-
Notifications
You must be signed in to change notification settings - Fork 2
/
main.py
146 lines (118 loc) · 4.31 KB
/
main.py
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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
from config import *
import time
import cv2
import tensorflow as tf
import numpy as np
import requests
import sys
from PIL import Image
import os
import face_recognition
import xlwings as xw
import atexit
import keras
# --init--
print("正在初始化表格")
app = xw.App(visible=True, add_book=False)
wb = app.books.open(r'./test.xlsx')
for i in range(len(face_names)):
wb.sheets['sheet1'].range(f'A{i + 1}').value = face_names[i]
# 行号
letters = []
for i in range(26):
letters.append(chr(ord("A") + i))
print("正在初始化变量")
file_name = "./test_data/nb.jpg"
IS_DRR = True
IS_SaveIMG = False
print("抓取视频流:", end="")
cap = cv2.VideoCapture(0)
print("成功")
# 加载模型
print("加载CNN模型")
model = keras.models.load_model('./model/model2.0.keras')
# --functions--
@atexit.register
def clean():
# wb.save()
wb.close()
app.quit()
def gesture_recognition(path):
img = keras.utils.load_img(
path, target_size=(112, 112)
)
img_array = keras.utils.img_to_array(img)
img_array = tf.expand_dims(img_array, 0) # Create a batch
predictions = model.predict(img_array)
score = tf.nn.softmax(predictions[0])
return class_names[np.argmax(score)], 100 * np.max(score)
def image_segmentation(path):
with open(path, "rb") as f:
response = requests.post(url, headers=headers, data=data, files={"image": f})
# Check for successful response
response.raise_for_status()
return response.json()
x = 2
if __name__ == "__main__":
if "-NoDRR" in sys.argv:
IS_DRR = False
if "-SaveIMG" in sys.argv:
IS_SaveIMG = True
print("开始执行--")
while True:
time_global = time.time()
for i in range(30):
ret, frame = cap.read()
cv2.imshow("A video", frame)
cv2.imwrite(file_name, frame)
image = Image.open(file_name)
image_backups = image
res = image_segmentation(file_name)
objects = res['data']
for i, obj in enumerate(objects):
t = time.time()
box = obj['box']
if obj['name'] != 'person':
continue
cropped_image = image_backups.crop((box['x1'], box['y1'], box['x2'], box['y2']))
cropped_image.save(f'./img/cropped_{i}.jpg')
class_name, confidence = gesture_recognition(f'./img/cropped_{i}.jpg')
picture = face_recognition.load_image_file(f'./img/cropped_{i}.jpg')
try:
encoding = face_recognition.face_encodings(picture)[0]
except IndexError as e:
print("没有人脸")
cv2.putText(frame,
class_name,
(int(box['x1']), int(box['y1']) + 30),
font, 1, (255, 255, 255), 1)
try:
res = face_recognition.face_distance(face_encodings_list, encoding)
face_name = face_names[np.argmin(res)]
print(
f"图片中的第{i}张人像:人名={face_name},行为={class_name},行为置信度={confidence},人像处理用时:{time.time() - t}")
face_index = face_names.index(face_name)
wb.sheets['sheet1'].range(f'{letters[x - 1]}{face_index + 1}').value = class_name
if IS_DRR:
cv2.putText(frame,
face_name,
(int(box['x1']), int(box['y1']) + 60),
font, 1, (255, 255, 255), 1)
except Exception as e:
pass
if not IS_SaveIMG:
os.remove(f'./img/cropped_{i}.jpg')
pass
if IS_DRR:
cv2.rectangle(frame, (int(box['x1']), int(box['y1'])), (int(box['x2']), int(box['y2'])),
(255, 255, 255), 1)
x += 1
if IS_DRR:
img_test1 = cv2.resize(frame, (int(frame.shape[0] / 2), int(frame.shape[1] / 2)))
cv2.imwrite(f"./debug/test{x} .jpg",frame)
cv2.imshow("A video", frame)
print(time_global - time.time())
c = cv2.waitKey(1)
if c == 27:
break
cv2.destroyAllWindows()