-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy patha9_test.py
132 lines (101 loc) · 2.99 KB
/
a9_test.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
import cv2
import a9
import numpy as np
class ImageIO:
def __init__(self, path:str) -> None:
self.path = path
def imread(self,fname:str)->np.ndarray:
im = cv2.imread(self.path+'Input/'+fname)
return np.array(im, dtype='float')/255
def imwrite(self,im:np.ndarray, fname:str)->None:
cv2.imwrite(self.path+'Output/'+fname, 255*im)
io = ImageIO('/Users/kartikeshmishra/Kartikesh/68370/a10/asst/')
def test_grad_descent():
im=io.imread('pru.png')
kernel=a9.gauss2D(1)
im_blur=a9.convolve3(im, kernel)
io.imwrite(im_blur, 'pru_blur.png')
im_sharp=a9.deconvGradDescent(im_blur, kernel)
io.imwrite(im_sharp, 'pru_sharp.png')
def test_conjugate_grad_descent():
im=io.imread('pru.png')
kernel=a9.gauss2D(1)
im_blur=a9.convolve3(im, kernel)
io.imwrite(im_blur, 'pru_blur.png')
im_sharp=a9.deconvCG(im_blur, kernel,20)
io.imwrite(im_sharp, 'pru_sharp_CG.png')
def test_real_psf():
im=io.imread('pru.png')
f=open('psf', 'r')
psf=[map(float, line.split(',')) for line in f ]
kernel=np.array(psf)
im_blur=a9.convolve3(im, kernel)
#kernel=kernel[::-1, ::-1]
io.imwrite(im_blur, 'pru_blur_real.png')
io.imwriteGrey(kernel/np.max(kernel), 'psf.png')
im_sharp=a9.deconvCG(im_blur, kernel, 4)
io.imwrite(im_sharp, 'pru_sharp_CG_real.png')
def test_conjugate_grad_descent_reg():
im=io.imread('pru.png')
kernel=a9.gauss2D(1)
im_blur=a9.convolve3(im, kernel)
noise=np.random.random(im_blur.shape)-0.5
im_blur_noisy=im_blur+0.05*noise
io.imwrite(im_blur_noisy, 'pru_blur_noise.png')
im_sharp=a9.deconvCG_reg(im_blur_noisy, kernel,10)
im_sharp_wo_reg=a9.deconvCG(im_blur_noisy, kernel,20)
io.imwrite(im_sharp, 'pru_sharp_CG_reg.png')
io.imwrite(im_sharp_wo_reg, 'pru_sharp_CG_wo_reg.png')
def test_naive_composite():
fg=io.imread('bear.png')
bg=io.imread('waterpool.png')
mask=io.imread('mask.png')
out=a9.naiveComposite(bg, fg, mask, 50, 1)
io.imwrite(out, 'naive_composite.png')
def test_Poisson():
y=50
x=10
fg=io.imread('bear.png')
bgn=io.imread('waterpool.png')
mask=io.imread('mask.png')
mask[mask>0.5]=1.0
mask[mask<0.6]=0.0
mask = a9.duplicate(mask)
bg = bgn[50:210,10:303]
bg[bg==0]=1e-4
fg[fg==0]=1e-4
bg=np.log(bg)+3
fg=np.log(fg)+3
tmp=a9.Poisson(bg, fg, mask)
w ,h , _ = fg.shape
out = np.array(bgn)
out[y:y+h, x:x+w]=np.exp(tmp-3)
out[y:y+h, x:x+w]=tmp
io.imwrite(out, 'poisson.png')
def test_PoissonCG():
y=50
x=10
fg=io.imread('bear.png')
bgn=io.imread('waterpool.png')
mask=io.imread('mask.png')
mask[mask>0.5]=1.0
mask[mask<0.6]=0.0
mask = a9.duplicate(mask)
bg = bgn[50:210,10:303]
bg[bg==0]=1e-4
fg[fg==0]=1e-4
bg=np.log(bg)+3
fg=np.log(fg)+3
tmp=a9.PoissonCG(bg, fg, mask)
w ,h , _ = fg.shape
out = np.array(bgn)
out[y:y+h, x:x+w]=np.exp(tmp-3)
out[y:y+h, x:x+w]=tmp
io.imwrite(out, 'poisson_CG.png')
# test_grad_descent()
# test_conjugate_grad_descent()
# test_real_psf()
test_conjugate_grad_descent_reg()
test_naive_composite()
# test_Poisson()
# test_PoissonCG()