-
Notifications
You must be signed in to change notification settings - Fork 0
/
cellfun.py
523 lines (450 loc) · 23.8 KB
/
cellfun.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
import cv2
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import cm
from matplotlib.colors import ListedColormap
import math
import copy
import csv
from matplotlib import patches
import sys
import h5py
# Function to find circles from an mp4 video input file
def trackingcircles(filename,micronperpix,centerxy,timegrab):
cap = cv2.VideoCapture(filename)
fps = cap.get(cv2.CAP_PROP_FPS)
cap.set(cv2.CAP_PROP_POS_FRAMES, int(round((timegrab[0]*60+timegrab[1])*fps)))
success, image = cap.read()
"""
image2 = image.copy()
cv2.namedWindow("Select Circle", cv2.WINDOW_NORMAL)
cv2.setWindowProperty("Select Circle", cv2.WND_PROP_FULLSCREEN, cv2.WINDOW_FULLSCREEN)
bbox1 = cv2.selectROI("Select Circle", image2)
cv2.circle(image2, (int(round(bbox1[0]+bbox1[2]/2)), int(round(bbox1[1]+bbox1[3]/2))), int(round((bbox1[2]+bbox1[3])/4)), (0, 0, 255), 2)
bbox2 = cv2.selectROI("Select Circle", image2)
cv2.circle(image2, (int(round(bbox2[0]+bbox2[2]/2)),int(round(bbox2[1]+bbox2[3]/2))),int(round((bbox2[2]+bbox2[3])/4)),(0,0,255),2)
bbox3 = cv2.selectROI("Select Circle", image2)
cv2.circle(image2, (int(round(bbox3[0]+bbox3[2]/2)),int(round(bbox3[1]+bbox3[3]/2))),int(round((bbox3[2]+bbox3[3])/4)),(0,0,255),2)
cv2.imshow("Select Circle",image2)
circlevec = np.array([[bbox1[0]+bbox1[2]/2,bbox1[1]+bbox1[3]/2,(bbox1[2]+bbox1[3])/4],[bbox2[0]+bbox2[2]/2,bbox2[1]+bbox2[3]/2,(bbox2[2]+bbox2[3])/4],[bbox3[0]+bbox3[2]/2,bbox3[1]+bbox3[3]/2,(bbox3[2]+bbox3[3])/4]])
np.save("Resources/hexagonaltrajectories/CEMtrajectories/6-11-2021-movie1-0-/circlesvid.npy", circlevec)
print("pixels: ",bbox2[0]+bbox2[2]/2-(bbox3[0]+bbox3[2]/2))
print("center circle: ",bbox1[0]+bbox1[2]/2,bbox1[1]+bbox1[3]/2)
# Continue with the rest of the circles
circlevec = np.load("Resources/hexagonaltrajectories/CEMtrajectories/6-11-2021-movie1-0-/circlesvid.npy")
while True:
for (x, y, r) in circlevec:
cv2.circle(image, (int(round(x)), int(round(y))), int(round(r)), (0, 0, 255), 2)
cv2.namedWindow("Select Circle", cv2.WINDOW_NORMAL)
cv2.setWindowProperty("Select Circle", cv2.WND_PROP_FULLSCREEN, cv2.WINDOW_FULLSCREEN)
bbox = cv2.selectROI("Select Circle", image)
if bbox[2] < 5:
break
circlevec = np.append(circlevec,[[bbox[0]+bbox[2]/2,bbox[1]+bbox[3]/2,(bbox[2]+bbox[3])/4]],axis=0)
np.save("Resources/hexagonaltrajectories/CEMtrajectories/6-11-2021-movie1-0-/circlesvid.npy", circlevec)
"""
# Draw the actual circles on the figure
circlevec = []
#with open("Resources/overlaidtrajectories/postpositions_actual.csv", newline='') as csvfile:
with open("Resources/overlaidtrajectories/postpositions_actual.csv", newline='') as csvfile:
circlereader = csv.reader(csvfile)
counter = 0
for row in circlereader:
counter += 1
if counter == 1:
continue
circlevec.append([float(row[0]),float(row[1]),float(50)])
circlevecactual = np.array(circlevec)
# Load the post positions from the video geometry
circlevecvid = np.load("Resources/hexagonaltrajectories/CEMtrajectories/6-11-2021-movie1-31380-41191/circlesvid.npy")
#np.savetxt("Resources/hexagonaltrajectories/CEMtrajectories/6-11-2021-movie1-31380-41191/circlepositions.txt",circlevecvid[:,:2])
circlevecvid[:, 0] = circlevecvid[:, 0]*micronperpix#+(3262.33343116-centerxy[0]*micronperpix)
circlevecvid[:, 1] = (1080-circlevecvid[:, 1])*micronperpix#+(1683.58621655-(1080-centerxy[1])*micronperpix)
differences = np.zeros_like(circlevecvid)
for i in range(circlevecvid.shape[0]):
counter = 0
mindist = 10000
for j in range(circlevecactual.shape[0]):
dist = math.sqrt((circlevecvid[i,0]-circlevecactual[j,0])**2+(circlevecvid[i,1]-circlevecactual[j,1])**2)
if dist < mindist:
mindist = dist
index = counter
counter += 1
differences[i,:] = circlevecactual[index,:]-circlevecvid[i,:]
"""
# Run for loop over all posts and find three closest distances for Ding
with open("Resources/highlowExpression/quadpoints_database.csv", 'w', newline='') as csvfile:
database = csv.writer(csvfile)
database.writerow(["post x position","post y position","x1","y1","x2","y2","x3","y3","x4","y4"])
# For loop over all posts to start
#dx = np.zeros((circlevecactual.shape[0], circlevecactual.shape[0]))
#dy = np.zeros((circlevecactual.shape[0], circlevecactual.shape[0]))
box_len = 400
quad_data = np.zeros((circlevecactual.shape[0],10))
for i in range(circlevecactual.shape[0]):
# One for loop to save numpy array of six closest posts from index of dist matrix
dy_min1 = 10000
dx_min2 = 10000
dy_min3 = 10000
dx_min4 = 10000
idx1flag = True
idx2flag = True
idx3flag = True
idx4flag = True
for j in range(circlevecactual.shape[0]):
dx = circlevecactual[j,0]-circlevecactual[i,0]
dy = circlevecactual[j,1]-circlevecactual[i,1]
if i == j:
dx = 10000
dy = 10000
if dx > 0 and dx < box_len and abs(dy) < box_len:
idx1flag = False
if abs(dy) < dy_min1:
dy_min1 = abs(dy)
idx1 = j
if dy < 0 and abs(dy) < box_len and abs(dx) < box_len:
idx2flag = False
if abs(dx) < dx_min2:
dx_min2 = abs(dx)
idx2 = j
if dx < 0 and abs(dx) < box_len and abs(dy) < box_len:
idx3flag = False
if abs(dy) < dy_min3:
dy_min3 = abs(dy)
idx3 = j
if dy > 0 and dy < box_len and abs(dx) < box_len:
idx4flag = False
if abs(dx) < dx_min4:
dx_min4 = abs(dx)
idx4 = j
quad_data[i,0] = circlevecactual[i,0]
quad_data[i,1] = circlevecactual[i,1]
if idx1flag:
quad_data[i,2:4] = np.nan
else:
quad_data[i,2] = circlevecactual[idx1,0]
quad_data[i,3] = circlevecactual[idx1,1]
if idx2flag:
quad_data[i,4:6] = np.nan
else:
quad_data[i,4] = circlevecactual[idx2, 0]
quad_data[i,5] = circlevecactual[idx2, 1]
if idx3flag:
quad_data[i,6:8] = np.nan
else:
quad_data[i,6] = circlevecactual[idx3,0]
quad_data[i,7] = circlevecactual[idx3,1]
if idx4flag:
quad_data[i,8:10] = np.nan
else:
quad_data[i,8] = circlevecactual[idx4, 0]
quad_data[i,9] = circlevecactual[idx4, 1]
database.writerow(quad_data[i])
#print("i: ",i)
"""
return image,circlevecvid,differences
def drawcircles(ax,circlevecvid=False):
# Draw the actual circles on the figure
circlevec = []
with open("Resources/overlaidtrajectories/postpositions_actual.csv", newline='') as csvfile:
#with open("Resources/highlowExpression/circles_wholechip.csv", newline='') as csvfile:
circlereader = csv.reader(csvfile)
counter = 0
for row in circlereader:
counter += 1
if counter == 1:
continue
circlevec.append([float(row[0]), float(row[1]), float(50)])
circlevecactual = np.array(circlevec)
#ax.set_xlim((-20, np.amax(circlevecactual[:, 0])+70))
#ax.set_ylim((-20, np.amax(circlevecactual[:, 1])+70))
ax.set_xlim((600, 1550))
ax.set_ylim((370, 840))
for i in range(circlevecactual.shape[0]):
circle = plt.Circle((circlevecactual[i, 0], circlevecactual[i, 1]), 50, color='gray',alpha=0.4)
#ax.add_artist(circle)
# Draw the circles from the microscope video if the circlevecvid is passed
if np.any(circlevecvid):
for i in range(circlevecvid.shape[0]):
circle = plt.Circle((circlevecvid[i, 0], circlevecvid[i, 1]),radius=78/2, color='gray')
ax.add_artist(circle)
def drawtraj(ax, fig,trackids,data,ctclabel=False,color=False,cellvel=False,size=False,fluidvel=False,stream=False,save=False,images=False,sequence=False):
if fluidvel:
# This is to plot COMSOL velocities:
fig, ax = plt.subplots()
values = np.load("Resources/overlaidtrajectories/fluidvelocity.npy")
ax.imshow(values,vmin=0,vmax=np.nanmax(values),extent=[0.6819762549592183,6819.0805733372235,0.3554905234788679,3554.549744265201],cmap="bone",alpha=0.75)
cbar = fig.colorbar(cm.ScalarMappable(cmap="bone"),ax=ax,ticks=np.linspace(0, 1, 10),orientation="horizontal",pad=.05,aspect=80)
cbar.ax.set_xticklabels([int(round(num)) for num in np.linspace(0, np.nanmax(values), 10)])
#cbar.ax.set_ylabel("Fluid Velocities (COMSOL) [um]",rotation=270,labelpad=20)
if stream:
counter = 0
with open("Resources/overlaidtrajectories/streamlines.csv",newline='') as csvfile:
streamreader = csv.reader(csvfile)
for row in streamreader:
if counter%2 == 0:
x = np.array([])
for j in range(len(row)):
x = np.append(x,float(row[j]))
if (counter+1)%2 == 0:
y = np.array([])
for j in range(len(row)):
y = np.append(y, float(row[j]))
ax.plot(x,y,c="grey", lw=.5)
counter += 1
if size:
ax.scatter([50000], [50000], label="< "+str(round(size-2.5,2))+" um", color=color)
ax.scatter([50000], [50000], label=str(round(size,2))+" um", color=color)
ax.scatter([50000], [50000], label=str(round(size+2,2))+" um", color=color)
ax.scatter([50000], [50000], label="> "+str(round(size+4,2))+" um", color=color)
lgnd = ax.legend(loc="upper right")
lgnd.legendHandles[0]._sizes = [0.5]
lgnd.legendHandles[1]._sizes = [3]
lgnd.legendHandles[2]._sizes = [5]
lgnd.legendHandles[3]._sizes = [7]
max_height = 0
max_height_half = 0
numtracks = len(trackids)
X = []
for i in range(numtracks):
ID = trackids[i]
if cellvel:
if ID == 171 or ID == 168 or ID == 102 or ID == 191:
#ax.plot(data["x"+str(ID)], data["y"+str(ID)], c="gray", lw=.05)
#ax.scatter(data["x"+str(ID)][np.isnan(data["v_mag"+str(ID)])],data["y"+str(ID)][np.isnan(data["v_mag"+str(ID)])], s=3,c="gray")
#ax.scatter(data["x"+str(ID)], data["y"+str(ID)], s=8,c=data["v_mag"+str(ID)], cmap='jet', vmin=0,vmax=data["maxvmag"])
ax.scatter(data["x"+str(ID)],data["y"+str(ID)],s=70,c=data["v_mag"+str(ID)],cmap='jet',vmin=0,vmax=1400)
if color:
#ax.plot(data["x"+str(ID)], data["y"+str(ID)], c=color, lw=.05)
if size:
ax.scatter(data["x"+str(ID)],data["y"+str(ID)], s=data["diameter"+str(ID)], c=color)
else:
ax.scatter(data["x" + str(ID)], data["y" + str(ID)], s=5, c=color)
#data["y"+str(ID)][0] >= 1700 and data["y"+str(ID)][0] <= 1800:
#ax.text(data["x"+str(ID)][-1],data["y"+str(ID)][-1],str(ID))
#ax.set_aspect('equal')
#plt.show()
if sequence:
#plt.close()
#fig, ax = plt.subplots()
# Call drawcircles() function to get image of circles and circles vector
#drawcircles(ax)
# Only take data that have a velocity through boolean array indexing
x_data_full = data["x"+str(ID)][~np.isnan(data["v_mag"+str(ID)])]
y_data_full = data["y"+str(ID)][~np.isnan(data["v_mag"+str(ID)])]
postdist1_full = data["postdist1"+str(ID)][~np.isnan(data["v_mag"+str(ID)])]
postdist2_full = data["postdist2"+str(ID)][~np.isnan(data["v_mag"+str(ID)])]
v_data_full = np.clip(data["v_mag"+str(ID)][~np.isnan(data["v_mag"+str(ID)])],a_min=None,a_max=2400)
tot_length = x_data_full.shape[0]
stride = 15
window = 22
num_examples = int((tot_length-window)/stride)
# Initialize the trajectory images with the number of images as the first index
# Loop over each training example and each trajectory point
for n in range(num_examples):
# Take all x_data and y_data less than range of W_p
# Find the last "full trajectory" image then move to the end to get last image
x_i = []
x_data = x_data_full[n*stride:(n*stride+window)]
y_data = y_data_full[n*stride:(n*stride+window)]
v_data = v_data_full[n*stride:(n*stride+window)]
postdist1 = postdist1_full[n*stride:(n*stride+window)]
postdist2 = postdist2_full[n*stride:(n*stride+window)]
for j in range(len(x_data)):
x_i.append([x_data[j], y_data[j],postdist1[j],postdist2[j],v_data[j]])
X.append(x_i)
if images:
#plt.close()
#fig, ax = plt.subplots()
# Call drawcircles() function to get image of circles and circles vector
drawcircles(ax)
# Only take data that have a velocity through boolean array indexing
x_data_full = data["x"+str(ID)][~np.isnan(data["v_mag"+str(ID)])]
y_data_full = data["y"+str(ID)][~np.isnan(data["v_mag"+str(ID)])]
v_data_full = np.clip(data["v_mag"+str(ID)][~np.isnan(data["v_mag"+str(ID)])],a_min=None,a_max=2400)
# First define the ranges of x and y in micrometers (try 700 and 200 microns) with minxlength in object_tracking = 900 mu
W_mu = 700
H_mu = 233
AR = H_mu/ W_mu
# Next define the ranges of the x and y in pixels of trajectory images but keep aspect ratio the same
H_p = 40
W_p = int(H_p/AR)
print("W_p:", W_p)
# Use CNN pixel equation: n_W^[l] = (n_W^[l-1]+2p-f)/s, to find total number of images
stride = 250
stride = 400
traj_len = np.amax(x_data_full)-np.amin(x_data_full)
#num_images = int(math.ceil((traj_len-W_mu)/stride) + 1)
num_images = int((traj_len-W_mu)/stride)+1+1
# Initialize the trajectory images with the number of images as the first index
traj_image = np.zeros((num_images, H_p, W_p))
# Loop over each image and each trajectory point
for n in range(num_images):
# Take all x_data and y_data less than range of W_p
# Find the last "full trajectory" image then move to the end to get last image
if n == num_images - 1:
# Add one pixel length here to ensure no index is -1
x_start =np.amax(x_data_full)-W_mu+W_mu/W_p
if np.amax(x_data_full)-x_end < 80:
continue
else:
x_end = np.amax(x_data_full)
bool = (x_data_full > x_start) & (x_data_full <= x_end)
else:
x_start = np.amin(x_data_full)+n*stride
x_end = np.amin(x_data_full)+W_mu+n*stride
bool = (x_data_full >= x_start) & (x_data_full<x_end)
x_data = x_data_full[bool]
y_data = y_data_full[bool]
v_data = v_data_full[bool]
min_y = np.amin(y_data)
max_y = np.amax(y_data)
# Karl's Proposal: just center it with max and min in the y direction because the max_height is less than with centroid method
# Make sure y values are in range of H_di
if (np.amax(y_data) - np.amin(y_data)) >= H_mu:
#max_height = (np.amax(y_data) - np.amin(y_data))
sys.exit("Error: y_data is larger than H_di range")
# If there are not more than a certain amount of velocity points then continue
# In addition, if the spacing between any two points is greater than max_space then continue
if x_data.shape[0] <= 15:
continue
max_space = 120
dontsave = False
# If either of the two ends are more than max_space than continue
if (x_data[0]-x_start) > max_space or (x_end-x_data[-1]) > max_space:
continue
offset_y = (H_p-(max_y-min_y)*(H_p/H_mu))/2
for j in range(len(x_data)):
if j != 0 and (x_data[j]-x_data[j-1]) > max_space:
dontsave = True
break
if n == num_images - 1:
x = int((x_data[j]-(x_end-W_mu))*(W_p/W_mu))-1
else:
x = int((x_data[j]-x_start)*(W_p/W_mu))
y = int((y_data[j]-min_y)*(H_p/H_mu)+offset_y)
# No if statement necessary because I throw an error before if it is not in the range
# Don't forget to flip the fucking image now
y = H_p - y - 1
traj_image[n, y, x] = v_data[j]
# If there is too much space in between points, then continue with images
if dontsave:
continue
# Normalize the velocities and save images right after
traj_image[n,:,:] *= 255/np.amax(traj_image[n,:,:])
#cv2.imwrite(images+str(ID)+"_"+str(n)+".png",traj_image[n])
if n == num_images - 1:
rect = patches.Rectangle((np.amax(x_data_full)-W_mu+W_mu/W_p,min_y-(H_mu-(max_y-min_y))/2),W_mu,H_mu,fill=False,color="black",lw=0.8)
else:
rect = patches.Rectangle((np.amin(x_data_full)+n*stride,min_y-(H_mu-(max_y-min_y))/2),W_mu,H_mu,fill=False,color="black",lw=0.8)
ax.add_patch(rect)
# To plot the cell velocity colorbar on the bottom
"""
cbar = fig.colorbar(cm.ScalarMappable(cmap='jet'),ax=ax,ticks=np.linspace(0, 1, 11),pad=0.07,fraction=0.08,orientation="horizontal",aspect=80)
# cbar.ax.set_yticklabels([int(round(num)) for num in np.linspace(0, data["maxvmag"], 10)])
labels = []
for num in np.linspace(0, 2400 - 2400 / 10, 10):
labels.append(str(int(num)))
labels.append("> 2400")
cbar.ax.set_xticklabels(labels)
#cbar.ax.set_xlabel("Cell Velocities [um/s]", labelpad=10,fontweight="semibold")
"""
ax.scatter(x_data_full, y_data_full, s=30, c=v_data_full,cmap='jet',vmin=0,vmax=np.amax(v_data_full))
#plt.savefig(images+str(ID)+".png")
#ax.set_xticks(range(2460,3200,35))
#ax.set_yticks(range(2000,2400,30))
#ax.grid(b=True, which="both", axis="both",color="black",alpha=0.8)
#ax.set_xticks([])
#ax.set_yticks([])
ax.set_aspect('equal')
plt.show()
# To plot the velocity colorbar on the bottom
if i == numtracks - 1:
if cellvel:
cbar = fig.colorbar(cm.ScalarMappable(cmap='jet'),ax=ax,ticks=np.linspace(0, 1, 11),pad=.12,fraction=0.04,orientation="horizontal",aspect=80)
# cbar.ax.set_yticklabels([int(round(num)) for num in np.linspace(0, data["maxvmag"], 10)])
labels = []
for num in np.linspace(0, 1400 - 1400 / 10, 10):
labels.append(str(int(num)))
labels.append("1400")
cbar.ax.set_xticklabels(labels)
# cbar.ax.set_xlabel("Cell Velocities [um/s]", labelpad=10,fontweight="semibold")
# To plot a legend with labels
"""
if ctclabel == "SKBR3 Cells":
ax.scatter([50000], [50000], label="PC3 Cells", color="g")
ax.scatter([50000], [50000], label=ctclabel, color=color)
#lgnd = ax.legend(loc="upper right")
#lgnd.legendHandles[0]._sizes = [30]
#lgnd.legendHandles[1]._sizes = [30]
"""
if save:
# This is to write the trajectory data to a .csv file
with open(save, 'a', newline='') as csvfile:
trajectorywriter = csv.writer(csvfile)
trajectorywriter.writerow(np.array([ID]))
trajectorywriter.writerow(np.array([data["maxvmag"]]))
trajectorywriter.writerow(np.array([data["diameter"+str(ID)]]))
#trajectorywriter.writerow(data["postdist1"+str(ID)])
#trajectorywriter.writerow(data["postdist2"+str(ID)])
trajectorywriter.writerow(data["x"+str(ID)])
trajectorywriter.writerow(data["y"+str(ID)])
trajectorywriter.writerow(data["v_x"+str(ID)])
trajectorywriter.writerow(data["v_y"+str(ID)])
trajectorywriter.writerow(data["v_mag"+str(ID)])
trajectorywriter.writerow('')
#print("max_height: ", max_height)
#print("len(X)",len(X))
#with h5py.File("Resources/overlaidtrajectories/"+ctclabel+".h5", "w") as hdf:
#hdf.create_dataset("examples", data=np.array(X))
return ax
# Function to draw blacked out pixels
def blackout(frame,blackpoints):
x_1 = blackpoints[0, :]
y_1 = blackpoints[1, :]
x_2 = blackpoints[2, :]
y_2 = blackpoints[3, :]
for i in range(len(x_1)):
frame[y_1[i]:y_2[i], x_1[i]:x_2[i]] = 0
def drawline(count,cap):
xpoints = []
ypoints = []
while True:
# Capture frame-by-frame
cap.set(cv2.CAP_PROP_POS_FRAMES, count)
success, frame = cap.read()
cv2.namedWindow("Select Point", cv2.WINDOW_NORMAL)
cv2.setWindowProperty("Select Point", cv2.WND_PROP_FULLSCREEN, cv2.WINDOW_FULLSCREEN)
bbox1 = cv2.selectROI("Select Point", frame)
if bbox1[2] > 50:
cv2.destroyWindow("Select Point")
break
xpoints.append(bbox1[0]+bbox1[2]/2)
ypoints.append(bbox1[1]+bbox1[3]/2)
count += 1
drawframe = 255*np.ones_like(frame)
for i in range(len(xpoints)):
if i == 0:
continue
cv2.line(drawframe, (int(round(xpoints[i-1])), int(round(ypoints[i-1]))), (int(round(xpoints[i])), int(round(ypoints[i]))), (0,0,0), 2)
return drawframe[:,:,0]
# Function to draw bounding box and trajectory for each image
def imdraw(img2, bbox, x_pos, y_pos):
x,y,w,h = int(bbox[0]),int(bbox[1]),int(bbox[2]),int(bbox[3])
cv2.rectangle(img2, (x,y), ((x+w), (y+h)), (255,0,0), 3, 1)
for i in range(len(x_pos)):
img2[int(round(y_pos[i])), int(round(x_pos[i])), :] = [0, 0, 255]
#img2[int(round(y_pos[i]) + 1), int(round(x_pos[i])), :] = [0, 0, 255]
#img2[int(round(y_pos[i]) - 1), int(round(x_pos[i])), :] = [0, 0, 255]
#img2[int(round(y_pos[i])), int(round(x_pos[i]) + 1), :] = [0, 0, 255]
#img2[int(round(y_pos[i])), int(round(x_pos[i]) - 1), :] = [0, 0, 255]
cv2.putText(img2, "Tracking", (850, 50), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 0), 2)
return img2
# Function to draw bounding boxes and trajectories for each image
def imdraw2(img2, bboxes):
xyposition = np.zeros((bboxes.shape[0], 2))
for m in range(bboxes.shape[0]):
(x, y, w, h) = [bboxes[m,0], bboxes[m,1], bboxes[m,2], bboxes[m,3]]
xyposition[m, :] = [(x+x+w)/2, (y+y+h)/2]
cv2.rectangle(img2, (int(round(x)), int(round(y))), (int(round(x + w)), int(round(y + h))), (0, 255, 0), 2)
return img2, xyposition