-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain_v5.py
198 lines (154 loc) · 6.49 KB
/
main_v5.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
'''
VIDEO MAGIC V5
This version is using Delta E to calculate the difference between the background and the background with actor.
Quality improved a lot! Now it is getting awesaome .. I already feel the beach
'''
import cv2
import numpy as np
import colour
BACKGROUND_IMAGE = "VideoMagic/beach.jpg"
#BACKGROUND_IMAGE = "VideoMagic/space.jpg"
THRESHOLD = 25
BLUR_FACTOR = 40
SHARP_FACTOR = 0.8
SHOW_MIXED = 0
SHOW_CAM = 1
SHOW_BACKGROUND = 2
SHOW_MASK = 3
SHOW_MASK_NOISY = 4
SHOW_MASK_BLURRED = 5
FONT = cv2.FONT_HERSHEY_SIMPLEX
FONT_SCALE = 0.5
FONT_COLOR = (256, 0, 0)
FONT_COLOR_BW = (0, 0, 0)
FONT_THICKNESS = 2
def main(display_mode = SHOW_MIXED, treshold=THRESHOLD, blur_factor=BLUR_FACTOR, sharp_factor=SHARP_FACTOR):
''' main
'''
cap = cv2.VideoCapture(0)
#####################################################################################
# PREPARATION
#####################################################################################
# capture background from cam
while True:
# capture from cam
_, background_image_cam = cap.read()
# place text on image
txt = "GET OUT OF THE SCREEN TO CAPTURE THE BACKGROUND THEN PRESS 'SPACE'"
background_image_cam_display = np.array (background_image_cam)
background_image_cam_display = cv2.putText(background_image_cam_display, txt, (5, 25), FONT, FONT_SCALE, FONT_COLOR)
# display cam image
cv2.imshow('video', background_image_cam_display)
# wait for user input
if cv2.waitKey(1) == ord(' '):
break
height, width, _ = background_image_cam.shape
# load new background image and prepare it (esp. scale size)
background_image_new = cv2.imread(BACKGROUND_IMAGE)
b_height, b_width, _ = background_image_new.shape
scale = width/b_width
padding = int((height - (b_height*scale))/2)
background_image_new = cv2.resize(background_image_new, (width, int(b_height*scale)))
background_image_new = cv2.copyMakeBorder(background_image_new, padding+1, padding+1, 0, 0, borderType=cv2.BORDER_CONSTANT, value=0)
# capture video and replace old background with new background
while True:
#####################################################################################
# IMAGE MAGIC
#####################################################################################
_, cam_image = cap.read()
# calcuate the image_mask = difference between the current cam image and the captured background image of the cam based on Delta-E Color Difference
# https://www.colour-science.org/
f = lambda pix: 1.0 if pix <= treshold else 0.0
vectorized_f = np.vectorize(f)
image_diff = abs(colour.delta_E(background_image_cam, cam_image))
image_mask_noisy = vectorized_f(image_diff)
# blur the image mask to reduce noise
if blur_factor > 0:
image_mask_blurred = cv2.blur(image_mask_noisy, (blur_factor, blur_factor))
else:
image_mask_blurred = image_mask_noisy
# generate sharper mask from blurred mask
f = lambda pix: 1.0 if pix > sharp_factor else 0.0
vectorized_f = np.vectorize(f)
image_mask = vectorized_f(image_mask_blurred)
#####################################################################################
# DISPLAY MODE
#####################################################################################
if display_mode == SHOW_CAM:
vis = np.array(cam_image)
elif display_mode == SHOW_BACKGROUND:
vis = np.array(background_image_new)
elif display_mode == SHOW_MASK:
vis = np.array(image_mask)
elif display_mode == SHOW_MASK_NOISY:
vis = np.array(image_mask_noisy)
elif display_mode == SHOW_MASK_BLURRED:
vis = np.array(image_mask_blurred)
else:
vis = cv2.copyTo(background_image_new, image_mask.astype(cam_image.dtype), np.array(cam_image))
# a little bluring magic dust ;-)
vis = cv2.blur(vis, (2, 2))
# place text on image
font_color = FONT_COLOR
if display_mode == SHOW_MASK or display_mode == SHOW_MASK_BLURRED or display_mode == SHOW_MASK_NOISY:
font_color = FONT_COLOR_BW
txt = "THRESHOLD "+str(treshold)
cv2.putText(vis, txt, (5, 25), FONT, FONT_SCALE, font_color)
txt = "BLUR FACTOR "+str(blur_factor)
cv2.putText(vis, txt, (5, 45), FONT, FONT_SCALE, font_color)
txt = "SHARPNESS "+str(sharp_factor)
cv2.putText(vis, txt, (5, 65), FONT, FONT_SCALE, font_color)
# display modified image
cv2.imshow('video', vis)
#####################################################################################
# COMMANDS
#####################################################################################
k = cv2.waitKey(1)
if k == -1:
continue
elif k == 27: # ESC
break
elif k == ord('+'):
treshold = treshold + 5
elif k == ord('-') and treshold >= 5:
treshold = treshold - 5
elif k == ord('b'):
blur_factor = blur_factor + 5
elif k == ord('n') and blur_factor >= 5:
blur_factor = blur_factor - 5
elif k == ord('s') and sharp_factor <= 0.9:
sharp_factor = sharp_factor + 0.1
elif k == ord('d') and sharp_factor >= 0.1:
sharp_factor = sharp_factor - 0.1
elif k == ord(" "):
display_mode = SHOW_MIXED
elif k == ord("0"):
display_mode = SHOW_CAM
elif k == ord("1"):
display_mode = SHOW_MASK_NOISY
elif k == ord("2"):
display_mode = SHOW_MASK_BLURRED
elif k == ord("3"):
display_mode = SHOW_MASK
else:
print ("Pressed:", k) # else print its value
cap.release()
cv2.destroyAllWindows()
###############################################################
print ('''
HOW-TO
==========================================================
Change display mode:
SPACE: mixed mode
1: noisy image
2: blurred image
3: sharp image mask
0: original cam
----------------------------------------------------------
+/-: threshold
b/n: blur factor
s/d: sharpness
==========================================================
ESC: Quit Video Magic
''')
main()