700字范文 > dlib实现人脸识别+活体检测

dlib实现人脸识别+活体检测

时间：2020-04-30 18:37:26

相关推荐

dlib实现人脸识别+活体检测

一：dlib的shape_predictor_68_face_landmarks模型

二、眨眼检测

三、张口检测

四、眨眼检测+张口检测

五、人脸识别

六、人脸识别+活体检测

七、人脸识别破解方法

八、参考资料及下载

一：dlib的shape_predictor_68_face_landmarks模型

该模型能够检测人脸的68个特征点（facial landmarks），定位图像中的眼睛，眉毛，鼻子，嘴巴，下颌线（ROI，Region of Interest）

下颌线[1,17]
左眼眉毛[18,22]
右眼眉毛[23,27]
鼻梁[28,31]
鼻子[32,36]
左眼[37,42]
右眼[43,48]
上嘴唇外边缘[49,55]
上嘴唇内边缘[66,68]
下嘴唇外边缘[56,60]
下嘴唇内边缘[61,65]

在使用的过程中对应的下标要减1，像数组的下标是从0开始。

模型链接：

/Lee_01/article/details/89140668

/Lee_01/article/details/89145740

二、眨眼检测

基本原理：计算眼睛长宽比 Eye Aspect Ratio，EAR.当人眼睁开时，EAR在某个值上下波动，当人眼闭合时，EAR迅速下降，理论上会接近于零，当时人脸检测模型还没有这么精确。所以我们认为当EAR低于某个阈值时，眼睛处于闭合状态。为检测眨眼次数，需要设置同一次眨眼的连续帧数。眨眼速度比较快，一般1~3帧就完成了眨眼动作。两个阈值都要根据实际情况设置。

原文链接：/Lee_01/article/details/89151044

from imutils.video import FileVideoStreamfrom imutils.video import VideoStreamfrom imutils import face_utilsimport numpy as npimport imutilsimport dlibimport cv2import sysdef _help():print("Usage:")print("python blink_detect.py")print("python blink_detect.py <path of a video>")print("For example:")print("python blink_detect.py video/lee.mp4")print("If the path of a video is not provided, the camera will be used as the input.Press q to quit.")def eye_aspect_ratio(eye):A = np.linalg.norm(eye[1] - eye[5])B = np.linalg.norm(eye[2] - eye[4])C = np.linalg.norm(eye[0] - eye[3])ear = (A + B) / (2.0 * C)return eardef blink_detection(vs, file_stream):# define three constants, one for the eye aspect ratio to indicate# blink and then the other constants for the min/max number of consecutive# frames the eye must be below the thresholdEAR_THRESH = 0.2EAR_CONSEC_FRAMES_MIN = 1EAR_CONSEC_FRAMES_MAX = 2# initialize the frame counters and the total number of blinksblink_counter = [0, 0] # left eye and right eyeblink_total = [0, 0] # left eye and right eyeprint("[INFO] loading facial landmark predictor...")detector = dlib.get_frontal_face_detector()predictor = dlib.shape_predictor("model/shape_predictor_68_face_landmarks.dat")# grab the indexes of the facial landmarks for the left and# right eye, respectively(lStart, lEnd) = face_utils.FACIAL_LANDMARKS_IDXS["left_eye"](rStart, rEnd) = face_utils.FACIAL_LANDMARKS_IDXS["right_eye"]print("[INFO] starting video stream thread...")while True:# if this is a file video stream, then we need to check if# there any more frames left in the buffer to processif file_stream and not vs.more():breakframe = vs.read()if frame is not None:frame = imutils.resize(frame)gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)rects = detector(gray, 0)if len(rects) == 1:rect = rects[0]shape = predictor(gray, rect)shape = face_utils.shape_to_np(shape)left_eye = shape[lStart:lEnd]right_eye = shape[rStart:rEnd]left_ear = eye_aspect_ratio(left_eye)right_ear = eye_aspect_ratio(right_eye)# compute the convex hull for the left and right eye, then# visualize each of the eyesleft_eye_hull = cv2.convexHull(left_eye)right_eye_hull = cv2.convexHull(right_eye)cv2.drawContours(frame, [left_eye_hull], -1, (0, 255, 0), 1)cv2.drawContours(frame, [right_eye_hull], -1, (0, 255, 0), 1)# check to see if the eye aspect ratio is below the blink# threshold, and if so, increment the blink frame counterif left_ear < EAR_THRESH:blink_counter[0] += 1# otherwise, the eye aspect ratio is not below the blink# thresholdelse:# if the eyes were closed for a sufficient number of# then increment the total number of blinksif EAR_CONSEC_FRAMES_MIN <= blink_counter[0] and blink_counter[0] <= EAR_CONSEC_FRAMES_MAX:blink_total[0] += 1blink_counter[0] = 0# draw the total number of blinks on the frame along with# the computed eye aspect ratio for the framecv2.putText(frame, "LBlinks: {}".format(blink_total[0]), (10, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "LEAR: {:.2f}".format(left_ear), (10, 60),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)# check to see if the eye aspect ratio is below the blink# threshold, and if so, increment the blink frame counterif right_ear < EAR_THRESH:blink_counter[1] += 1# otherwise, the eye aspect ratio is not below the blink# thresholdelse:# if the eyes were closed for a sufficient number of# then increment the total number of blinksif EAR_CONSEC_FRAMES_MIN <= blink_counter[1] and blink_counter[1] <= EAR_CONSEC_FRAMES_MAX:blink_total[1] += 1blink_counter[1] = 0# draw the total number of blinks on the frame along with# the computed eye aspect ratio for the framecv2.putText(frame, "RBlinks: {}".format(blink_total[1]), (200, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "REAR: {:.2f}".format(right_ear), (200, 60),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)elif len(rects) == 0:cv2.putText(frame, "No face!", (10, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)else:cv2.putText(frame, "More than one face!", (10, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.namedWindow("Frame", cv2.WINDOW_NORMAL)cv2.imshow("Frame", frame)# if the `q` key was pressed, break from the loopif cv2.waitKey(1) & 0xFF == ord('q'):breakcv2.destroyAllWindows()vs.stop()if len(sys.argv) > 2 or "-h" in sys.argv or "--help" in sys.argv:_help()elif len(sys.argv) == 2:vs = FileVideoStream(sys.argv[1]).start()file_stream = Trueblink_detection(vs, file_stream)else:vs = VideoStream(src=0).start()file_stream = Falseblink_detection(vs, file_stream)

dlib模型官网下载地址：/files/shape_predictor_68_face_landmarks.dat.bz2

三、张口检测

检测原理：类似眨眼检测，计算Mouth Aspect Ratio,MAR.当MAR大于设定的阈值时，认为张开了嘴巴。

四、眨眼检测+张口检测

from imutils.video import FileVideoStreamfrom imutils.video import VideoStreamfrom imutils import face_utilsimport numpy as npimport dlibimport cv2import sysdef _help():print("Usage:")print("python liveness_detect.py")print("python liveness_detect.py <path of a video>")print("For example:")print("python liveness_detect.py video/lee.mp4")print("If the path of a video is not provided, the camera will be used as the input.Press q to quit.")def eye_aspect_ratio(eye):# (|e1-e5|+|e2-e4|) / (2|e0-e3|)A = np.linalg.norm(eye[1] - eye[5])B = np.linalg.norm(eye[2] - eye[4])C = np.linalg.norm(eye[0] - eye[3])ear = (A + B) / (2.0 * C)return eardef mouth_aspect_ratio(mouth):# (|m2-m9|+|m4-m7|)/(2|m0-m6|)A = np.linalg.norm(mouth[2] - mouth[9]) # 51, 59B = np.linalg.norm(mouth[4] - mouth[7]) # 53, 57C = np.linalg.norm(mouth[0] - mouth[6]) # 49, 55mar = (A + B) / (2.0 * C)return mardef liveness_detection(vs, file_stream):EAR_THRESH = 0.15EAR_CONSEC_FRAMES_MIN = 1EAR_CONSEC_FRAMES_MAX = 2MAR_THRESH = 0.5# 初始化眨眼的连续帧数以及总的眨眼次数blink_counter = 0blink_total = 0print("[INFO] loading facial landmark predictor...")detector = dlib.get_frontal_face_detector()predictor = dlib.shape_predictor("model/shape_predictor_68_face_landmarks.dat")(lStart, lEnd) = face_utils.FACIAL_LANDMARKS_IDXS["left_eye"](rStart, rEnd) = face_utils.FACIAL_LANDMARKS_IDXS["right_eye"](mStart, mEnd) = face_utils.FACIAL_LANDMARKS_IDXS["mouth"]print("[INFO] starting video stream thread...")while True:# if this is a file video stream, then we need to check if# there any more frames left in the buffer to processif file_stream and not vs.more():breakframe = vs.read()if frame is not None:gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)rects = detector(gray, 0)# 只能处理一张人脸if len(rects) == 1:shape = predictor(gray, rects[0]) # 保存68个特征点坐标的<class 'dlib.dlib.full_object_detection'>对象shape = face_utils.shape_to_np(shape) # 将shape转换为numpy数组，数组中每个元素为特征点坐标left_eye = shape[lStart:lEnd]right_eye = shape[rStart:rEnd]left_ear = eye_aspect_ratio(left_eye)right_ear = eye_aspect_ratio(right_eye)ear = (left_ear + right_ear) / 2.0mouth = shape[mStart:mEnd]mar = mouth_aspect_ratio(mouth)left_eye_hull = cv2.convexHull(left_eye)right_eye_hull = cv2.convexHull(right_eye)mouth_hull = cv2.convexHull(mouth)cv2.drawContours(frame, [left_eye_hull], -1, (0, 255, 0), 1)cv2.drawContours(frame, [right_eye_hull], -1, (0, 255, 0), 1)cv2.drawContours(frame, [mouth_hull], -1, (0, 255, 0), 1)# EAR低于阈值，有可能发生眨眼，眨眼连续帧数加一次if ear < EAR_THRESH:blink_counter += 1# EAR高于阈值，判断前面连续闭眼帧数，如果在合理范围内，说明发生眨眼else:# if the eyes were closed for a sufficient number of# then increment the total number of blinksif EAR_CONSEC_FRAMES_MIN <= blink_counter and blink_counter <= EAR_CONSEC_FRAMES_MAX:blink_total += 1blink_counter = 0cv2.putText(frame, "Blinks: {}".format(blink_total), (0, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "Mouth: {}".format("open" if mar > MAR_THRESH else "closed"),(130, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "EAR: {:.2f}".format(ear), (300, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "MAR: {:.2f}".format(mar), (450, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)elif len(rects) == 0:cv2.putText(frame, "No face!", (0, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)else:cv2.putText(frame, "More than one face!", (0, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.namedWindow("Frame", cv2.WINDOW_NORMAL)cv2.imshow("Frame", frame)# 按下q键退出循环（鼠标要点击一下图片使图片获得焦点）if cv2.waitKey(1) & 0xFF == ord('q'):breakcv2.destroyAllWindows()vs.stop()if len(sys.argv) > 2 or "-h" in sys.argv or "--help" in sys.argv:_help()elif len(sys.argv) == 2:video_stream = FileVideoStream(sys.argv[1]).start()file_stream = Trueliveness_detection(video_stream, file_stream)else:video_stream = VideoStream(src=0).start()file_stream = Falseliveness_detection(video_stream, file_stream)

五、人脸识别

# -*-coding:GBK -*-import face_recognitionimport osimport cv2from PIL import Image, ImageFont, ImageDrawimport numpy as npimport datetimeimport threadingclass Recorder:passrecord_dic = {}unknown_pic = []flag_over = 0 # 定义一个是否进行来访记录的标记# 定时去保存对比图像信息，并且将位置人员的图像保存下来def save_recorder(name, frame):global record_dicglobal flag_overglobal unknown_picif flag_over == 1: returntry:record = record_dic[name]seconds_diff = (datetime.datetime.now() - record.times[-1]).total_seconds()if seconds_diff < 60 * 10:returnrecord.times.append(datetime.datetime.now())print('更新记录', record_dic, record.times)except KeyError:newRec = Recorder()newRec.times = [datetime.datetime.now()]record_dic[name] = newRecprint('添加记录', record_dic, newRec.times)if name == '未知头像':s = str(record_dic[name].times[-1])# print(s)# 未知人员的图片名称filename = s[:10]+s[-6:] + '.jpg'cv2.imwrite(filename, frame)unknown_pic.append(filename)# 解析已有人员的所有照片并得到照片名和人物面部编码信息def load_img(path):print('正在加载已知人员的图片...')for dirpath, dirnames, filenames in os.walk(path):print(filenames)facelib = []for filename in filenames:filepath = os.sep.join([dirpath, filename])# 把对应每张图片加载进来face_image = face_recognition.load_image_file(filepath)face_encoding = face_recognition.face_encodings(face_image)[0]facelib.append(face_encoding)return facelib,filenamesfacelib, facenames = load_img('facelib')# print(facenames)video_capture = cv2.VideoCapture(0)while True:ret, frame = video_capture.read()# 通过缩小图片（缩小为1/4），提高对比效率small_frame = cv2.resize(frame, (0,0), fx=0.25, fy=0.25)rgb_small_frame = small_frame[:,:,::-1] # 将opencv的BGR格式转换为RGB格式face_locations = face_recognition.face_locations(rgb_small_frame)face_encodings = face_recognition.face_encodings(rgb_small_frame, face_locations)face_names = []# 循环多张人脸for face_encoding in face_encodings:matches = pare_faces(facelib, face_encoding, tolerance=0.39)name = '未知头像'if True in matches:# 如果摄像头里面的头像匹配了已知人物头像，则取出第一个True的位置first_match_index = matches.index(True)name = facenames[first_match_index][:-4] # 取出文件上对应的人名face_names.append(name)for (top, right, bottom, left), name in zip(face_locations, face_names):# 还原原图片大小top *= 4right *= 4bottom *= 4left *= 4cv2.rectangle(frame, (left, top), (right, bottom), (0,0,255), thickness=2) # 标注人脸信息img_PIL = Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB))font = ImageFont.truetype('simhei.ttf', 40)draw = ImageDraw.Draw(img_PIL)draw.text((left+6, bottom-6), name, font=font, fill=(255,255,255))frame = cv2.cvtColor(np.asarray(img_PIL),cv2.COLOR_RGB2BGR)save_recorder(name, frame)cv2.imshow('capture', frame)if cv2.waitKey(1) & 0xFF == ord('q'):breakvideo_capture.release()

六、人脸识别+活体检测

import face_recognitionfrom imutils import face_utilsfrom PIL import Image, ImageDraw,ImageFontimport numpy as npimport threading#导入threading模块import yagmailimport dlibimport datetimeimport timeimport cv2import osimport sys# 初始化眨眼次数blink_total = 0# 初始化张嘴次数mouth_total = 0# 设置图片存储路径# pic_path = './dataset'# 图片数量pic_total = 0# 初始化眨眼的连续帧数以及总的眨眼次数blink_counter = 0# 初始化张嘴状态为闭嘴mouth_status_open = 0# 眼长宽比例值EAR_THRESH = 0.15EAR_CONSEC_FRAMES_MIN = 1EAR_CONSEC_FRAMES_MAX = 5 # 当EAR小于阈值时，接连多少帧一定发生眨眼动作# 嘴长宽比例值MAR_THRESH = 0.15# 人脸检测器detector = dlib.get_frontal_face_detector()# 特征点检测器predictor = dlib.shape_predictor("modles/shape_predictor_68_face_landmarks.dat")# 获取左眼的特征点(lStart, lEnd) = face_utils.FACIAL_LANDMARKS_IDXS["left_eye"]# 获取右眼的特征点(rStart, rEnd) = face_utils.FACIAL_LANDMARKS_IDXS["right_eye"]# 获取嘴巴特征点(mStart, mEnd) = face_utils.FACIAL_LANDMARKS_IDXS["inner_mouth"]class Recorder:passred_dict = {}unknownjpg = []def sendemail(title,contents,fileslist):#将照片和访问记录上传云端# yag = yagmail.SMTP("发件人邮箱",'密码' ,'', 465)# yag.send(['收件人邮箱1','收件人邮箱2'],title,contents,fileslist)yag=yagmail.SMTP("xxx@",'mbiwgdukvqaadfei','',465)#发件人邮箱yag.send(['xxx@','xxx@'],title,contents,fileslist)#收件人邮箱(注意和上面邮箱不同)def dicttostr():#生成来访记录列表strlist = []listkey =list(sorted(red_dict.keys()))#取字典的keyfor item in listkey:#通过循环，合成每一条来访记录strlist.extend([item + ','+str(onetime) for onetime in red_dict[item].times])return strlistflagover = 0#全局标志，用来控制是否保持来访记录def saveRecorder(name, frame):#保存和添加来访记录global red_dictglobal flagoverglobal unknownjpgif flagover == 1:#响应全局标志，如果为1时，关闭来访记录returntry:red = red_dict[name]#如果多次识别，比较时间secondsDiff = (datetime.datetime.now() - red.times[-1]).total_seconds()if secondsDiff < 5*60: # 如果两次识别在5分钟内，将被过滤掉returnred.times.append(datetime.datetime.now())print('更新记录', red_dict, red.times)except (KeyError):newRed = Recorder()newRed.times = [datetime.datetime.now()]red_dict[name] = newRedprint('添加记录', red_dict, newRed.times)if name == 'Unknown':s = str(red_dict[name].times[-1])print('写入', s[:10] + s[-6:])filename = s[:10] + s[-6:] + '.jpg'cv2.imwrite(filename, frame)unknownjpg.append(filename)def loop_timer_headle(): # 定时器循环触发函数print('————————Timer headle!————————', str(datetime.datetime.now()))global timer2global flagoverglobal red_dictglobal unknownjpgflagover = 1timer2 = threading.Timer(60 * 5, loop_timer_headle) # 创建定时器 5分钟timer2.start()# 发送邮件sendemail("来访统计记录", '\n'.join(dicttostr()), unknownjpg)red_dict.clear()unknownjpg.clear()print("清空")time.sleep(10)print("重新开始")flagover = 0timer2 = threading.Timer(2, loop_timer_headle)timer2.start()def load_img(sample_dir):#导入数据库照片print('loading sample face..')for (dirpath, dirnames, filenames) in os.walk(sample_dir): # 一级一级的文件夹递归print(dirpath, dirnames, filenames)facelib = []for filename in filenames:filename_path = os.sep.join([dirpath, filename])print(filename_path)faceimage = face_recognition.load_image_file(filename_path)# 由于我们每个图像只有一个脸，我只关心每个图像中的第一个编码，所以我取索引0face_encoding = face_recognition.face_encodings(faceimage)[0]facelib.append(face_encoding)return facelib, filenames# def getFaceEncoding(src):#获取人脸编码#image = face_recognition.load_image_file(src) # 加载人脸图片## 获取图片人脸定位[(top,right,bottom,left )]#face_locations = face_recognition.face_locations(image)#img_ = image[face_locations[0][0]:face_locations[0][2], face_locations[0][3]:face_locations[0][1]]#img_ = cv2.cvtColor(img_, cv2.COLOR_BGR2RGB)## display(img_)#face_encoding = face_recognition.face_encodings(image, face_locations)[0] # 默认人脸数为1，对人脸图片进行编码#return face_encoding#对比两张照片距离# def simcos(a, b):#a = np.array(a)#b = np.array(b)#dist = np.linalg.norm(a - b) # 二范数#sim = 1.0 / (1.0 + dist) ##return sim# 提供对外比对的接口返回比对的相似度# def comparison(face_src1, face_src2):#xl1 = getFaceEncoding(face_src1)#xl2 = getFaceEncoding(face_src2)#value = simcos(xl1, xl2)#print(value)# 眼长宽比例def eye_aspect_ratio(eye):# (|e1-e5|+|e2-e4|) / (2|e0-e3|)A = np.linalg.norm(eye[1] - eye[5])B = np.linalg.norm(eye[2] - eye[4])C = np.linalg.norm(eye[0] - eye[3])ear = (A + B) / (2.0 * C)return ear# 嘴长宽比例def mouth_aspect_ratio(mouth):A = np.linalg.norm(mouth[1] - mouth[7]) # 61, 67B = np.linalg.norm(mouth[3] - mouth[5]) # 63, 65C = np.linalg.norm(mouth[0] - mouth[4]) # 60, 64mar = (A + B) / (2.0 * C)return mardef main():global blink_total # 使用global声明blink_total，在函数中就可以修改全局变量的值global mouth_totalglobal pic_totalglobal blink_counterglobal mouth_status_open# video_path, src = sys.argv[1], sys.argv[2]facelib, facename = load_img('dataset')vs = cv2.VideoCapture(0)face_locations = [] # 定义列表存放人脸位置face_encodings = [] # 定义列表存放人脸特征编码process_this_frame = True # 定义信号量while True:ret, frame = vs.read() # 捕获一帧图片small_frame = cv2.resize(frame, (0, 0), fx=0.25, fy=0.25) # 将图片缩小1/4，为人脸识别提速rgb_small_frame = small_frame[:, :, ::-1] # 将opencv的BGR格式转为RGB格式if process_this_frame: # 使用信号量对当前的处理进行保护# 找到人脸位置，并生成特征码face_locations = face_recognition.face_locations(rgb_small_frame)face_encodings = face_recognition.face_encodings(rgb_small_frame, face_locations)face_names = [] # 定义列表，放置识别结果for face_encoding in face_encodings: # 循环多张人脸matches = pare_faces(facelib, face_encoding) # 人脸识别name = "Unknown" # 定义默认的识别结果为Unknownif True in matches: # 如果识别出来，就将名称取出first_match_index = matches.index(True)name = facename[first_match_index][:-4]face_names.append(name) # 保存识别结果process_this_frame = not process_this_frame # 信号量保护结束# 显示结果for (top, right, bottom, left), name in zip(face_locations, face_names):top *= 4 # 还原人脸的原始尺寸right *= 4bottom *= 4left *= 4cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 255), 2) # 标注人脸img_PIL = Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)) # 转换图片格式font = ImageFont.truetype('simhei.ttf', 40) # 加载字体position = (left + 6, bottom - 6) # 指定文字输出位置draw = ImageDraw.Draw(img_PIL) # 绘制照片draw.text(position, name, font=font, fill=(255, 255, 255)) # 绘制文字frame = cv2.cvtColor(np.asarray(img_PIL), cv2.COLOR_RGB2BGR) # 将图片转回OpenCV格式saveRecorder(name, frame) # 过滤并保存记录gray = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)rects = detector(gray, 0) # 人脸检测# 只能处理一张人脸if len(rects) == 1:shape = predictor(gray, rects[0]) # 保存68个特征点坐标的<class 'dlib.dlib.full_object_detection'>对象shape = face_utils.shape_to_np(shape) # 将shape转换为numpy数组，数组中每个元素为特征点坐标left_eye = shape[lStart:lEnd] # 取出左眼对应的特征点right_eye = shape[rStart:rEnd] # 取出右眼对应的特征点left_ear = eye_aspect_ratio(left_eye) # 计算左眼EARright_ear = eye_aspect_ratio(right_eye) # 计算右眼EARear = (left_ear + right_ear) / 2.0 # 求左右眼EAR的均值inner_mouth = shape[mStart:mEnd] # 取出嘴巴对应的特征点mar = mouth_aspect_ratio(inner_mouth) # 求嘴巴mar的均值# left_eye_hull = cv2.convexHull(left_eye) # 寻找左眼轮廓# right_eye_hull = cv2.convexHull(right_eye) # 寻找右眼轮廓# mouth_hull = cv2.convexHull(inner_mouth) # 寻找内嘴巴轮廓# cv2.drawContours(frame, [left_eye_hull], -1, (0, 255, 0), 1) # 绘制左眼轮廓# cv2.drawContours(frame, [right_eye_hull], -1, (0, 255, 0), 1) # 绘制右眼轮廓# cv2.drawContours(frame, [mouth_hull], -1, (0, 255, 0), 1) # 绘制嘴巴轮廓# EAR低于阈值，有可能发生眨眼，眨眼连续帧数加一次if ear < EAR_THRESH:blink_counter += 1# EAR高于阈值，判断前面连续闭眼帧数，如果在合理范围内，说明发生眨眼else:# if the eyes were closed for a sufficient number of# then increment the total number of blinksif EAR_CONSEC_FRAMES_MIN <= blink_counter <= EAR_CONSEC_FRAMES_MAX:blink_total += 1blink_counter = 0# 通过张、闭来判断一次张嘴动作if mar > MAR_THRESH:mouth_status_open = 1else:if mouth_status_open:mouth_total += 1mouth_status_open = 0cv2.putText(frame, "Blinks: {}".format(blink_total), (0, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "Mouth: {}".format(mouth_total),(130, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "EAR: {:.2f}".format(ear), (300, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "MAR: {:.2f}".format(mar), (450, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)elif len(rects) == 0:cv2.putText(frame, "No face!", (0, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)else:cv2.putText(frame, "More than one face!", (0, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)# cv2.namedWindow("Frame", cv2.WINDOW_NORMAL)# cv2.imshow('Frame', frame) # 将图片显示出来# liveness_detection(vs)if blink_total >= 1 and mouth_total >= 1:cv2.putText(frame, "True", (200, 200),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)blink_total = 0mouth_total = 0cv2.namedWindow("Frame", cv2.WINDOW_NORMAL)cv2.imshow('Frame', frame) # 将图片显示出来k = cv2.waitKey(1) & 0xFF # 按键判断if k == ord(' '):break;# comparison(pic_path, src)vs.release()time.sleep(2) # 休眠2秒if __name__ == '__main__':main()