forked from ys388/MCGPU
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathfan_beam_reconstruction.py
176 lines (145 loc) · 6.17 KB
/
fan_beam_reconstruction.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
import codecs
import numpy as np
import matplotlib.pyplot as plt
import os
from PIL import Image
from reconstruct_metrics import Sinogram2Img
from skimage.metrics import structural_similarity as ssim
from skimage.metrics import peak_signal_noise_ratio as psnr
class init:
def __init__(self):
self.param = {}
# image
self.param['nx'] = 340
self.param['ny'] = 340
self.param['image_pixels'] = 512
self.param['sod'] = 600
self.param['odd'] = 300
self.param['views'] = 360
self.param['nbins'] = 640
self.param['z'] = 32
self.param['detector_length'] = 800
## detector
self.param['u_water'] = 0.192 #0.0205
self.param['angles'] = np.linspace(0, 2 * np.pi, self.param['views'], False)
self.param['ratio'] = self.param['image_pixels'] / self.param['nx']
self.param['du'] = self.param['detector_length']/self.param['nbins']
def reshape_(f,shape,z):
line = f.readline() # 以行的形式进行读取文件
all = [] # 设置x y z数组
scat_1 = []
scat_2 = []
scat_3 = []
no_scat = []
while line:
if line[0] != '#' and line != '\n':
line_temp = line.split(' ') # 每行数据分隔情况，此数据以“,”分隔
no_scatter = float(line_temp[0])
scatter_1 = float(line_temp[1])
scatter_2 = float(line_temp[2])
scatter_3 = float(line_temp[3])
no_scat.append(no_scatter)
scat_1.append(scatter_1) # 将其添加在列表之中
scat_2.append(scatter_2)
scat_3.append(scatter_3)
all.append(no_scatter + scatter_1 + scatter_2 + scatter_3)
line = f.readline()
f.close() # close文件 # 对获取的txt前两列数据进行保存
no_scat_np = np.array(no_scat)
scat_1_np = np.array(scat_1)
scat_2_np = np.array(scat_2)
scat_3_np = np.array(scat_3)
all_np = np.array(all)
no_scat_np = no_scat_np.reshape(z, shape)
scat_1_np = scat_1_np.reshape(z, shape)
scat_2_np = scat_2_np.reshape(z, shape)
scat_3_np = scat_3_np.reshape(z, shape)
all_np = all_np.reshape(z, shape)
return no_scat_np,all_np
def npy2tif_save(npy_,path):
tif_ = Image.fromarray(npy_)
tif_.save(path)
def air2energy(path_ori_air,param):
detectors, views = param.param['nbins'], param.param['views']
z = param.param['z']
get_line = int(z / 2)
air_file = codecs.open(path_ori_air, mode='r')
file_no, file_all = reshape_(air_file, detectors, z)
air_line = file_all[get_line, :]
return air_line
def trunc(npy_, w_l, w_r):
npy_[npy_ < w_l] = w_l
npy_[npy_ > w_r] = w_r
return npy_
def raw2energy(path_ori, energy_no_path, energy_all_path, signal_sim, param,number_point = 6):
detectors, views = param.param['nbins'], param.param['views']
z = param.param['z']
get_line = int(z/2)
energy_map_no = np.zeros((views, detectors), dtype=np.float64)
energy_map_all = np.zeros((views, detectors), dtype=np.float64)
##取出air_file中间一行
for path in os.listdir(path_ori):
if 'raw' in path:
os.remove(os.path.join(path_ori, path))
paths = sorted([path for path in os.listdir(path_ori) if signal_sim in path])
paths.sort(key=lambda x: int(x[number_point:]))
for i in range(len(paths)):
num = int(i) ##720 views
# num = int(i/2) ##360 views
path = paths[i]
path_ = os.path.join(path_ori, path)
mat_file = codecs.open(path_, mode='r')
file_no, file_all = reshape_(mat_file, detectors, z)
# plt.imshow(file_all,'gray')
# plt.show()
line_no = file_no[get_line, :]
line_all = file_all[get_line, :] # mc_dat0....
# num = int(path[9:])
energy_map_no[num, :] = line_no
energy_map_all[num, :] = line_all
npy2tif_save(energy_map_no,energy_no_path)
npy2tif_save(energy_map_all, energy_all_path)
return energy_map_no, energy_map_all
def energy2sino(energy_map, air_line, param):
detectors, views = param.param['nbins'], param.param['views']
sino_map = np.zeros((views, detectors), dtype=np.float64)
for i in range(energy_map.shape[0]):
for j in range(energy_map.shape[1]):
sino_map[i][j] = -np.log(energy_map[i][j] / air_line[j])
return sino_map
def show_fig(ct_no, ct_all,w_l,w_r):
f, ax = plt.subplots(1, 2, figsize=(20, 10))
ax[0].imshow(ct_no, cmap=plt.cm.gray, vmin=w_l, vmax=w_r)
ax[0].set_title('Scatter-free', fontsize=20)
ax[1].imshow(ct_all, cmap=plt.cm.gray, vmin=w_l, vmax=w_r)
ax[1].set_title('Scatter-contaminated', fontsize=20)
plt.xticks([])
plt.yticks([])
plt.axis('off')
plt.show()
plt.savefig('compare.png')
if __name__ == '__main__':
path_ori = 'D:/CT_Project/lung_data/mayo_simu/mayo_views_13/'
path_ori_air = path_ori + 'air.dat'
os.makedirs(path_ori + 'test_save', exist_ok='True')
energy_no_path = path_ori + 'test_save/energy_no.tif'
energy_all_path = path_ori + 'test_save/energy_all.tif'
ct_no_path = path_ori + 'test_save/ct_no.tif'
ct_all_path = path_ori + 'test_save/ct_all.tif'
signal_sim = 'mc' ###用于查找MC文件的标记
w_l, w_r = 0.002, 0.030
param = init()
load_flag = False ###是否需要重新加载数据
### ct和energy的numpy文件
energy_map_no, energy_map_all = raw2energy(path_ori, energy_no_path, energy_all_path, signal_sim, param)
# energy_map_no = np.load(energy_no_path)
# energy_map_all = np.load(energy_all_path)
air_line = air2energy(path_ori_air, param)
sino_map_no = energy2sino(energy_map_no, air_line, param)
sino_map_all = energy2sino(energy_map_all, air_line, param)
ct_map_no = Sinogram2Img(sino_map_no, param)
ct_map_all = Sinogram2Img(sino_map_all, param)
ct_trunc_no, ct_trunc_all = trunc(ct_map_no, w_l, w_r), trunc(ct_map_all, w_l, w_r)
npy2tif_save(ct_trunc_no, ct_no_path)
npy2tif_save(ct_trunc_all, ct_all_path)
show_fig(ct_trunc_no, ct_trunc_all, w_l, w_r)