人脸任务在计算机视觉领域中十分重要，本项目主要使用了两类技术：人脸检测+人脸识别。
代码分为两部分内容：人脸注册 和 人脸识别
- 人脸注册：将人脸特征存储进数据库，这里用feature.csv代替
- 人脸识别：将人脸特征与CSV文件中人脸特征进行比较，如果成功匹配则写入考勤文件attendance.csv
文章前半部分为一步步实现流程介绍，最后会有整理过后的完整项目代码。
一、项目实现
A. 注册：
导入相关包
import cv2
import numpy as np
import dlib
import time
import csv
# from argparse import ArgumentParser
from PIL import Image, ImageDraw, ImageFont设计注册功能
注册过程我们需要完成的事：
- 打开摄像头获取画面图片
- 在图片中检测并获取人脸位置
- 根据人脸位置获取68个关键点
- 根据68个关键点生成特征描述符
- 保存
- (优化)展示界面，加入注册时成功提示等
1、基本步骤
我们首先进行前三步：
# 检测人脸，获取68个关键点，获取特征描述符
def faceRegister(faceId=1, userName='default', interval=3, faceCount=3, resize_w=700, resize_h=400):'''faceId:人脸IDuserName: 人脸姓名faceCount: 采集该人脸图片的数量interval: 采集间隔'''cap = cv2.VideoCapture(0)# 人脸检测模型hog_face_detector = dlib.get_frontal_face_detector()# 关键点 检测模型shape_detector = dlib.shape_predictor('./weights/shape_predictor_68_face_landmarks.dat')# resnet模型face_descriptor_extractor = dlib.face_recognition_model_v1('./weights/dlib_face_recognition_resnet_model_v1.dat')while True:ret, frame = cap.read()# 镜像frame = cv2.flip(frame,1)# 转为灰度图frame_gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)# 检测人脸detections = hog_face_detector(frame,1)for face in detections:# 人脸框坐标 左上和右下l, t, r, b = face.left(), face.top(), face.right(), face.bottom()# 获取68个关键点points = shape_detector(frame,face)# 绘制关键点for point in points.parts():cv2.circle(frame,(point.x,point.y),2,(0,255,0),1)# 绘制矩形框cv2.rectangle(frame,(l,t),(r,b),(0,255,0),2)cv2.imshow("face",frame)if cv2.waitKey(10) & 0xFF == ord('q'):breakcap.release()cv2.destroyAllWindowsfaceRegister() 此时一张帅脸如下：
2、描述符的采集
之后，我们根据参数，即faceCount 和 Interval 进行描述符的生成和采集。
(这里我默认是faceCount=3，Interval=3，即每3秒采集一次，共3次）
def faceRegister(faceId=1, userName='default', interval=3, faceCount=3, resize_w=700, resize_h=400):'''faceId:人脸IDuserName: 人脸姓名faceCount: 采集该人脸图片的数量interval: 采集间隔'''cap = cv2.VideoCapture(0)# 人脸检测模型hog_face_detector = dlib.get_frontal_face_detector()# 关键点 检测模型shape_detector = dlib.shape_predictor('./weights/shape_predictor_68_face_landmarks.dat')# resnet模型face_descriptor_extractor = dlib.face_recognition_model_v1('./weights/dlib_face_recognition_resnet_model_v1.dat')# 开始时间start_time = time.time()# 执行次数collect_times = 0while True:ret, frame = cap.read()# 镜像frame = cv2.flip(frame,1)# 转为灰度图frame_gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)# 检测人脸detections = hog_face_detector(frame,1)for face in detections:# 人脸框坐标 左上和右下l, t, r, b = face.left(), face.top(), face.right(), face.bottom()# 获取68个关键点points = shape_detector(frame,face)# 绘制人脸关键点for point in points.parts():cv2.circle(frame, (point.x, point.y), 2, (0, 255, 0), 1)# 绘制矩形框cv2.rectangle(frame, (l, t), (r, b), (0, 255, 0), 2)# 采集：if collect_times < faceCount:# 获取当前时间now = time.time()# 时间限制if now - start_time > interval:# 获取特征描述符face_descriptor = face_descriptor_extractor.compute_face_descriptor(frame,points)# dlib格式转为数组face_descriptor = [f for f in face_descriptor]collect_times += 1start_time = nowprint("成功采集{}次".format(collect_times))else:# 时间间隔不到intervalprint("等待进行下一次采集")passelse:# 已经成功采集完3次了print("采集完毕")cap.release()cv2.destroyAllWindows()returncv2.imshow("face",frame)if cv2.waitKey(10) & 0xFF == ord('q'):breakcap.release()cv2.destroyAllWindows()faceRegister() 等待进行下一次采集 ... 成功采集1次 等待进行下一次采集 ... 成功采集2次 等待进行下一次采集 ... 成功采集3次 采集完毕
3、完整的注册
最后就是写入csv文件
这里加入了注册成功等的提示，且把一些变量放到了全局，因为后面人脸识别打卡时也会用到。
# 加载人脸检测器
hog_face_detector = dlib.get_frontal_face_detector()
cnn_detector = dlib.cnn_face_detection_model_v1('./weights/mmod_human_face_detector.dat')
haar_face_detector = cv2.CascadeClassifier('./weights/haarcascade_frontalface_default.xml')# 加载关键点检测器
points_detector = dlib.shape_predictor('./weights/shape_predictor_68_face_landmarks.dat')
# 加载resnet模型
face_descriptor_extractor = dlib.face_recognition_model_v1('./weights/dlib_face_recognition_resnet_model_v1.dat')# 绘制中文
def cv2AddChineseText(img, text, position, textColor=(0, 255, 0), textSize=30):if (isinstance(img, np.ndarray)): # 判断是否OpenCV图片类型img = Image.fromarray(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))# 创建一个可以在给定图像上绘图的对象draw = ImageDraw.Draw(img)# 字体的格式fontStyle = ImageFont.truetype("./fonts/songti.ttc", textSize, encoding="utf-8")# 绘制文本draw.text(position, text, textColor, font=fontStyle)# 转换回OpenCV格式return cv2.cvtColor(np.asarray(img), cv2.COLOR_RGB2BGR)
# 绘制左侧信息
def drawLeftInfo(frame, fpsText, mode="Reg", detector='haar', person=1, count=1):# 帧率cv2.putText(frame, "FPS: " + str(round(fpsText, 2)), (30, 50), cv2.FONT_ITALIC, 0.8, (0, 255, 0), 2)# 模式：注册、识别cv2.putText(frame, "Mode: " + str(mode), (30, 80), cv2.FONT_ITALIC, 0.8, (0, 255, 0), 2)if mode == 'Recog':# 检测器cv2.putText(frame, "Detector: " + detector, (30, 110), cv2.FONT_ITALIC, 0.8, (0, 255, 0), 2)# 人数cv2.putText(frame, "Person: " + str(person), (30, 140), cv2.FONT_ITALIC, 0.8, (0, 255, 0), 2)# 总人数cv2.putText(frame, "Count: " + str(count), (30, 170), cv2.FONT_ITALIC, 0.8, (0, 255, 0), 2)# 注册人脸
def faceRegiser(faceId=1, userName='default', interval=3, faceCount=3, resize_w=700, resize_h=400):# 计数count = 0# 开始注册时间startTime = time.time()# 视频时间frameTime = startTime# 控制显示打卡成功的时长show_time = (startTime - 10)# 打开文件f = open('./data/feature.csv', 'a', newline='')csv_writer = csv.writer(f)cap = cv2.VideoCapture(0)while True:ret, frame = cap.read()frame = cv2.resize(frame, (resize_w, resize_h))frame = cv2.flip(frame, 1)# 检测face_detetion = hog_face_detector(frame, 1)for face in face_detetion:# 识别68个关键点points = points_detector(frame, face)# 绘制人脸关键点for point in points.parts():cv2.circle(frame, (point.x, point.y), 2, (0, 255, 0), 1)# 绘制框框l, t, r, b = face.left(), face.top(), face.right(), face.bottom()cv2.rectangle(frame, (l, t), (r, b), (0, 255, 0), 2)now = time.time()if (now - show_time) < 0.5:frame = cv2AddChineseText(frame,"注册成功 {count}/{faceCount}".format(count=(count + 1), faceCount=faceCount),(l, b + 30), textColor=(255, 0, 255), textSize=30)# 检查次数if count < faceCount:# 检查时间if now - startTime > interval:# 特征描述符face_descriptor = face_descriptor_extractor.compute_face_descriptor(frame, points)face_descriptor = [f for f in face_descriptor]# 描述符增加进data文件line = [faceId, userName, face_descriptor]# 写入csv_writer.writerow(line)# 保存照片样本print('人脸注册成功 {count}/{faceCount}，faceId:{faceId}，userName:{userName}'.format(count=(count + 1),faceCount=faceCount,faceId=faceId,userName=userName))frame = cv2AddChineseText(frame,"注册成功 {count}/{faceCount}".format(count=(count + 1), faceCount=faceCount),(l, b + 30), textColor=(255, 0, 255), textSize=30)show_time = time.time()# 时间重置startTime = now# 次数加一count += 1else:print('人脸注册完毕')f.close()cap.release()cv2.destroyAllWindows()returnnow = time.time()fpsText = 1 / (now - frameTime)frameTime = now# 绘制drawLeftInfo(frame, fpsText, 'Register')cv2.imshow('Face Attendance Demo: Register', frame)if cv2.waitKey(10) & 0xFF == ord('q'):breakf.close()cap.release()cv2.destroyAllWindows()此时执行：
faceRegiser(3,"用户B")
人脸注册成功 1/3，faceId:3，userName:用户B 人脸注册成功 2/3，faceId:3，userName:用户B 人脸注册成功 3/3，faceId:3，userName:用户B 人脸注册完毕
其features文件：
B. 识别、打卡
识别步骤如下：
- 打开摄像头获取画面
- 根据画面中的图片获取里面的人脸特征描述符
- 根据特征描述符将其与feature.csv文件里特征做距离判断
- 获取ID、NAME
- 考勤记录写入attendance.csv里
这里与上面流程相似，不过是加了一个对比功能，距离小于阈值，则表示匹配成功。就加快速度不一步步来了，代码如下：
# 刷新右侧考勤信息
def updateRightInfo(frame, face_info_list, face_img_list):# 重新绘制逻辑：从列表中每隔3个取一批显示，新增人脸放在最前面# 如果有更新，重新绘制# 如果没有，定时往后移动left_x = 30left_y = 20resize_w = 80offset_y = 120index = 0frame_h = frame.shape[0]frame_w = frame.shape[1]for face in face_info_list[:3]:name = face[0]time = face[1]face_img = face_img_list[index]# print(face_img.shape)face_img = cv2.resize(face_img, (resize_w, resize_w))offset_y_value = offset_y * indexframe[(left_y + offset_y_value):(left_y + resize_w + offset_y_value), -(left_x + resize_w):-left_x] = face_imgcv2.putText(frame, name, ((frame_w - (left_x + resize_w)), (left_y + resize_w) + 15 + offset_y_value),cv2.FONT_ITALIC, 0.5, (0, 255, 0), 1)cv2.putText(frame, time, ((frame_w - (left_x + resize_w)), (left_y + resize_w) + 30 + offset_y_value),cv2.FONT_ITALIC, 0.5, (0, 255, 0), 1)index += 1return frame# 返回DLIB格式的face
def getDlibRect(detector='hog', face=None):l, t, r, b = None, None, None, Noneif detector == 'hog':l, t, r, b = face.left(), face.top(), face.right(), face.bottom()if detector == 'cnn':l = face.rect.left()t = face.rect.top()r = face.rect.right()b = face.rect.bottom()if detector == 'haar':l = face[0]t = face[1]r = face[0] + face[2]b = face[1] + face[3]nonnegative = lambda x: x if x >= 0 else 0return map(nonnegative, (l, t, r, b))# 获取CSV中信息
def getFeatList():print('加载注册的人脸特征')feature_list = Nonelabel_list = []name_list = []# 加载保存的特征样本with open('./data/feature.csv', 'r') as f:csv_reader = csv.reader(f)for line in csv_reader:# 重新加载数据faceId = line[0]userName = line[1]face_descriptor = eval(line[2])label_list.append(faceId)name_list.append(userName)# 转为numpy格式face_descriptor = np.asarray(face_descriptor, dtype=np.float64)# 转为二维矩阵，拼接face_descriptor = np.reshape(face_descriptor, (1, -1))# 初始化if feature_list is None:feature_list = face_descriptorelse:# 拼接feature_list = np.concatenate((feature_list, face_descriptor), axis=0)print("特征加载完毕")return feature_list, label_list, name_list# 人脸识别
def faceRecognize(detector='haar', threshold=0.5, write_video=False, resize_w=700, resize_h=400):# 视频时间frameTime = time.time()# 加载特征feature_list, label_list, name_list = getFeatList()face_time_dict = {}# 保存name,time人脸信息face_info_list = []# numpy格式人脸图像数据face_img_list = []# 侦测人数person_detect = 0# 统计人脸数face_count = 0# 控制显示打卡成功的时长show_time = (frameTime - 10)# 考勤记录f = open('./data/attendance.csv', 'a')csv_writer = csv.writer(f)cap = cv2.VideoCapture(0)# resize_w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))//2# resize_h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) //2videoWriter = cv2.VideoWriter('./record_video/out' + str(time.time()) + '.mp4', cv2.VideoWriter_fourcc(*'MP4V'), 15,(resize_w, resize_h))while True:ret, frame = cap.read()frame = cv2.resize(frame, (resize_w, resize_h))frame = cv2.flip(frame, 1)# 切换人脸检测器if detector == 'hog':face_detetion = hog_face_detector(frame, 1)if detector == 'cnn':face_detetion = cnn_detector(frame, 1)if detector == 'haar':frame_gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)face_detetion = haar_face_detector.detectMultiScale(frame_gray, minNeighbors=7, minSize=(100, 100))person_detect = len(face_detetion)for face in face_detetion:l, t, r, b = getDlibRect(detector, face)face = dlib.rectangle(l, t, r, b)# 识别68个关键点points = points_detector(frame, face)cv2.rectangle(frame, (l, t), (r, b), (0, 255, 0), 2)# 人脸区域face_crop = frame[t:b, l:r]# 特征face_descriptor = face_descriptor_extractor.compute_face_descriptor(frame, points)face_descriptor = [f for f in face_descriptor]face_descriptor = np.asarray(face_descriptor, dtype=np.float64)# 计算距离distance = np.linalg.norm((face_descriptor - feature_list), axis=1)# 最小距离索引min_index = np.argmin(distance)# 最小距离min_distance = distance[min_index]predict_name = "Not recog"if min_distance < threshold:# 距离小于阈值，表示匹配predict_id = label_list[min_index]predict_name = name_list[min_index]# 判断是否新增记录：如果一个人距上次检测时间>3秒，或者换了一个人，将这条记录插入need_insert = Falsenow = time.time()if predict_name in face_time_dict:if (now - face_time_dict[predict_name]) > 3:# 刷新时间face_time_dict[predict_name] = nowneed_insert = Trueelse:# 还是上次人脸need_insert = Falseelse:# 新增数据记录face_time_dict[predict_name] = nowneed_insert = Trueif (now - show_time) < 1:frame = cv2AddChineseText(frame, "打卡成功", (l, b + 30), textColor=(0, 255, 0), textSize=40)if need_insert:# 连续显示打卡成功1sframe = cv2AddChineseText(frame, "打卡成功", (l, b + 30), textColor=(0, 255, 0), textSize=40)show_time = time.time()time_local = time.localtime(face_time_dict[predict_name])# 转换成新的时间格式(2016-05-05 20:28:54)face_time = time.strftime("%H:%M:%S", time_local)face_time_full = time.strftime("%Y-%m-%d %H:%M:%S", time_local)# 开始位置增加face_info_list.insert(0, [predict_name, face_time])face_img_list.insert(0, face_crop)# 写入考勤表line = [predict_id, predict_name, min_distance, face_time_full]csv_writer.writerow(line)face_count += 1# 绘制人脸点cv2.putText(frame, predict_name + " " + str(round(min_distance, 2)), (l, b + 30), cv2.FONT_ITALIC, 0.8,(0, 255, 0), 2)# 处理下一张脸now = time.time()fpsText = 1 / (now - frameTime)frameTime = now# 绘制drawLeftInfo(frame, fpsText, 'Recog', detector=detector, person=person_detect, count=face_count)# 舍弃face_img_list、face_info_list后部分，节约内存if len(face_info_list) > 10:face_info_list = face_info_list[:9]face_img_list = face_img_list[:9]frame = updateRightInfo(frame, face_info_list, face_img_list)if write_video:videoWriter.write(frame)cv2.imshow('Face Attendance Demo: Recognition', frame)if cv2.waitKey(10) & 0xFF == ord('q'):breakf.close()videoWriter.release()cap.release()cv2.destroyAllWindows()然后效果就和我们宿舍楼下差不多了~
我年轻的时候，我大概比现在帅个几百倍吧，哎。
二、总代码
上文其实把登录和注册最后一部分代码放在一起就是了，这里就不再复制粘贴了，相关权重文件下载链接：opencv/data at master · opencv/opencv · GitHub
当然本项目还有很多需要优化的地方，比如设置用户不能重复、考勤打卡每天只能一次、把csv改为链接成数据库等等，后续代码优化完成后就可以部署然后和室友**了。
