-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
536 lines (506 loc) · 25.4 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
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
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
"""
--- Edith ---
1. Use OpenCV to read video stream from camera on Paspberry Pi
2. Identify 3-dimensional coordinates of key points of both hands
3. Activate circle-drawing when the index fingers stop moving
4. Trigger Recognition: when the circles corresponding to the index
fingers of both hands are fully formed, the recognition begins
5. If there are any characters, OCR begins; otherwise, Object-dection begins
"""
import cv2
import time
import math
import torch
import requests
import numpy as np
import mediapipe as mp
# new import for streaming
import queue
import threading
# new import for voice broadcast
import os
import pyttsx3
import socket
import hashlib
from paddleocr import PaddleOCR
from PIL import Image, ImageDraw, ImageFont
font = cv2.FONT_HERSHEY_SIMPLEX
rtmp_str = 'rtsp://172.20.10.5:9554/webcam'
remote = False # use remote camera or not
q = queue.Queue() # store the frames
q_result = queue.Queue() # store Lable and Description
def engine_init():
engine_name = pyttsx3.init() # initialize voice engine
engine_name.setProperty('rate', 100) # set voice rate
engine_name.setProperty('volume', 1) # set voice volume
voices = engine_name.getProperty('voices')
engine_name.setProperty('voice', voices[1].id) # set first speech synthesizer
return engine_name
def voice_broadcast():
ip = "172.20.10.4"
port = 6969
ip_port = (ip, port)
server = socket.socket()
server.bind(ip_port)
server.listen(5)
engine = engine_init()
print("connection start..\n")
while True:
conn, addr = server.accept()
print("conn:", conn, "\naddr:", addr)
while True:
filename = "test.mp3"
# generate vioce file(in this demo the context is fixed)
# text = "Hello world, voice broadcast test 001."
text = q_result.get()
# engine.say(text)
engine.save_to_file(text, filename)
engine.runAndWait()
engine.stop()
if os.path.isfile(filename): # if the file exist
# 1.send file size
size = os.stat(filename).st_size
header = str(size) + ' ' + filename
conn.send(header.encode("utf-8"))
print("file size：", size)
# 2.send file
conn.recv(1024) # wait for client's ready signal
m = hashlib.md5()
f = open(filename, "rb")
for line in f:
conn.send(line) # send data
m.update(line)
f.close()
# 3.send md5 to verify
md5 = m.hexdigest()
conn.send(md5.encode("utf-8"))
print("md5:", md5)
# sleep 15 second (for this demo)
# time.sleep(15)
server.close()
def Receive():
print("start Reveive")
cap = cv2.VideoCapture(rtmp_str)
ret, frame = cap.read()
q.put(frame)
while ret:
ret, frame = cap.read()
q.put(frame)
# drawing class
class DrawSomeInfo:
def __init__(self):
self.hand_num = 0
# record imformation of left and right hands
# coordinates
self.last_finger_cord_x = {'Left': 0, 'Right': 0}
self.last_finger_cord_y = {'Left': 0, 'Right': 0}
# circles degree
self.last_finger_arc_degree = {'Left': 0, 'Right': 0}
# right hand circle
self.right_hand_circle_list = []
# initialize retention time
now = time.time()
self.stop_time = {'Left': now, 'Right': now}
# circles color
self.handedness_color = {'Left': (90, 208, 10), 'Right': (50, 52, 53)}
# finger movement range(adjust according to your camera)
self.float_distance = 10
# time to trigger
self.activate_duration = 0.3
self.last_thumb_img = None
# OCR text recognition model
self.pp_ocr = PaddleOCR(use_angle_cls=True, lang="ch", use_gpu=False,
rec_model_dir='./inference/ch_ppocr_server_v2.0_rec_infer/',
cls_model_dir='./inference/ch_ppocr_mobile_v2.0_cls_infer/',
det_model_dir='./inference/ch_ppocr_server_v2.0_det_infer/')
# Yolov5 object recognition model
self.yolov5_dete = torch.hub.load('./yolov5', 'custom', path='./weights/yolov5n.pt', source='local')
self.yolov5_dete.conf = 0.4
# last detection result
self.last_detect_res = {'detection': None, 'ocr': 'None'}
# description of last detection result
self.last_detect_des = {'detection': None, 'None': 'None'}
# add text
def cv2AddText(self, img, text, position, textColor=(0, 255, 0), textSize=30):
# judge type of the image
if (isinstance(img, np.ndarray)):
img = Image.fromarray(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
# instantiate a object to draw on the gave image
draw = ImageDraw.Draw(img)
# form of text
fontStyle = ImageFont.truetype(
"./fonts/FederationBold.TTF", textSize, encoding="utf-8")
# draw text
draw.text(position, text, textColor, font=fontStyle)
# converse the type back to openCV
return cv2.cvtColor(np.asarray(img), cv2.COLOR_RGB2BGR)
# generate text area on top right-hand corner
def generateOcrTextArea(self, ocr_text, line_text_num, line_num, x, y, w, h, frame):
# first we crop the sub-rect from the image
sub_img = frame[y:y + h, x:x + w]
green_rect = np.ones(sub_img.shape, dtype=np.uint8) * 0
res = cv2.addWeighted(sub_img, 0.5, green_rect, 0.5, 1.0)
for i in range(line_num):
text = ocr_text[(i * line_text_num):(i + 1) * line_text_num]
res = self.cv2AddText(res, text, (10, 30 * i + 10), textColor=(242, 242, 230), textSize=18)
return res
# generate label area
def generateLabelArea(self, text, x, y, w, h, frame):
# crop the sub-rect from the image
sub_img = frame[y:y + h, x:x + w]
green_rect = np.ones(sub_img.shape, dtype = np.uint8) * 0
res = cv2.addWeighted(sub_img, 0.5, green_rect, 0.5, 1.0)
if res is not None:
res = self.cv2AddText(res, text, (10, 10), textColor = (242, 242, 230), textSize = 25)
else:
return 0
return res
# generate thumbnail on top right-hand corner
def generateThumb(self, raw_img, frame):
# object recognition
if self.last_detect_res['detection'] == None:
results = self.yolov5_dete(raw_img)
# nothing detected
if len(results) == 0:
print("nothing detected")
if len(results) > 0:
results_np = results.pandas().xyxy[0].to_numpy()
if len(results_np) > 0:
label_id = results_np[0][5]
label_name = results_np[0][6]
self.last_detect_res['detection'] = [label_id, label_name]
# wiki search
keyword = self.last_detect_res['detection'][1]
url = "https://www.wikidata.org/w/api.php"
params = {
'action': 'wbsearchentities',
'format': 'json',
'search': keyword,
'language': 'en',
'type': 'item',
'limit': 1 # max num
}
# visit
get = requests.get(url=url, params=params)
# turn to json data
re_json = get.json()
self.last_detect_des['detection'] = re_json["search"][0]["display"]["description"]["value"]
lable = self.last_detect_res['detection'][1]
intro = self.last_detect_des['detection']
# print("Label Name: ", lable)
# print("Description: ", intro)
q_result.put(lable + ' ' + intro)
else:
self.last_detect_res['detection'] = [0, 'None']
self.last_detect_des['detection'] = ['None', 'None']
else:
self.last_detect_res['detection'] = [0, 'None']
self.last_detect_des['detection'] = ['None', 'None']
# complete image
frame_height, frame_width, _ = frame.shape
# cover
raw_img_h, raw_img_w, _ = raw_img.shape
thumb_img_w = 200
thumb_img_h = math.ceil(raw_img_h * thumb_img_w / raw_img_w)
thumb_img = cv2.resize(raw_img, (thumb_img_w, thumb_img_h))
rect_weight = 4
# draw square frame on thumbnail
thumb_img = cv2.rectangle(thumb_img, (0, 0), (thumb_img_w, thumb_img_h), (0, 139, 247), rect_weight)
# generate label
x, y, w, h = (frame_width - thumb_img_w), thumb_img_h, thumb_img_w, 50
# put the image back to its position
text = self.last_detect_res['detection'][1]
frame = frame.copy()
frame[y:y + h, x:x + w] = self.generateLabelArea(text, x, y, w, h, frame)
# OCR
ocr_text = ''
if self.last_detect_res['ocr'] == 'None':
result = self.pp_ocr.ocr(raw_img, cls=True)
src_im = raw_img
text_list = [line[1][0] for line in result]
thumb_img = cv2.resize(src_im, (thumb_img_w, thumb_img_h))
if len(text_list) > 0:
ocr_text = ''.join(text_list)
# record detected text
self.last_detect_res['ocr'] = ocr_text
else:
# nothing detected
self.last_detect_res['ocr'] = 'checked_no'
else:
ocr_text = self.last_detect_res['ocr']
frame[0:thumb_img_h, (frame_width - thumb_img_w):frame_width, :] = thumb_img
# whether required display
if ocr_text != '' and ocr_text != 'checked_no':
line_text_num = 15
line_num = math.ceil(len(ocr_text) / line_text_num)
y, h = (y + h + 20), (32 * line_num)
frame[y:y + h, x:x + w] = self.generateOcrTextArea(ocr_text, line_text_num, line_num, x, y, w, h, frame)
self.last_thumb_img = thumb_img
return frame
# draw circles
def drawArc(self, frame, point_x, point_y, arc_radius=150, end=360, color=(255, 0, 255), width=20):
img = Image.fromarray(frame)
shape = [(point_x - arc_radius, point_y - arc_radius),
(point_x + arc_radius, point_y + arc_radius)]
img1 = ImageDraw.Draw(img)
img1.arc(shape, start=0, end=end, fill=color, width=width)
frame = np.asarray(img)
return frame
# clear current mode
def clear(self):
self.right_hand_circle_list = []
self.last_finger_arc_degree = {'Left': 0, 'Right': 0}
# check whether the retention period is longer than 0.3s. Start to draw independently is time is over 0.3s
def checkIndexFingerMove(self, handedness, finger_cord, frame, frame_copy):
# calculate distance
x_distance = abs(finger_cord[0] - self.last_finger_cord_x[handedness])
y_distance = abs(finger_cord[1] - self.last_finger_cord_y[handedness])
# lock mode
# if not moving
if (x_distance <= self.float_distance) and (y_distance <= self.float_distance):
# retention time longer than time of trigger
if (time.time() - self.stop_time[handedness]) > self.activate_duration:
# draw circles, increase by 5 degrees/0.01s
arc_degree = 5 * ((time.time() - self.stop_time[handedness] - self.activate_duration) // 0.01)
if arc_degree <= 360:
frame = self.drawArc(
frame, finger_cord[0], finger_cord[1], arc_radius=50, end=arc_degree,
color=self.handedness_color[handedness], width=15)
else:
frame = self.drawArc(
frame, finger_cord[0], finger_cord[1], arc_radius=50, end=360,
color=self.handedness_color[handedness], width=15)
# make the degrees be 360
self.last_finger_arc_degree[handedness] = 360
# recognition starts when 2 circles are formed
if (self.last_finger_arc_degree['Left'] >= 360) and (
self.last_finger_arc_degree['Right'] >= 360):
# get the coordinates
if self.last_finger_cord_x['Right'] < self.last_finger_cord_x['Left']:
rect_l = (self.last_finger_cord_x['Right'], self.last_finger_cord_y['Left'])
rect_r = (self.last_finger_cord_x['Right'], self.last_finger_cord_y['Left'])
else:
rect_l = (self.last_finger_cord_x['Left'], self.last_finger_cord_y['Right'])
rect_r = (self.last_finger_cord_x['Left'], self.last_finger_cord_y['Right'])
# generate square frame of the recognition area
frame = cv2.rectangle(frame, rect_l, rect_r, (0, 255, 0), 2)
# generate label for square frame of the recognition area
if self.last_detect_res['detection']:
x, y, w, h = self.last_finger_cord_x['Left'], (
self.last_finger_cord_y['Left'] - 50), 200, 50
frame = frame.copy()
text = self.last_detect_res['detection'][1]
frame[y:y + h, x:x + w] = self.generateLabelArea(text, x, y, w, h, frame)
# initialize recognition results
self.last_detect_res = {'detection': None, 'ocr': 'None'}
# transmit results to thumbnail
y1 = min(self.last_finger_cord_y['Right'], self.last_finger_cord_y['Left'])
y2 = max(self.last_finger_cord_y['Right'], self.last_finger_cord_y['Left'])
x1 = min(self.last_finger_cord_x['Right'], self.last_finger_cord_x['Left'])
x2 = max(self.last_finger_cord_x['Right'], self.last_finger_cord_x['Left'])
raw_img = frame_copy[
y1:y2,
x1:x2, ]
frame = self.generateThumb(raw_img, frame)
# if (self.hand_num == 1) and (self.last_finger_arc_degree['Right'] == 360):
# self.right_hand_circle_list.append((finger_cord[0], finger_cord[1]))
else:
# restart clock when palms start moving dramatically
self.stop_time[handedness] = time.time()
self.last_finger_arc_degree[handedness] = 0
# refresh position
self.last_finger_cord_x[handedness] = finger_cord[0]
self.last_finger_cord_y[handedness] = finger_cord[1]
return frame
# recognition-controlling class
class VirtualFingerReader:
def __init__(self):
# initialize mediapipe
self.mp_drawing = mp.solutions.drawing_utils
self.mp_drawing_styles = mp.solutions.drawing_styles
self.mp_hands = mp.solutions.hands
self.image = None
def process(self):
th1 = threading.Thread(target=VirtualFingerReader.recognize, args=(self,))
th1.start()
th1.join()
# check the index of left and right hands in the array
# NOTICE: Mediapipe use mirror image
def checkHandsIndex(self, handedness):
# judge number of palms
if len(handedness) == 1:
handedness_list = ['Left' if handedness[0].classification[0].label == 'Right' else 'Right']
else:
handedness_list = [handedness[1].classification[0].label, handedness[0].classification[0].label]
return handedness_list
# main function
def recognize(self):
# initialize drawing class
drawInfo = DrawSomeInfo()
# calculate FPS
fpsTime = time.time()
# use OpenCV to read video stream
if remote == True:
print("Using web camera")
cap = cv2.VideoCapture('rtsp://172.20.10.5:9554/webcam')
else:
print("Using local camera")
cap = cv2.VideoCapture(0)
# resolution ratio
resize_w = 960
resize_h = 720
fps = cap.get(cv2.CAP_PROP_FPS)
fps = 18
videoWriter = cv2.VideoWriter('./record_video/out' + str(time.time()) + '.mp4', cv2.VideoWriter_fourcc(*'H264'),
fps, (resize_w, resize_h))
with self.mp_hands.Hands(min_detection_confidence=0.7,
min_tracking_confidence=0.5,
max_num_hands=2) as hands:
flag_empty = False
while cap.isOpened():
if remote == True:
if q.qsize() > 5:
flag_empty = False
if flag_empty != True:
# empty the queue when the init is done
print("empty")
while q.qsize() > 1:
q.get()
flag_empty = True
if q.empty() != True:
self.image = q.get()
if cv2.waitKey(1) & 0xFF == ord('q'):
break
else:
success, self.image = cap.read()
if not success:
print("Blank Frame")
continue
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# resize image
self.image = cv2.resize(self.image, (resize_w, resize_h))
# adjust according to camera position
# self.image = cv2.rotate( self.image, cv2.ROTATE_180)
# improve performance
self.image.flags.writeable = False
# BGR to RGB
self.image = cv2.cvtColor(self.image, cv2.COLOR_BGR2RGB)
# mirror image(adjust according to the camera)
# self.image = cv2.flip(self.image, 1)
# process image with mediapipe
results = hands.process(self.image)
self.image.flags.writeable = True
self.image = cv2.cvtColor(self.image, cv2.COLOR_RGB2BGR)
# save thumbnail
if isinstance(drawInfo.last_thumb_img, np.ndarray):
self.image = drawInfo.generateThumb(drawInfo.last_thumb_img, self.image)
hand_num = 0
# whether palms exists
if results.multi_hand_landmarks:
# record the index of hands
handedness_list = self.checkHandsIndex(results.multi_handedness)
hand_num = len(handedness_list)
drawInfo.hand_num = hand_num
# copy the original clean frame
frame_copy = self.image.copy()
# traverse all the palms
for hand_index, hand_landmarks in enumerate(results.multi_hand_landmarks):
# fault-tolerance
if hand_index > 1:
hand_index = 1
# draw finger skeleton
self.mp_drawing.draw_landmarks(
self.image,
hand_landmarks,
self.mp_hands.HAND_CONNECTIONS,
self.mp_drawing_styles.get_default_hand_landmarks_style(),
self.mp_drawing_styles.get_default_hand_connections_style())
# used for store finger coordiates
landmark_list = []
# coordinates used for store the moving range
paw_x_list = []
paw_y_list = []
for landmark_id, finger_axis in enumerate(
hand_landmarks.landmark):
landmark_list.append([
landmark_id, finger_axis.x, finger_axis.y,
finger_axis.z
])
paw_x_list.append(finger_axis.x)
paw_y_list.append(finger_axis.y)
if landmark_list:
# ratio resize
ratio_x_to_pixel = lambda x: math.ceil(x * resize_w)
ratio_y_to_pixel = lambda y: math.ceil(y * resize_h)
# top-left and bottom-right coordinates of palms
paw_left_top_x, paw_right_bottom_x = map(ratio_x_to_pixel,
[min(paw_x_list), max(paw_x_list)])
paw_left_top_y, paw_right_bottom_y = map(ratio_y_to_pixel,
[min(paw_y_list), max(paw_y_list)])
# get the coordinates of index finger
index_finger_tip = landmark_list[8]
index_finger_tip_x = ratio_x_to_pixel(index_finger_tip[1])
index_finger_tip_y = ratio_y_to_pixel(index_finger_tip[2])
# get the coordinates of middle finger
middle_finger_tip = landmark_list[12]
middle_finger_tip_x = ratio_x_to_pixel(middle_finger_tip[1])
middle_finger_tip_y = ratio_y_to_pixel(middle_finger_tip[2])
# draw coordinates
label_height = 30
label_wdith = 130
cv2.rectangle(self.image, (paw_left_top_x - 30, paw_left_top_y - label_height - 30),
(paw_left_top_x + label_wdith, paw_left_top_y - 30), (0, 139, 247), -1)
l_r_hand_text = handedness_list[hand_index][:1]
cv2.putText(self.image,
"{hand} x:{x} y:{y}".format(
hand = l_r_hand_text,
x = index_finger_tip_x,
y = index_finger_tip_y),
(paw_left_top_x - 30 + 10, paw_left_top_y - 40),
cv2.FONT_HERSHEY_PLAIN, 1, (242, 242, 230), 2)
# draw square frame for palms
cv2.rectangle(self.image, (paw_left_top_x - 30, paw_left_top_y - 30),
(paw_right_bottom_x + 30, paw_right_bottom_y + 30), (0, 139, 247), 1)
# release current mode
line_len = math.hypot((index_finger_tip_x - middle_finger_tip_x),
(index_finger_tip_y - middle_finger_tip_y))
if line_len < 50 and handedness_list[hand_index] == 'Right':
drawInfo.clear()
drawInfo.last_thumb_img = None
# transfer image to drawing class. start to draw circles when retention rime is longer than 0.3s
self.image = drawInfo.checkIndexFingerMove(handedness_list[hand_index],
[index_finger_tip_x, index_finger_tip_y],
self.image, frame_copy)
# display FPS and palms number
cTime = time.time()
fps_text = 1 / (cTime - fpsTime)
fpsTime = cTime
self.image = drawInfo.cv2AddText(self.image, "FPS: " + str(int(fps_text)),
(10, 10), textColor = (242, 242, 230), textSize = 35)
self.image = drawInfo.cv2AddText(self.image, "Palm num: " + str(hand_num),
(10, 45), textColor = (242, 242, 230), textSize = 35)
# display image
# self.image = cv2.resize(self.image, (resize_w//2, resize_h//2))
cv2.imshow('EDITH', self.image)
videoWriter.write(self.image)
if cv2.waitKey(5) & 0xFF == 27:
break
cap.release()
def test():
while True:
text = input("input:\t")
q_result.put(text)
if __name__ == '__main__':
control = VirtualFingerReader()
p1 = threading.Thread(target=Receive)
p2 = threading.Thread(target=voice_broadcast)
p3 = threading.Thread(target=test)
if remote == True:
p1.start()
p2.start()
# p3.start()
control.process()
else:
# p2.start()
control.recognize()