-
Notifications
You must be signed in to change notification settings - Fork 3
/
imagemorph.py
159 lines (127 loc) · 4.75 KB
/
imagemorph.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
import sys
import math
import random
import numpy as np
from PIL import Image
class Pixel:
def __init__(self, r, g, b):
self.r = r
self.g = g
self.b = b
def read_img(filename):
image = Image.open(filename)
pixels = list(image.getdata())
global w, h
w, h = image.size[0], image.size[1]
print(image.size)
return [[Pixel(r, g, b) for (r, g, b) in pixels[i * w:(i + 1) * w]] for i in range(h)]
def write_img(image, filename):
pixels = [tuple([pixel.r, pixel.g, pixel.b]) for row in image for pixel in row]
img = Image.new("RGB", (w, h))
img.putdata(pixels)
img.save(filename, quality=100)
def compute_displacement_field(amp, sigma):
d_x = [[0.0 for _ in range(w)] for _ in range(h)]
d_y = [[0.0 for _ in range(w)] for _ in range(h)]
da_x = [[0.0 for _ in range(w)] for _ in range(h)]
da_y = [[0.0 for _ in range(w)] for _ in range(h)]
kws = int(2.0 * sigma)
ker = [math.exp(-float(k * k) / (sigma * sigma)) for k in range(-kws, kws + 1)]
for i in range(h):
for j in range(w):
d_x[i][j] = -1.0 + 2.0 * random.random()
d_y[i][j] = -1.0 + 2.0 * random.random()
for i in range(h):
for j in range(w):
sum_x = 0.0
sum_y = 0.0
for k in range(-kws, kws + 1):
v = j + k
if v < 0:
v = -v
if v >= w:
v = 2 * w - v - 1
sum_x += d_x[i][v] * ker[abs(k)]
sum_y += d_y[i][v] * ker[abs(k)]
da_x[i][j] = sum_x
da_y[i][j] = sum_y
for j in range(w):
for i in range(h):
sum_x = 0.0
sum_y = 0.0
for k in range(-kws, kws + 1):
u = i + k
if u < 0:
u = -u
if u >= h:
u = 2 * h - u - 1
sum_x += da_x[u][j] * ker[abs(k)]
sum_y += da_y[u][j] * ker[abs(k)]
d_x[i][j] = sum_x
d_y[i][j] = sum_y
avg = sum(math.sqrt(d_x[i][j] ** 2 + d_y[i][j] ** 2) for i in range(h) for j in range(w)) / (h * w)
for i in range(h):
for j in range(w):
d_x[i][j] = amp * d_x[i][j] / avg
d_y[i][j] = amp * d_y[i][j] / avg
return d_x, d_y
def apply_displacement_field(input, d_x, d_y):
output = [[Pixel(0, 0, 0) for _ in range(w)] for _ in range(h)]
for i in range(h):
for j in range(w):
p1 = i + d_y[i][j]
p2 = j + d_x[i][j]
u0 = int(math.floor(p1))
v0 = int(math.floor(p2))
f1 = p1 - u0
f2 = p2 - v0
sumr, sumg, sumb = 0.0, 0.0, 0.0
for idx in range(4):
if idx == 0:
u, v = u0, v0
f = (1.0 - f1) * (1.0 - f2)
elif idx == 1:
u, v = u0 + 1, v0
f = f1 * (1.0 - f2)
elif idx == 2:
u, v = u0, v0 + 1
f = (1.0 - f1) * f2
else:
u, v = u0 + 1, v0 + 1
f = f1 * f2
u = max(0, min(u, h - 1))
v = max(0, min(v, w - 1))
val = input[u][v].r
sumr += f * val
val = input[u][v].g
sumg += f * val
val = input[u][v].b
sumb += f * val
output[i][j].r = int(sumr)
output[i][j].g = int(sumg)
output[i][j].b = int(sumb)
return output
def rubbersheet(input, amp, sigma):
if sigma > h / 2.5 or sigma > w / 2.5:
sys.stderr.write("- Warning: Gaussian smoothing kernel too large for the input image.\n")
return
if sigma < 1E-5:
sys.stderr.write("- Warning: Gaussian smoothing kernel with negative/zero spread.\n")
return
d_x, d_y = compute_displacement_field(amp, sigma)
output = apply_displacement_field(input, d_x, d_y)
return output
if __name__ == "__main__":
# Process command line arguments
# Example run: python3 imagemorph.py test.jpg 2.0 50.0 output.jpg
if len(sys.argv) != 5:
sys.stderr.write(f"Program usage: {sys.argv[0]} [path_to_image] [displacement] [smoothing radius] [output_file]\n")
sys.exit(1)
path_to_image = sys.argv[1]
amp = float(sys.argv[2])
sigma = float(sys.argv[3])
output_file = sys.argv[4]
input_image = read_img(path_to_image)
output = rubbersheet(input_image, amp, sigma)
write_img(output, 'output.jpg')
print(f'Output image saved to {output_file}')