-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathmain_rpi.py
269 lines (227 loc) · 7.8 KB
/
main_rpi.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
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
import socket, cv2, pickle, struct, imutils, time
from imutils.video import VideoStream
from funcs import tcp_read, tcp_write
from utils import *
import time
import tensorflow as tf
import numpy as np
# Set up code for GPIO
import RPi.GPIO as GPIO
pause_btn = 27
switch_model_btn = 5
GPIO.setmode(GPIO.BCM)
GPIO.setwarnings(False)
GPIO.setup(pause_btn, GPIO.IN, pull_up_down=GPIO.PUD_UP)
GPIO.setup(switch_model_btn, GPIO.IN, pull_up_down=GPIO.PUD_UP)
started = False
normal_model = GPIO.input(switch_model_btn)
#*************** For Normal Image Classifier *****************
def addText(img,text,thickness=0.7,color=(255,0,0),location=(50,50)):
cv2.putText(img,text,location,cv2.FONT_HERSHEY_SIMPLEX,thickness,color,1,cv2.LINE_AA)
def showscore(img,scores,legend,model_flag):
treshold=0.5
if model_flag:
treshold = 0.7
else:
treshold = 0.65
name="unknown"
gesture_ind = np.argmax(scores)
confidence = scores[0, gesture_ind]
if confidence>=treshold:
name=legend[gesture_ind]
addText(img,name)
print(confidence,name)
return name
def pred_word(imga,input_details,output_details):
word_interpreter.set_tensor(input_details[0]['index'], imga)
word_interpreter.invoke()
output_data = word_interpreter.get_tensor(output_details[0]['index'])
return showscore(img,output_data,word_legend,model_flag=True)
def pred_letter(imga,input_details,output_details):
letter_interpreter.set_tensor(input_details[0]['index'], imga)
letter_interpreter.invoke()
output_data = letter_interpreter.get_tensor(output_details[0]['index'])
return showscore(img,output_data,letter_legend,model_flag=False)
letter_model_dir = "/home/pi/python_client/normal_models/letter_sobel_III.tflite"
word_model_dir = "/home/pi/python_client/normal_models/word_sobel_II.tflite"
letter_legend = {
0:"A",
1:"B",
2:"C",
3:"D",
4:"E",
5:"F",
6:"G",
7:"H",
8:"I",
9:"J",
10:"K",
11:"L",
12:"M",
13:"N",
14:"O",
15:"P",
16:"Q",
17:"R",
18:"S",
19:"T",
20:"U",
21:"V",
22:"W",
23:"X",
24:"Y",
25:"Z",
26:"switch"
}
word_legend = {
0:"brother",
1:"hello",
2:"i",
3:"learn",
4:"learner_1",
5:"learner_2",
6:"name",
7:"no",
8:"switch"
}
scale = 2
delta = 0
ddepth = cv2.CV_16S
letter_interpreter = tf.lite.Interpreter(model_path=letter_model_dir)
letter_interpreter.allocate_tensors()
letter_input_details = letter_interpreter.get_input_details()
letter_output_details = letter_interpreter.get_output_details()
word_interpreter = tf.lite.Interpreter(model_path=word_model_dir)
word_interpreter.allocate_tensors()
word_input_details = word_interpreter.get_input_details()
word_output_details = word_interpreter.get_output_details()
word_pred = True
#*************************************************************
prev_pred=""
prev_confirmed_pred=""
k=0
def switch(ev = None):
global started
started = not started
print("started:", started)
def switch_model(ev = None):
global normal_model, server_socket, prev_pred, prev_confirmed_pred, k
if not normal_model:
try:
server_socket.close()
except:
print("Socket not closeable")
normal_model = not normal_model
# Reset params
prev_pred = ""
prev_confirmed_pred = ""
k = 0
print("normal_model:", normal_model)
GPIO.add_event_detect(pause_btn, GPIO.BOTH, callback=switch, bouncetime=300)
GPIO.add_event_detect(switch_model_btn, GPIO.BOTH, callback=switch_model, bouncetime=300)
server_socket = 0
vid = VideoStream(src=0, usePiCamera=True).start()
tts_("loaded")
while True:
if normal_model:
if not started:
continue
img = vid.read()
try:
img = cv2.resize(img,(240,240))
except:
continue
edges = preprocess_(img)
cv2.imshow("edges",edges)
imga = np.expand_dims(edges, 2)
imga = np.expand_dims(imga, 0).astype('float32')/255.
if word_pred==True:
label = pred_word(imga,word_input_details,word_output_details)
else:
label = pred_letter(imga,letter_input_details,letter_output_details)
if prev_pred == "":
prev_pred=label
prev_confirmed_pred="unknown"
elif prev_pred==label:
k+=1
if k>=8:
k=0
if label=="leaner_2" and prev_confirmed_pred == "leaner_1":
addText(img,"LEARNER",1,(0,0,255),(100,100))
elif label=="switch" and prev_confirmed_pred != "switch":
word_pred= not word_pred
print("Switching models")
addText(img,label,1,(0,0,255),(100,100))
else:
addText(img,label,1,(0,0,255),(100,100))
prev_confirmed_pred=label
tts_(label)
else:
prev_pred=label
k=0
cv2.imshow("test",img)
key = cv2.waitKey(1)
if key & 0xFF == 27:
break
else:
cv2.destroyAllWindows()
# Socket Create
server_socket = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
host_name = socket.gethostname()
host_ip = '192.168.1.6'
# host_ip = socket.gethostbyname(host_name)
print('HOST IP:',host_ip)
port = 9999
socket_address = (host_ip,port)
# Socket Accept
while True:
if normal_model:
server_socket.close()
break
try:
server_socket.connect(socket_address)
except:
print("error")
break
print("Connected")
if server_socket:
if normal_model:
server_socket.close()
break
while True:
if normal_model:
server_socket.close()
break
if not started:
continue
img = vid.read()
try:
frame = img[:]
except:
continue
a = pickle.dumps(frame)
message = struct.pack("Q",len(a))+a
server_socket.sendall(message)
# cv2.imshow('TRANSMITTING VIDEO',frame)
guess = tcp_read(server_socket)
if len(guess) > 1:
if guess == prev_pred:
k += 1
if k >= 3:
tts_(guess)
k = 0
prev_pred = ''
time.sleep(1)
else:
prev_pred = guess
k = 0
else:
prev_pred = ''
k = 0
"""key = cv2.waitKey(1) & 0xFF
if key ==ord('q'):
client_socket.close()"""
server_socket.close()
print("closed connection")
print("exited")
cv2.destroyAllWindows()