JAVA實現人臉識別，活體檢測之百度API
阿新 • • 發佈：2019-01-27
-----------------------8/16更新---------------------------
有人問原始碼:
連結：https://pan.baidu.com/s/1WP37IBacu6VZwtNOUDbSYg 密碼：vdzm
在這
--------------------------------------------------
然後是官方的使用說明文件：地址；
以上，感謝。
首先引入jar：
maven:
<dependency> <groupId>com.baidu.aip</groupId> <artifactId>java-sdk</artifactId> <version>4.3.2</version> </dependency>
然後獲取所需的AppID、APIKey、SecretKey：
進入百度雲，建立應用：
建立完成之後可看見如下：
獲取AppID、APIKey、SecretKey後，匯入一下程式碼：
public class FaceSpot { private static final String AppID = ""; private static final String APIKey = ""; private static final String SecretKey = ""; static AipFace client = null; static { client = new AipFace(AppID, APIKey, SecretKey); // 可選：設定網路連線引數 client.setConnectionTimeoutInMillis(2000); client.setSocketTimeoutInMillis(60000); } public static void main(String[] args) throws IOException { String filePath = "F:/3.jpg"; byte[] imgData = FileToByte(new File(filePath)); System.out.println(detectFace(imgData,"1")); // String filePath1 = "F:/3.jpg"; // String filePath2 = "F:/7.jpg"; // byte[] imgData1 = FileUtil.readFileByBytes(filePath1); // byte[] imgData2 = FileUtil.readFileByBytes(filePath2); // System.out.println(faceverify(imgData1)); } /** * 人臉檢測 * * @return * @throws IOException */ public static String detectFace(File file, String max_face_num) { try { return detectFace(FileToByte(file), max_face_num); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } return null; } /** * 人臉檢測 * * @return * @throws IOException */ public static String detectFace(byte[] arg0, String max_face_num) { try { HashMap<String, String> options = new HashMap<String, String>(); options.put("face_field", "age,beauty,expression,faceshape,gender,glasses,race,qualities"); options.put("max_face_num", "2"); options.put("face_type", "LIVE"); // 圖片資料 String imgStr = Base64Util.encode(arg0); String imageType = "BASE64"; JSONObject res = client.detect(imgStr, imageType, options); System.out.println(res.toString(2)); return res.toString(); } catch (Exception e) { e.printStackTrace(); } return null; } /** * 人臉比對 * @param file1 * @param file2 * @return */ public static String matchFace(File file1, File file2) { try { return matchFace(FileToByte(file1), FileToByte(file2)); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } return null; } /** * 人臉比對 * * @param arg0 * 人臉1 * @param arg1 * 人臉2 * @return */ public static String matchFace(byte[] arg0, byte[] arg1) { String imgStr1 = Base64Util.encode(arg0); String imgStr2 = Base64Util.encode(arg1); MatchRequest req1 = new MatchRequest(imgStr1, "BASE64"); MatchRequest req2 = new MatchRequest(imgStr2, "BASE64"); ArrayList<MatchRequest> requests = new ArrayList<MatchRequest>(); requests.add(req1); requests.add(req2); JSONObject res = client.match(requests); return res.toString(); } /** * 人臉搜尋 * @param file * @param groupIdList * @param userId * @return */ public static String searchFace(File file, String groupIdList, String userId) { try { return searchFace(FileToByte(file), groupIdList, userId); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } return null; } /** * 人臉搜尋 * * @param arg0 * @param groupIdList * @return */ public static String searchFace(byte[] arg0, String groupIdList, String userId) { String imgStr = Base64Util.encode(arg0); String imageType = "BASE64"; HashMap<String, String> options = new HashMap<String, String>(); options.put("quality_control", "NORMAL"); options.put("liveness_control", "LOW"); if (userId != null) { options.put("user_id", userId); } options.put("max_user_num", "1"); JSONObject res = client.search(imgStr, imageType, groupIdList, options); return res.toString(2); } /** * 增加使用者 * @param file * @param userInfo * @param userId * @param groupId * @return */ public static String addUser(File file, String userInfo, String userId, String groupId) { try { return addUser(FileToByte(file), userInfo, userId, groupId); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } return null; } /** * 增加使用者 * * @param arg0 * @param userInfo * @param userId * @param groupId * @return */ public static String addUser(byte[] arg0, String userInfo, String userId, String groupId) { String imgStr = Base64Util.encode(arg0); String imageType = "BASE64"; HashMap<String, String> options = new HashMap<String, String>(); options.put("user_info", userInfo); options.put("quality_control", "NORMAL"); options.put("liveness_control", "LOW"); JSONObject res = client.addUser(imgStr, imageType, groupId, userId, options); return res.toString(2); } public static String updateUser(File file, String userInfo, String userId, String groupId) { try { return updateUser(FileToByte(file), userInfo, userId, groupId); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } return null; } /** * 更新使用者 * * @param arg0 * @param userInfo * @param userId * @param groupId * @return */ public static String updateUser(byte[] arg0, String userInfo, String userId, String groupId) { String imgStr = Base64Util.encode(arg0); String imageType = "BASE64"; HashMap<String, String> options = new HashMap<String, String>(); if (userInfo != null) { options.put("user_info", userInfo); } options.put("quality_control", "NORMAL"); options.put("liveness_control", "LOW"); JSONObject res = client.updateUser(imgStr, imageType, groupId, userId, options); return res.toString(2); } /** * 刪除使用者人臉資訊 * @param userId * @param groupId * @param faceToken * @return */ public static String deleteUserFace(String userId, String groupId, String faceToken) { HashMap<String, String> options = new HashMap<String, String>(); // 人臉刪除 JSONObject res = client.faceDelete(userId, groupId, faceToken, options); return res.toString(); } /** * 查詢使用者資訊 * @param userId * @param groupId * @return */ public static String searchUserInfo(String userId, String groupId) { HashMap<String, String> options = new HashMap<String, String>(); // 使用者資訊查詢 JSONObject res = client.getUser(userId, groupId, options); return res.toString(2); } /** * 獲取使用者人臉列表 * @param userId * @param groupId * @return */ public static String getUserFaceList(String userId, String groupId) { HashMap<String, String> options = new HashMap<String, String>(); // 獲取使用者人臉列表 JSONObject res = client.faceGetlist(userId, groupId, options); return res.toString(2); } /** * 獲取一組使用者 * @param groupId * @param returnNum * @return */ public static String getGroupUsers(String groupId, String returnNum) { HashMap<String, String> options = new HashMap<String, String>(); options.put("start", "0"); if (returnNum != null) { options.put("length", returnNum); } // 獲取使用者列表 JSONObject res = client.getGroupUsers(groupId, options); return res.toString(2); } /** * 組使用者複製 * @param userId * @param srcGroupId * @param dstGroupId * @return */ public static String userCopy(String userId, String srcGroupId, String dstGroupId) { HashMap<String, String> options = new HashMap<String, String>(); options.put("src_group_id", srcGroupId); options.put("dst_group_id", dstGroupId); // 複製使用者 JSONObject res = client.userCopy(userId, options); return res.toString(2); } /** * 刪除使用者 * @param userId * @param groupId * @return */ public static String deleteUser(String userId, String groupId) { HashMap<String, String> options = new HashMap<String, String>(); // 人臉刪除 JSONObject res = client.deleteUser(groupId, userId, options); return res.toString(); } /** * 增加組資訊 * @param groupId * @return */ public static String addGroup(String groupId) { HashMap<String, String> options = new HashMap<String, String>(); // 建立使用者組 JSONObject res = client.groupAdd(groupId, options); return res.toString(); } /** * 刪除 * @param groupId * @return */ public static String deleteGroup(String groupId) { HashMap<String, String> options = new HashMap<String, String>(); // 建立使用者組 JSONObject res = client.groupDelete(groupId, options); return res.toString(); } /** * 獲取組列表 * @param length * @return */ public static String getGroupList(String length) { HashMap<String, String> options = new HashMap<String, String>(); options.put("start", "0"); options.put("length", length); // 組列表查詢 JSONObject res = client.getGroupList(options); return res.toString(); } /** * 活體檢測 * @param arg0 * @return */ public static String faceverify(byte[] arg0){ String imgStr = Base64Util.encode(arg0); String imageType = "BASE64"; FaceVerifyRequest req = new FaceVerifyRequest(imgStr, imageType); ArrayList<FaceVerifyRequest> list = new ArrayList<FaceVerifyRequest>(); list.add(req); JSONObject res = client.faceverify(list); return res.toString(); } private static byte[] FileToByte(File file) throws IOException { // 將資料轉為流 InputStream content = new FileInputStream(file); ByteArrayOutputStream swapStream = new ByteArrayOutputStream(); byte[] buff = new byte[100]; int rc = 0; while ((rc = content.read(buff, 0, 100)) > 0) { swapStream.write(buff, 0, rc); } // 獲得二進位制陣列 return swapStream.toByteArray(); } }
填入自己的AppID, APIKey, SecretKey；
函式的引數可自行根據官方文件自己改一改，檔案讀取那邊提供一下小帥的FileUtil並且處理好了異常：
public class FileUtil { /** * 根據檔案路徑讀取byte[] 陣列 */ public static byte[] readFileByBytes(String filePath) throws IOException { File file = new File(filePath); if (!file.exists()) { throw new FileNotFoundException(filePath); } else { ByteArrayOutputStream bos = new ByteArrayOutputStream((int) file.length()); BufferedInputStream in = null; try { in = new BufferedInputStream(new FileInputStream(file)); short bufSize = 1024; byte[] buffer = new byte[bufSize]; int len1; while (-1 != (len1 = in.read(buffer, 0, bufSize))) { bos.write(buffer, 0, len1); } byte[] var7 = bos.toByteArray(); return var7; } finally { try { if (in != null) { in.close(); } } catch (IOException var14) { var14.printStackTrace(); } bos.close(); } } } }
然後就是測試啦，測試的話，隨便百度了一張美女的照片，然後看返回：
{
"timestamp": 1528422299,
"result": {
"face_list": [{
"expression": {
"probability": 0.9997190833,
"type": "smile"
},
"face_probability": 1,
"glasses": {
"probability": 0.9999995232,
"type": "none"
},
"location": {
"height": 102,
"rotation": -2,
"width": 109,
"left": 102.9398575,
"top": 56.39219284
},
"age": 19,
"gender": {
"probability": 0.9999992847,
"type": "female"
},
"face_shape": {
"probability": 0.7517765164,
"type": "heart"
},
"face_token": "706eaee3240ef0cc679ab209b1b71e0d",
"beauty": 69.47167206,
"race": {
"probability": 0.9999864101,
"type": "yellow"
},
"angle": {
"yaw": -2.858289719,
"roll": -2.302281618,
"pitch": 9.867022514
}
}],
"face_num": 1
},
"cached": 0,
"error_code": 0,
"log_id": 3671200838,
"error_msg": "SUCCESS"
}嗯，確實是美女，
"beauty": 69.47167206好，然後通過瀏覽器開啟攝像頭來實時的試下，先上效果圖：
好，大家不用在意這些細節，比如顏值啥的。。
我覺得顏值這項可能是假的，因為我做搞怪的表情時分數反而高一些？？？？？？？？？？
上程式碼吧。。
首先是Controller：
@Controller
@RequestMapping(value = "/faceRecognition")
public class faceRecognitionController {
/**
* 人臉檢測測試頁面
* @return
* @throws Exception
*/
@RequestMapping(value = "/test.do")
public ModelAndView queryVoi() throws Exception {
ModelAndView modelAndView = new ModelAndView();
modelAndView.setViewName("/artificialIntelligence/faceRecognition/test");
return modelAndView;
}
/**
* 請求人臉檢測
* @return
* @throws Exception
*/
@RequestMapping(value = "/save.do")
@ResponseBody
public Map<String, Object> queryService(@RequestParam("the_file") MultipartFile file) {
Map<String, Object> modelMap = new HashMap<String, Object>();
try {
//將資料轉為流
InputStream content = file.getInputStream();
ByteArrayOutputStream swapStream = new ByteArrayOutputStream();
byte[] buff = new byte[100];
int rc = 0;
while ((rc = content.read(buff, 0, 100)) > 0) {
swapStream.write(buff, 0, rc);
}
//獲得二進位制陣列
byte[] in2b = swapStream.toByteArray();
//呼叫人臉檢測的方法
String str = FaceSpot.detectFace(in2b,""+1);
JSONObject job = new JSONObject(FaceSpot.faceverify(in2b));
System.out.println(job.toString());
JSONObject testData = job.getJSONObject("result");
//System.out.println(testData.get("face_liveness"));
JSON json = JSON.parseObject(str);
FaceV3DetectBean bean = JSON.toJavaObject(json, FaceV3DetectBean.class);
JSONArray arr = new JSONArray();
for(int i=0;i<bean.getResult().getFace_list().size();i++){
JSONObject jsonObject = new JSONObject();
//獲取年齡
int ageOne = bean.getResult().getFace_list().get(i).getAge();
//處理年齡
String age =String.valueOf(new BigDecimal(ageOne).setScale(0, BigDecimal.ROUND_HALF_UP));
jsonObject.put("age", age);
//獲取美醜打分
Double beautyOne = (Double) bean.getResult().getFace_list().get(i).getBeauty();
//處理美醜打分
String beauty =String.valueOf(new BigDecimal(beautyOne).setScale(0, BigDecimal.ROUND_HALF_UP));
jsonObject.put("beauty", beauty);
//獲取性別 male(男)、female(女)
String gender = (String) bean.getResult().getFace_list().get(i).getGender().getType();
jsonObject.put("gender", gender);
//獲取是否帶眼睛 0-無眼鏡，1-普通眼鏡，2-墨鏡
String glasses = bean.getResult().getFace_list().get(i).getGlasses().getType();
jsonObject.put("glasses", String.valueOf(glasses));
//獲取是否微笑，0，不笑；1，微笑；2，大笑
String expression = bean.getResult().getFace_list().get(i).getExpression().getType();
jsonObject.put("expression", String.valueOf(expression));
arr.put(jsonObject);
}
modelMap.put("strjson", arr.toString());
modelMap.put("face_liveness", testData.get("face_liveness"));
modelMap.put("success", true);
} catch (Exception e) {
e.printStackTrace();
modelMap.put("success", false);
modelMap.put("data", e.getMessage());
}
return modelMap;
}
}用的是fastjson，自行加jar包。
然後是jsp：
<%@ page language="java" contentType="text/html; charset=UTF-8" pageEncoding="UTF-8"%>
<%@ taglib prefix="c" uri="http://java.sun.com/jsp/jstl/core"%>
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1">
<title>人臉檢測</title>
<script src="js/jquery-2.1.1.js" type="text/javascript" charset="utf-8"></script>
<script>
//判斷瀏覽器是否支援HTML5 Canvas
window.onload = function () {
try {
//動態建立一個canvas元 ，並獲取他2Dcontext。如果出現異常則表示不支援 document.createElement("canvas").getContext("2d");
document.getElementById("support").innerHTML = "瀏覽器支援HTML5 CANVAS";
}
catch (e) {
document.getElementByIdx("support").innerHTML = "瀏覽器不支援HTML5 CANVAS";
}
};
//這段代 主要是獲取攝像頭的視訊流並顯示在Video 籤中
window.addEventListener("DOMContentLoaded", function () {
var canvas = document.getElementById("canvas"),
context = canvas.getContext("2d"),
video = document.getElementById("video"),
videoObj = { "video": true },
errBack = function (error) {
console.log("Video capture error: ", error.code);
};
//拍照按鈕
// $("#snap").click(function () {
// context.drawImage(video, 0, 0, 330, 250);
// })
//拍照每秒一次
setInterval(function(){
context.drawImage(video, 0, 0, 330, 250)
CatchCode();
},1000);
//navigator.getUserMedia這個寫法在Opera中好像是navigator.getUserMedianow
//更新相容火狐瀏覽器
if (navigator.getUserMedia || navigator.webkitGetUserMedia || navigator.mozGetUserMedia) {
navigator.getUserMedia=navigator.getUserMedia || navigator.webkitGetUserMedia || navigator.mozGetUserMedia;
navigator.getUserMedia(videoObj, function (stream) {
video.srcObject = stream;
video.play();
}, errBack);
}
}, false);
function dataURItoBlob(base64Data) {
var byteString;
if (base64Data.split(',')[0].indexOf('base64') >= 0)
byteString = atob(base64Data.split(',')[1]);
else
byteString = unescape(base64Data.split(',')[1]);
var mimeString = base64Data.split(',')[0].split(':')[1].split(';')[0];
var ia = new Uint8Array(byteString.length);
for (var i = 0; i < byteString.length; i++) {
ia[i] = byteString.charCodeAt(i);
}
return new Blob([ia], {type:mimeString});
}
//上傳伺服器
function CatchCode() {
var canvans = document.getElementById("canvas");
//獲取瀏覽器頁面的畫布物件
//以下開始編 資料
var imageBase64 = canvans.toDataURL();
var blob = dataURItoBlob(imageBase64); // 上一步中的函式
var fd = new FormData(document.forms[0]);
fd.append("the_file", blob, 'image.png');
//將影象轉換為base64資料
$.ajax({
type:"POST",
url:"faceRecognition/save.do",
processData: false, // 必須
contentType: false, // 必須
data:fd,
datatype: "json",
success:function(data){
var mes = eval(data);
//alert(mes.success);
//var jsonObj = $.parseJSON(mes.strjson);
//alert(jsonObj[0].age);
if (mes.success) {
//alert(mes.strjson);
var jsonObj = $.parseJSON(mes.strjson);
//alert(jsonObj);
var age = jsonObj[0].age;
var beauty = jsonObj[0].beauty;
var gendergender = jsonObj[0].gender;
var glasses = jsonObj[0].glasses;
var expression = jsonObj[0].expression
$("#age").html(age);
$("#beauty").html(beauty);
$("#faceverify").html(mes.face_liveness);
if(gendergender == 'male'){
$("#gendergender").html("男");
}else{
$("#gendergender").html("女");
}
if(glasses == 'none'){
$("#glasses").html("未戴眼鏡");
}else if(glasses == 'common'){
$("#glasses").html("戴了普通眼鏡");
}else{
$("#glasses").html("戴了墨鏡");
}
if(expression == 'none'){
$("#expression").html("不笑");
}else if(expression == 'smile'){
$("#expression").html("微笑");
}else{
$("#expression").html("大笑");
}
}
},
error: function(){
//請求出錯處理
alert("出情況了");
}
});
}
</script>
<style>
.div-a{ float:left;width:60%;height:60%;border:1px solid #F00}
.div-b{ float:left;width:39%;height:60%;border:1px solid #000}
span{ font-size:25px }
</style>
</head>
<body>
<!-- 左邊區域 -->
<div class="div-a" id="contentHolder">
<video id="video" width="100%" height="60%" autoplay></video>
<canvas style="" hidden="hidden" id="canvas" width="520" height="250"></canvas>
</div>
<!-- 右邊區域 -->
<div class="div-b" >
<!-- 測試按鈕 -->
<!-- <input type="button" id="snap" style="width:100px;height:35px;" value="拍 照" /> -->
<!-- <input type="button" onclick="CatchCode();" style="width:100px;height:35px;" value="上傳伺服器" /> -->
<h1>人臉檢測實時資料</h1>
<span>年齡：</span><span id="age"></span><br/>
<span>顏值：</span><span id="beauty" ></span><br/>
<span>性別：</span><span id="gendergender"></span><br/>
<span>是否戴眼鏡:</span><span id="glasses"></span><br/>
<span>表情：</span><span id="expression"></span><br/>
<span>活體分數：</span><span id="faceverify"></span><br/>
</div>
</body>
</html>這個貌似只支援火狐？？我用其他瀏覽器攝像頭並未開啟，有哪位大神知道程式碼還請告訴在下，謝謝指教。
這邊有個活體指數，我測試了下，貌似還挺準，我首先百度了一張胡歌的照片放在攝像頭上，emmmm.....顏值80，但是活體指數就很低，然後本人上，活體指數一直在0.95之上浮動，還是可靠的。
好，告辭！