該demo時是基于springboot的人臉識別和后期判斷的小項目，利用的是第三發的虹軟人臉識別的SDK。
下面主要展示頁面功能：照相或照片的方式解析圖片，提取特征；照相或照片的方式對比圖片，判斷相似度；


在引入相關的依賴后和輸入想過的appid和sdk key后，我們主要對調用sdk識別的代碼進行分析
1、JS調用攝像頭拍照，或從本地文件上傳
將照片或者照片文件轉化為64位編碼，向后端傳送

function getMedia() {$("#mainDiv").empty();let videoComp = " <video id='video' width='500px' height='500px' autoplay='autoplay' style='margin-top: 20px'></video><canvas id='canvas' width='500px' height='500px' style='display: none'></canvas>";$("#mainDiv").append(videoComp);let constraints = {video: {width: 500, height: 500},audio: true};//獲得video攝像頭區域let video = document.getElementById("video");//這里介紹新的方法，返回一個 Promise對象// 這個Promise對象返回成功后的回調函數帶一個 MediaStream 對象作為其參數// then()是Promise對象里的方法// then()方法是異步執行，當then()前的方法執行完后再執行then()內部的程序// 避免數據沒有獲取到let promise = navigator.mediaDevices.getUserMedia(constraints);promise.then(function (MediaStream) {video.srcObject = MediaStream;video.play();});// var t1 = window.setTimeout(function() {// takePhoto();// },2000)} //拍照事件function takePhoto() {let mainComp = $("#mainDiv");if(mainComp.has('video').length){let userNameInput = $("#userName").val();if(userNameInput == ""){alert("姓名不能為空!");return false;}//獲得Canvas對象let video = document.getElementById("video");let canvas = document.getElementById("canvas");let ctx = canvas.getContext('2d');ctx.drawImage(video, 0, 0, 500, 500);var formData = new FormData();var base64File = canvas.toDataURL();var userName = $("#userName").val();formData.append("file", base64File);formData.append("name", userName);formData.append("groupId", "101");$.ajax({type: "post",url: "/faceAdd",data: formData,contentType: false,processData: false,async: false,success: function (text) {var res = JSON.stringify(text)if (text.code == 0) {alert("注冊成功")} else {alert(text.message)}},error: function (error) {alert(JSON.stringify(error))}});}else{var formData = new FormData();let userName = $("#userName").val();formData.append("groupId", "101");var file = $("#file0")[0].files[0];var reader = new FileReader();reader.readAsDataURL(file);reader.onload = function () {var base64 = reader.result;formData.append("file", base64);formData.append("name",userName);$.ajax({type: "post",url: "/faceAdd",data: formData,contentType: false,processData: false,async: false,success: function (text) {var res = JSON.stringify(text)if (text.code == 0) {alert("注冊成功")} else {alert(text.message)}},error: function (error) {alert(JSON.stringify(error))}});location.reload();}}}

2、后端解析圖片，提取人像特征
人像特征的提取主要是靠FaceEngine引擎
人臉添加

/*人臉添加*/@RequestMapping(value = "/faceAdd", method = RequestMethod.POST)@ResponseBodypublic Result<Object> faceAdd(@RequestParam("file") String file, @RequestParam("groupId") Integer groupId, @RequestParam("name") String name) {try {if (file == null) {return Results.newFailedResult("file is null");}if (groupId == null) {return Results.newFailedResult("groupId is null");}if (name == null) {return Results.newFailedResult("name is null");}byte[] decode = Base64.decode(base64Process(file));ImageInfo imageInfo = ImageFactory.getRGBData(decode);//人臉特征獲取byte[] bytes = faceEngineService.extractFaceFeature(imageInfo);if (bytes == null) {return Results.newFailedResult(ErrorCodeEnum.NO_FACE_DETECTED);}UserFaceInfo userFaceInfo = new UserFaceInfo();userFaceInfo.setName(name);userFaceInfo.setGroupId(groupId);userFaceInfo.setFaceFeature(bytes);userFaceInfo.setFaceId(RandomUtil.randomString(10));//人臉特征插入到數據庫userFaceInfoService.insertSelective(userFaceInfo);logger.info("faceAdd:" + name);return Results.newSuccessResult("");} catch (Exception e) {logger.error("", e);}return Results.newFailedResult(ErrorCodeEnum.UNKNOWN);}

在人臉添加的控制層，用到了service層邏輯處理，大致看注釋就能明白

@Overridepublic byte[] extractFaceFeature(ImageInfo imageInfo) throws InterruptedException {FaceEngine faceEngine = null;try {//獲取引擎對象faceEngine = faceEngineObjectPool.borrowObject();//人臉檢測得到人臉列表List<FaceInfo> faceInfoList = new ArrayList<FaceInfo>();//人臉檢測int i = faceEngine.detectFaces(imageInfo.getImageData(), imageInfo.getWidth(), imageInfo.getHeight(), imageInfo.getImageFormat(), faceInfoList);if (CollectionUtil.isNotEmpty(faceInfoList)) {FaceFeature faceFeature = new FaceFeature();//提取人臉特征faceEngine.extractFaceFeature(imageInfo.getImageData(), imageInfo.getWidth(), imageInfo.getHeight(), imageInfo.getImageFormat(), faceInfoList.get(0), faceFeature);return faceFeature.getFeatureData();}} catch (Exception e) {logger.error("", e);} finally {if (faceEngine != null) {//釋放引擎對象faceEngineObjectPool.returnObject(faceEngine);}}return null;}

然后將faceFeature存入數據庫即可。

下面是人臉對比分析，性別年齡相似度分析
先看controller層

/*人臉識別*/@RequestMapping(value = "/faceSearch", method = RequestMethod.POST)@ResponseBodypublic Result<FaceSearchResDto> faceSearch(String file, Integer groupId) throws Exception {if (groupId == null) {return Results.newFailedResult("groupId is null");}byte[] decode = Base64.decode(base64Process(file));BufferedImage bufImage = ImageIO.read(new ByteArrayInputStream(decode));ImageInfo imageInfo = ImageFactory.bufferedImage2ImageInfo(bufImage);//人臉特征獲取byte[] bytes = faceEngineService.extractFaceFeature(imageInfo);if (bytes == null) {return Results.newFailedResult(ErrorCodeEnum.NO_FACE_DETECTED);}//人臉比對，獲取比對結果List<FaceUserInfo> userFaceInfoList = faceEngineService.compareFaceFeature(bytes, groupId);if (CollectionUtil.isNotEmpty(userFaceInfoList)) {FaceUserInfo faceUserInfo = userFaceInfoList.get(0);FaceSearchResDto faceSearchResDto = new FaceSearchResDto();BeanUtil.copyProperties(faceUserInfo, faceSearchResDto);List<ProcessInfo> processInfoList = faceEngineService.process(imageInfo);if (CollectionUtil.isNotEmpty(processInfoList)) {//人臉檢測List<FaceInfo> faceInfoList = faceEngineService.detectFaces(imageInfo);int left = faceInfoList.get(0).getRect().getLeft();int top = faceInfoList.get(0).getRect().getTop();int width = faceInfoList.get(0).getRect().getRight() - left;int height = faceInfoList.get(0).getRect().getBottom() - top;Graphics2D graphics2D = bufImage.createGraphics();graphics2D.setColor(Color.RED);//紅色BasicStroke stroke = new BasicStroke(5f);graphics2D.setStroke(stroke);graphics2D.drawRect(left, top, width, height);ByteArrayOutputStream outputStream = new ByteArrayOutputStream();System.out.println("111");System.out.println(bufImage.getType());ImageIO.write(bufImage, "gif", outputStream);System.out.println("111");byte[] bytes1 = outputStream.toByteArray();faceSearchResDto.setImage("data:image/jpeg;base64," + Base64Utils.encodeToString(bytes1));faceSearchResDto.setAge(processInfoList.get(0).getAge());faceSearchResDto.setGender(processInfoList.get(0).getGender().equals(1) ? "女" : "男");}return Results.newSuccessResult(faceSearchResDto);}return Results.newFailedResult(ErrorCodeEnum.FACE_DOES_NOT_MATCH);}

再看service層

@Overridepublic List<FaceUserInfo> compareFaceFeature(byte[] faceFeature, Integer groupId) throws InterruptedException, ExecutionException {List<FaceUserInfo> resultFaceInfoList = Lists.newLinkedList();//識別到的人臉列表FaceFeature targetFaceFeature = new FaceFeature();targetFaceFeature.setFeatureData(faceFeature);List<FaceUserInfo> faceInfoList = userFaceInfoMapper.getUserFaceInfoByGroupId(groupId); //從數據庫中取出人臉庫List<List<FaceUserInfo>> faceUserInfoPartList = Lists.partition(faceInfoList, 1000);//分成1000一組，多線程處理CompletionService<List<FaceUserInfo>> completionService = new ExecutorCompletionService(executorService);for (List<FaceUserInfo> part : faceUserInfoPartList) {completionService.submit(new CompareFaceTask(part, targetFaceFeature));}for (int i = 0; i < faceUserInfoPartList.size(); i++) {List<FaceUserInfo> faceUserInfoList = completionService.take().get();if (CollectionUtil.isNotEmpty(faceInfoList)) {resultFaceInfoList.addAll(faceUserInfoList);}}resultFaceInfoList.sort((h1, h2) -> h2.getSimilarValue().compareTo(h1.getSimilarValue()));//從大到小排序return resultFaceInfoList;}private class CompareFaceTask implements Callable<List<FaceUserInfo>> {private List<FaceUserInfo> faceUserInfoList;private FaceFeature targetFaceFeature;public CompareFaceTask(List<FaceUserInfo> faceUserInfoList, FaceFeature targetFaceFeature) {this.faceUserInfoList = faceUserInfoList;this.targetFaceFeature = targetFaceFeature;}@Overridepublic List<FaceUserInfo> call() throws Exception {FaceEngine faceEngine = null;List<FaceUserInfo> resultFaceInfoList = Lists.newLinkedList();//識別到的人臉列表try {faceEngine = faceEngineObjectPool.borrowObject();for (FaceUserInfo faceUserInfo : faceUserInfoList) {FaceFeature sourceFaceFeature = new FaceFeature();sourceFaceFeature.setFeatureData(faceUserInfo.getFaceFeature());FaceSimilar faceSimilar = new FaceSimilar();faceEngine.compareFaceFeature(targetFaceFeature, sourceFaceFeature, faceSimilar);Integer similarValue = plusHundred(faceSimilar.getScore());//獲取相似值if (similarValue > passRate) {//相似值大于配置預期，加入到識別到人臉的列表FaceUserInfo info = new FaceUserInfo();info.setName(faceUserInfo.getName());info.setFaceId(faceUserInfo.getFaceId());info.setSimilarValue(similarValue);resultFaceInfoList.add(info);}}} catch (Exception e) {logger.error("", e);} finally {if (faceEngine != null) {faceEngineObjectPool.returnObject(faceEngine);}}return resultFaceInfoList;}

大致流程

總結

以上是生活随笔為你收集整理的人脸识别SDK调用与分析的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。