-
Notifications
You must be signed in to change notification settings - Fork 0
/
gf.py
184 lines (142 loc) · 5.83 KB
/
gf.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
import numpy as np
import scipy as sp
import scipy.ndimage
import imageio
def box_filter(img, r):
"""Apply a box filter to the image.
Args:
img (ndarray): Input image.
r (int): Radius of the box filter.
Returns:
ndarray: Filtered image.
"""
(rows, cols) = img.shape[:2]
im_dst = np.zeros_like(img)
tile = [1] * img.ndim
tile[0] = r
im_cum = np.cumsum(img, 0)
im_dst[0:r + 1, :, ...] = im_cum[r:2 * r + 1, :, ...]
im_dst[r + 1:rows - r, :, ...] = im_cum[2 * r + 1:rows, :, ...] - im_cum[0:rows - 2 * r - 1, :, ...]
im_dst[rows - r:rows, :, ...] = np.tile(im_cum[rows - 1:rows, :, ...], tile) - im_cum[ rows - 2 * r - 1:rows - r - 1, :, ...]
tile = [1] * img.ndim
tile[1] = r
im_cum = np.cumsum(im_dst, 1)
im_dst[:, 0:r + 1, ...] = im_cum[:, r:2 * r + 1, ...]
im_dst[:, r + 1:cols - r, ...] = im_cum[:, 2 * r + 1:cols, ...] - im_cum[:, 0:cols - 2 * r - 1, ...]
im_dst[:, cols - r: cols, ...] = np.tile(im_cum[:, cols - 1:cols, ...], tile) - im_cum[ :, cols - 2 * r - 1:cols - r - 1, ...]
return im_dst
def guided_filter(I, p, r, eps, s=None):
"""Guided filter for image dehazing.
Args:
I (ndarray): Guide image.
p (ndarray): Filtering input.
r (int): Window radius.
eps (float): Regularization parameter.
s (int): Subsampling factor for fast guided filter.
Returns:
ndarray: Filtered output.
"""
if p.ndim == 2:
p3 = p[:, :, np.newaxis]
else:
p3 = p
out = np.zeros_like(p3)
for ch in range(p3.shape[2]):
out[:, :, ch] = _gf_colorgray(I, p3[:, :, ch], r, eps, s)
return np.squeeze(out) if p.ndim == 2 else out
def _gf_colorgray(I, p, r, eps, s=None):
"""Automatically choose color or gray guided filter based on image dimensions."""
if I.ndim == 2 or I.shape[2] == 1:
return _gf_gray(I, p, r, eps, s)
elif I.ndim == 3 and I.shape[2] == 3:
return _gf_color(I, p, r, eps, s)
else:
raise ValueError("Invalid guide dimensions:", I.shape)
def _gf_gray(I, p, r, eps, s=None):
"""Grayscale guided filter."""
if s is not None:
Isub = sp.ndimage.zoom(I, 1 / s, order=1)
Psub = sp.ndimage.zoom(p, 1 / s, order=1)
r = round(r / s)
else:
Isub = I
Psub = p
(rows, cols) = Isub.shape
N = box_filter(np.ones([rows, cols]), r)
meanI = box_filter(Isub, r) / N
meanP = box_filter(Psub, r) / N
corrI = box_filter(Isub * Isub, r) / N
corrIp = box_filter(Isub * Psub, r) / N
varI = corrI - meanI * meanI
covIp = corrIp - meanI * meanP
a = covIp / (varI + eps)
b = meanP - a * meanI
meanA = box_filter(a, r) / N
meanB = box_filter(b, r) / N
if s is not None:
meanA = sp.ndimage.zoom(meanA, s, order=1)
meanB = sp.ndimage.zoom(meanB, s, order=1)
q = meanA * I + meanB
return q
def _gf_color(I, p, r, eps, s=None):
"""Color guided filter."""
fullI = I
fullP = p
if s is not None:
I = sp.ndimage.zoom(fullI, [1 / s, 1 / s, 1], order=1)
p = sp.ndimage.zoom(fullP, [1 / s, 1 / s], order=1)
r = round(r / s)
h, w = p.shape[:2]
N = box_filter(np.ones((h, w)), r)
mI_r = box_filter(I[:, :, 0], r) / N
mI_g = box_filter(I[:, :, 1], r) / N
mI_b = box_filter(I[:, :, 2], r) / N
mP = box_filter(p, r) / N
mIp_r = box_filter(I[:, :, 0] * p, r) / N
mIp_g = box_filter(I[:, :, 1] * p, r) / N
mIp_b = box_filter(I[:, :, 2] * p, r) / N
covIp_r = mIp_r - mI_r * mP
covIp_g = mIp_g - mI_g * mP
covIp_b = mIp_b - mI_b * mP
var_I_rr = box_filter(I[:, :, 0] * I[:, :, 0], r) / N - mI_r * mI_r
var_I_rg = box_filter(I[:, :, 0] * I[:, :, 1], r) / N - mI_r * mI_g
var_I_rb = box_filter(I[:, :, 0] * I[:, :, 2], r) / N - mI_r * mI_b
var_I_gg = box_filter(I[:, :, 1] * I[:, :, 1], r) / N - mI_g * mI_g
var_I_gb = box_filter(I[:, :, 1] * I[:, :, 2], r) / N - mI_g * mI_b
var_I_bb = box_filter(I[:, :, 2] * I[:, :, 2], r) / N - mI_b * mI_b
a = np.zeros((h, w, 3))
for i in range(h):
for j in range(w):
sig = np.array([
[var_I_rr[i, j], var_I_rg[i, j], var_I_rb[i, j]],
[var_I_rg[i, j], var_I_gg[i, j], var_I_gb[i, j]],
[var_I_rb[i, j], var_I_gb[i, j], var_I_bb[i, j]]
])
covIp = np.array([covIp_r[i, j], covIp_g[i, j], covIp_b[i, j]])
a[i, j, :] = np.linalg.solve(sig + eps * np.eye(3), covIp)
b = mP - a[:, :, 0] * mI_r - a[:, :, 1] * mI_g - a[:, :, 2] * mI_b
meanA = box_filter(a, r) / N[..., np.newaxis]
meanB = box_filter(b, r) / N
if s is not None:
meanA = sp.ndimage.zoom(meanA, [s, s, 1], order=1)
meanB = sp.ndimage.zoom(meanB, [s, s], order=1)
q = np.sum(meanA * fullI, axis=2) + meanB
return q
def test_gf():
"""Test the guided filter."""
cat = imageio.imread('cat.bmp').astype(np.float32) / 255
tulips = imageio.imread('tulips.bmp').astype(np.float32) / 255
r = 8
eps = 0.05
cat_smoothed = guided_filter(cat, cat, r, eps)
cat_smoothed_s4 = guided_filter(cat, cat, r, eps, s=4)
imageio.imwrite('cat_smoothed.png', cat_smoothed)
imageio.imwrite('cat_smoothed_s4.png', cat_smoothed_s4)
tulips_smoothed4s = np.zeros_like(tulips)
for i in range(3):
tulips_smoothed4s[:, :, i] = guided_filter(tulips, tulips[:, :, i], r, eps, s=4)
imageio.imwrite('tulips_smoothed4s.png', tulips_smoothed4s)
tulips_smoothed = np.zeros_like(tulips)
for i in range(3):
tulips_smoothed[:, :, i] = guided_filter(tulips, tulips[:, :, i], r, eps)
imageio.imwrite('tulips_smoothed.png', tulips_smoothed)