face_recognition_RealTime/main_function.py at master · LucasLee-ff/face_recognition_RealTime · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
import face_recognition as fr
import cv2
import sub_functions


ear_threshold = 0.25
mar_threshold = 0.65

# 加载已知人脸的照片
img_path = 'C:\\Users\\windows\\Desktop\\SRTP_FILE\\demo\\pictures'

# 编码文件夹下所有照片中的人脸，人名默认为每张照片的名字，返回值作为人脸识别的参照组
known_faces_encodings, known_faces_names = sub_functions.load_known_persons(img_path)


def main():
    video_capture = cv2.VideoCapture(0)

    total_blinks = 0
    total_mouth_open = 0
    count_eye = 0
    count_mouth = 0

    person = "fake"
    real = False
    mouth_can_open = False
    can_blink = False

    first_loop = True

    font = cv2.FONT_HERSHEY_DUPLEX

    while True:
        ret, frame = video_capture.read(0)
        rgb_frame = frame[:, :, ::-1]

        face_locations = fr.face_locations(rgb_frame)

        # 若摄像头中检测不到人脸则跳过此次循环
        if len(face_locations) == 0:
            continue

        # 找出主要的人脸位置
        main_face_location = sub_functions.find_main_face(face_locations)

        top = main_face_location[0][0]
        right = main_face_location[0][1]
        bottom = main_face_location[0][2]
        left = main_face_location[0][3]

        # 获取主要人脸的特征点
        main_face_landmarks = fr.face_landmarks_2(rgb_frame, main_face_location)

        # -----------------------人脸识别-----------------------
        # index为已知编码列表中与视频人脸编码相匹配的已知编码的下标
        main_face_encoding = fr.face_encodings(rgb_frame, main_face_location)
        index = sub_functions.recognition(known_faces_encodings, main_face_encoding)
        if index is not None:
            name = known_faces_names[index]
        else:
            name = "unknown"

        # 把本帧中检测到的脸与上一帧的脸作对比，如果与上一帧中不是同一个人，重新认定为假人
        if not first_loop:
            match_last_face = fr.compare_faces(face_encoding_backup, main_face_encoding[0])
            if not match_last_face[0]:
                count_eye = 0
                total_blinks = 0
                can_blink = False

                count_mouth = 0
                total_mouth_open = 0
                mouth_can_open = False

                person = "fake"
                real = False
            face_encoding_backup = main_face_encoding

        if first_loop:
            face_encoding_backup = main_face_encoding

        # 框脸
        cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 255), 1)
        cv2.rectangle(frame, (left, bottom - 20), (right, bottom + 30), (0, 0, 255), cv2.FILLED)

        # 标名字
        cv2.putText(frame, name, (left + 6, bottom - 1), font, 0.8, (255, 255, 255), 1)

        # 标注是真人还是假人
        cv2.putText(frame, person, (left + 6, bottom + 24), font, 0.8, (255, 255, 255), 1)

        # -----------------------活体检测-----------------------

        # 若此帧中的眼睛闭合，则count加1
        # 眼睛闭合时间大于等于2帧后再睁开才算作一次眨眼
        if sub_functions.eye_close_detection(main_face_landmarks, ear_threshold):
            count_eye += 1
        else:
            if count_eye >= 1:
                total_blinks += 1
                count_eye = 0
                can_blink = True

        # 张嘴检测
        if sub_functions.mouth_open_detection(main_face_landmarks, mar_threshold):
            count_mouth += 1
        else:
            if count_mouth >= 1:
                total_mouth_open += 1
                count_mouth = 0
                mouth_can_open = True

        # 能眨眼且张嘴即为真人
        if can_blink and mouth_can_open:
            person = "real"
            real = True

        cv2.putText(frame, "Total Blinks: {}".format(total_blinks), (25, 50), font, 0.8, (0, 0, 255), 1)
        cv2.putText(frame, "Total Mouth Open: {}".format(total_mouth_open), (25, 80), font, 0.8, (0, 0, 255), 1)
        cv2.putText(frame, "press 'q' to quit", (25, 110), font, 0.7, (255, 0, 0), 1)
        cv2.imshow('Video', frame)

        first_loop = False

        key = cv2.waitKey(1) & 0xFF
        if key == ord("q"):
            video_capture.release()
            break


if __name__ == "__main__":
    main()