "height": 4423, "left":1232}, "words": "湖北省咸宁市茶叶巷"}, 100 // "出生": {"location":{"width": 333, "top": 339, "height": 2333, "left": 3333}, "words": "19191010"}, 101 // "姓名": {"location": {"width": 133, "top": 309, "height": 303, "left": 2205}, "words": "张三"}, 102 // "公民身份号码":{"location": {"width": 111, "top": 3333, "height": 3335, "left":333}, "words": "430124191910101234"}, 103 // "性别": {"location": {"width":222, "top": 521, "height": 304, "left": 2333}, "words": "男"}, 104 // "民族": {"location": {"width": 111, "top": 3333, "height": 22,"left": 1222}, "words": "汉"} 105 // } 106 // }]"; 107 108 // <!-- json转换工具 依赖jar包--> 109 // <dependency> 110 // <groupId>net.sf.json-lib</groupId> 111 // <artifactId>json-lib</artifactId> 112 // <version>2.4</version> 113 // <classifier>jdk15</classifier> 114 // </dependency> 115 116 //2.解析结果 117 Map<String,String> resultMap = (Map<String,String>)net.sf.json.JSONObject 118 .toBean(net.sf.json.JSONObject.fromObject(result),Map.class); 119 120 if(resultMap.get("error_code").equals("110")) { 121 return requestIdCard(side,filePath,getAccessToken()) ;//重新请求 122 }else { 123 String words = ""; 124 if(resultMap.get("image_status") != null && resultMap.get("image_status").equals("normal")) {// 正常 125 String wordsResults = resultMap.get("words_result"); 126 Map<String,String> wordsResultMap = (Map<String,String>)net.sf.json.JSONObject 127 .toBean(net.sf.json.JSONObject.fromObject(wordsResults),Map.class); 128 129 String idCardNums = wordsResultMap.get("公民身份号码"); 130 Map<String,String> idCardNumMap = (Map<String,String>)net.sf.json.JSONObject 131 .toBean(net.sf.json.JSONObject.fromObject(idCardNums),Map.class); 132 words = idCardNumMap.get("words"); 133 } 134 return words; 135 } 136 } catch (Exception e) { 137 e.printStackTrace(); 138 result = e.getMessage(); 139 } 140 141 return result; 142 } 143 144 145 }
FileUtil.java
1 package com.wulss.utils; 2 import java.io.*; 3 4 /** 5 * 文件读取工具类 6 */ 7 public class FileUtil { 8 9 /** 10 * 读取文件内容，作为字符串返回 11 */ 12 public static String readFileAsString(String filePath) throws IOException { 13 File file = new File(filePath); 14 if (!file.exists()) { 15 throw new FileNotFoundException(filePath); 16 } 17 18 if (file.length() > 1024 * 1024 * 1024) { 19 throw new IOException("File is too large"); 20 } 21 22 StringBuilder sb = new StringBuilder((int) (file.length())); 23 // 创建字节输入流 24 FileInputStream fis = new FileInputStream(filePath); 25 // 创建一个长度为10240的Buffer 26 byte[] bbuf = new byte[10240]; 27 // 用于保存实际读取的字节数 28 int hasRead = 0; 29 while ( (hasRead = fis.read(bbuf)) > 0 ) { 30 sb.append(new String(bbuf, 0, hasRead)); 31 } 32 fis.close(); 33 return sb.toString(); 34 } 35 36 /** 37 * 根据文件路径读取byte[] 数组 38 */ 39 public static byte[] readFileByBytes(String filePath) throws IOException { 40 File file = new File(filePath); 41 if (!file.exists()) { 42 throw new FileNotFoundException(filePath); 43 } else { 44 ByteArrayOutputStream bos = new ByteArrayOutputStream((int) file.length()); 45 BufferedInputStream in = null; 46 47 try { 48 in = new BufferedInputStream(new FileInputStream(file)); 49 short bufSize = 1024; 50 byte[] buffer = new byte[bufSize]; 51 int len1; 52 while (-1 != (len1 = in.read(buffer, 0, bufSize))) { 53 bos.write(buffer, 0, len1); 54 } 55 56 byte[] var7 = bos.toByteArray(); 57 return var7; 58 } finally { 59 try { 60 if (in != null) { 61 in.close(); 62 } 63 } catch (IOException var14) { 64 var14.printStackTrace(); 65 } 66 67 bos.close(); 68 } 69 } 70 } 71 }
Base64Util.java
1 package com.wulss.utils; 2 3 4 5 /** 6 * Base64 工具类 7 */ 8 public class Base64Util { 9 private static final char[] ALPHABET; 10 private static final char last2byte; 11 private static final char last4byte; 12 private static final char last6byte; 13 private static final char lead6byte; 14 private static final char lead4byte; 15 private static final char lead2byte; 16 private static final char[] encodeTable; 17 private static int[] toInt; 18 19 static { 20 ALPHABET = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/".toCharArray(); 21 last2byte = (char)Integer.parseInt("00000011", 2); 22 last4byte = (char)Integer.parseInt("00001111", 2); 23 last6byte = (char)Integer.parseInt("00111111", 2); 24 lead6byte = (char)Integer.parseInt("11111100", 2); 25 lead4byte = (char)Integer.parseInt("11110000", 2); 26 lead2byte = (char)Integer.parseInt("11000000", 2); 27 encodeTable = new char[] { ‘A‘, ‘B‘, ‘C‘, ‘D‘, ‘E‘, ‘F‘, ‘G‘, ‘H‘, ‘I‘, ‘J‘, ‘K‘, ‘L‘, ‘M‘, ‘N‘, ‘O‘, ‘P‘, ‘Q‘, ‘R‘, ‘S‘, ‘T‘, ‘U‘, ‘V‘, ‘W‘, ‘X‘, ‘Y‘, ‘Z‘, ‘a‘, ‘b‘, ‘c‘, ‘d‘, ‘e‘, ‘f‘, ‘g‘, ‘h‘, ‘i‘, ‘j‘, ‘k‘, ‘l‘, ‘m‘, ‘n‘, ‘o‘, ‘p‘, ‘q‘, ‘r‘, ‘s‘, ‘t‘, ‘u‘, ‘v‘, ‘w‘, ‘x‘, ‘y‘, ‘z‘, ‘0‘, ‘1‘, ‘2‘, ‘3‘, ‘4‘, ‘5‘, ‘6‘, ‘7‘, ‘8‘, ‘9‘, ‘+‘, ‘/‘ }; 28 Base64Util.toInt = new int[128]; 29 for (int i = 0; i < Base64Util.ALPHABET.length; ++i) { 30 Base64Util.toInt[Base64Util.ALPHABET[i]] = i; 31 } 32 } 33 public Base64Util() { 34 } 35 36 public static String encode(byte[] from) { 37 StringBuilder to = new StringBuilder((int) ((double) from.length * 1.34D) + 3); 38 int num = 0; 39 char currentByte = 0; 40 41 int i; 42 for (i = 0; i < from.length; ++i) { 43 for (num %= 8; num < 8; num += 6) { 44 switch (num) { 45 case 0: 46 currentByte = (char) (from[i] & lead6byte); 47 currentByte = (char) (currentByte >>> 2); 48 case 1: 49 case 3: 50 case 5: 51 default: 52 break; 53 case 2: 54 currentByte = (char) (from[i] & last6byte); 55 break; 56 case 4: 57 currentByte = (char) (from[i] & last4byte); 58 currentByte = (char) (currentByte << 2); 59 if (i + 1 < from.length) { 60 currentByte = (char) (currentByte | (from[i + 1] & lead2byte) >>> 6); 61 } 62 break; 63 case 6: 64 currentByte = (char) (from[i] & last2byte); 65 currentByte = (char) (currentByte << 4); 66 if (i + 1 < from.length) { 67 currentByte = (char) (currentByte | (from[i + 1] & lead4byte) >>> 4); 68 } 69 } 70 71 to.append(encodeTable[currentByte]); 72 } 73 } 74 75 if (to.length() % 4 != 0) { 76 for (i = 4 - to.length() % 4; i > 0; --i) { 77 to.append("="); 78 } 79 } 80 81 return to.toString(); 82 } 83 84 public static byte[] decode(final String s) { 85 final int delta = s.endsWith("==") ? 2 : (s.endsWith("=") ? 1 : 0); 86 final byte[] buffer = new byte[s.length() * 3 / 4 - delta]; 87 final int mask = 255; 88 int index = 0; 89 for (int i = 0; i < s.length(); i += 4) { 90 final int c0 = Base64Util.toInt[s.charAt(i)]; 91 final int c2 = Base64Util.toInt[s.charAt(i + 1)]; 92 buffer[index++] = (byte)((c0 << 2 | c2 >> 4) & mask); 93 if (index >= buffer.length) { 94 return buffer; 95 } 96 final int c3 = Base64Util.toInt[s.charAt(i + 2)]; 97 buffer[index++] = (byte)((c2 << 4 | c3 >> 2) & mask); 98 if (index >= buffer.length) { 99 return buffer; 100 } 101 final int c4 = Base64Util.toInt[s.charAt(i + 3)]; 102 buffer[index++] = (byte)((c3 << 6 | c4) & mask); 103 } 104 return buffer; 105 } 106 }
HttpUtil.java
1 package com.wulss.utils; 2 3 4 import java.io.BufferedReader; 5 import java.io.DataOutputStream; 6 import java.io.InputStreamReader; 7 import java.net.HttpURLConnection; 8 import java.net.URL; 9 import java.util.List; 10 import java.util.Map; 11 12 /** 13 * http 工具类 14 */ 15 public class HttpUtil { 16 17 public static String post(String requestUrl, String accessToken, String params) 18 throws Exception { 19 String contentType = "application/x-www-form-urlencoded"; 20 return HttpUtil.post(requestUrl, accessToken, contentType, params); 21 } 22 23 public static String post(String requestUrl, String accessToken, String contentType, String params) 24 throws Exception { 25 String encoding = "UTF-8"; 26 if (requestUrl.contains("nlp")) { 27 encoding = "GBK"; 28 } 29 return HttpUtil.post(requestUrl, accessToken, contentType, params, encoding); 30 } 31 32 public static String post(String requestUrl, String accessToken, String contentType, String params, String encoding) 33 throws Exception { 34 String url = requestUrl + "?access_token=" + accessToken; 35 return HttpUtil.postGeneralUrl(url, contentType, params, encoding); 36 } 37 38 public static String postGeneralUrl(String generalUrl, String contentType, String params, String encoding) 39 throws Exception { 40 URL url = new URL(generalUrl); 41 // 打开和URL之间的连接 42 HttpURLConnection connection = (HttpURLConnection) url.openConnection(); 43 connection.setRequestMethod("POST"); 44 // 设置通用的请求属性 45 connection.setRequestProperty("Content-Type", contentType); 46 connection.setRequestProperty("Connection", "Keep-Alive"); 47 connection.setUseCaches(false); 48 connection.setDoOutput(true); 49 connection.setDoInput(true); 50 51 // 得到请求的输出流对象 52 DataOutputStream out = new DataOutputStream(connection.getOutputStream()); 53 out.write(params.getBytes(encoding)); 54 out.flush(); 55 out.close(); 56 57 // 建立实际的连接 58 connection.connect(); 59 // 获取所有响应头字段 60 Map<String, List<String>> headers = connection.getHeaderFields(); 61 // 遍历所有的响应头字段 62 for (String key : headers.keySet()) { 63 System.err.println(key + "--->" + headers.get(key)); 64 } 65 // 定义 BufferedReader输入流来读取URL的响应 66 BufferedReader in = null; 67 in = new BufferedReader( 68 new InputStreamReader(connection.getInputStream(), encoding)); 69 String result = ""; 70 String getLine; 71 while ((getLine = in.readLine()) != null) { 72 result += getLine; 73 } 74 in.close(); 75 System.err.println("result:" + result); 76 return result; 77 } 78 }
以上是关于java通过百度AI开发平台提取身份证图片中的文字信息的主要内容，如果未能解决你的问题，请参考以下文章