-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathFitcir_diameter_multiCPU.py
169 lines (163 loc) · 6.56 KB
/
Fitcir_diameter_multiCPU.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
'''
@author: xiaoyangliu
'''
from skimage import io
import numpy as np
import matplotlib.pyplot as plt
import glob
import os
from scipy import ndimage
import pandas as pd
import time
from itertools import product
from joblib import Parallel, delayed
#start = time.time()
def circle(cen, r, size):
row, col = size[0], size[1]
x = np.arange(col)
y = np.arange(row)
x,y = np.meshgrid(x,y)
img_circle = np.zeros([row, col],dtype=bool)
img_circle[(x-cen[1])**2 + (y-cen[0])**2 < r**2] = True
#img_circle = np.float32(img_circle)
return img_circle
# -----------------------------------------------------improved method---------------------------------------------------------------
def fitcir(img, xy_shift, xy_step, r0, r_shift_left, r_shift_right, r_step):
img = img.astype(bool)
img_true = np.count_nonzero(img==True)
size = img.shape
mass_cen = ndimage.measurements.center_of_mass(img)
y = np.arange(mass_cen[0]-xy_shift,mass_cen[0]+xy_shift, xy_step) #original method
x = np.arange(mass_cen[1]-xy_shift,mass_cen[1]+xy_shift, xy_step)#original method
#r = np.arange(r0-r_shift, r0+r_shift, r_step)
r_l = np.arange(r0-r_shift_left,r0+r_shift_right,r_step*2)
#all_potential = list(product(r,y,x))
all_potential_l = list(product(r_l,y,x))
xy_p = list(product(y,x))
#cross = np.zeros([len(y),len(x),len(r)])
#r_diff = np.zeros([len(y),len(x)])
for i in np.arange(0,len(all_potential_l),1):
img_circle = circle((all_potential_l[i][1],all_potential_l[i][2]),all_potential_l[i][0],size)
over = img & img_circle
cross = np.count_nonzero(over==True)
ratio_i = cross/img_true
if ratio_i > 0.9915:
s = True
#print(ratio_i)
y_fit = all_potential_l[i][1]
x_fit = all_potential_l[i][2]
r_fit = all_potential_l[i][0]
cir = circle((y_fit,x_fit),r_fit,img.shape)
cir = np.where(cir==True, -2, cir)
sub = img.astype(int) - cir
break
y_fit = 'wrong'
x_fit = 'wrong'
r_fit = 'wrong'
ratio = 'N/A'
s = False
sub = img
for xy in range(len(xy_p)):
img_circle = circle((xy_p[xy][0],xy_p[xy][1]),r_fit-r_step,size)
over = img & img_circle
cross = np.count_nonzero(over==True)
ratio_p = cross/img_true
#print(ratio_p)
if ratio_p > 0.994:
y_fit = xy_p[xy][0]
x_fit = xy_p[xy][1]
r_fit = r_fit-r_step
ratio = ratio_p
cir = circle((y_fit,x_fit),r_fit,img.shape)
cir = np.where(cir==True, -2, cir)
sub = img.astype(int) - cir
break
return s, y_fit, x_fit, r_fit,ratio,sub
'''
#-------------------------- original succussful method--------------------------------
#for yi in range(len(y)):
for xi in range(len(x)):
for k in range(len(r)):
#if r[k] < mass_cen[0] and r[k] < mass_cen[1] and r[k] < img.shape[0]-mass_cen[0] and r[k] < img.shape[1]-mass_cen[1]:
img_circle = circle((y[yi],x[xi]), r[k], size)
#count_true = 0
#img_circle = np.float32(img_circle)
over = img & img_circle
cross[yi,xi,k] = np.count_nonzero(over == True)
#cross[yi,xi,k] = count_true
ratio = cross[yi,xi,:]/img_true
#print(ratio)
#ratio = cross[yi,xi,:]/img_total_square
for i in range(len(ratio)):
if ratio[i] > 0.988:
r_diff[yi,xi] = i
break
else:
r_diff[yi,xi] = 999
r_diff_min = np.min(r_diff)
r_diff_ind = np.unravel_index(np.argmin(r_diff, axis=None), r_diff.shape)
if r_diff_min != 999:
status = True
y_fit = y[np.int(r_diff_ind[0])]
x_fit = x[np.int(r_diff_ind[1])]
r_fit = r[np.int(r_diff_min)]
#cir = circle((y_fit#, x_fit), r_fit, img.shape)
#sub_fit = cir-img
#io.imshow(sub_fit)
#print(f'Fitting slice {slice_num} succeeded!')
return status, y_fit, x_fit, r_fit,cross,r_diff
else:
status = False
#message = 'Fitting failed!'
y_fit = 'wrong fit'
x_fit = 'wrong fit'
r_fit = 'wrong fit'
return status, y_fit, x_fit, r_fit, cross, r_diff
#---------------------End-----------------------------------------
'''
file = ['/home/karenchen-wiegart/ChenWiegartgroup/Xiaoyang/201909_FXI_segimg/crop_norm_img/rotate_img/crop_norm_30598_x14_crop137_306_seg1p47.tif']
savefolder = '/home/karenchen-wiegart/ChenWiegartgroup/Xiaoyang/201909_FXI_segimg/crop_norm_img/rotate_img/'
#file = '/home/karenchen-wiegart/ChenWiegartgroup/Xiaoyang/201909_FXI_segimg/'
#file = 'C:\\ChenWiegartGroup\\Xiaoyang\\2019FXI_corrosiondistance_cal_xiaoyang_20200416\\test\\'
#files = glob.glob(file +'*.tif')
file_name_list = []
r_mean_list = []
r_std_list = []
scan_slice_csv = pd.DataFrame()
for f in file:
start_1 = time.time()
img_slice = io.imread(f)
img_slice.astype(bool)
z = img_slice.shape[0]
ran_sli = np.arange(0,z,1)
b = np.zeros([ran_sli.shape[0],img_slice.shape[1],img_slice.shape[2]])
file_name = os.path.splitext(os.path.basename(f))[0][0:5]
print(file_name)
file_name_list.append(file_name)
dat = Parallel(n_jobs=-2,verbose=10)(delayed(fitcir)(img_slice[ran_sli[zi],:,:].astype(bool),30,2,270.0,4,40,2) for zi in range(ran_sli.shape[0]))
s,y,x,r,rat,sub = zip(*dat)
if all(s) == True:
b = np.asarray(sub)
else:
#print(r)
print('Fitting failed!')
#cen_list.append('wrong fit')
#r_list[z] = 0
b = np.int16(b)
io.imsave(savefolder+str(file_name)+'_0p9915_fitoutcir_all.tif',b)
r_array = np.asarray(r)
r_mean = np.mean(r_array)
r_mean_list.append(r_mean)
r_std = np.std(r_array)
r_std_list.append(r_std)
print(file_name,'r_mean',r_mean,'r_std',r_std)
scan_slice_csv[str(file_name)+'centerY'] = y
scan_slice_csv[str(file_name)+'centerX'] = x
scan_slice_csv[str(file_name)+'radius'] = r
scan_slice_csv[' '] = ''
end_1 = time.time()
print('finishscan,time',file_name,end_1-start_1)
scan_slice_csv.to_csv(savefolder+str(file_name)+'_30598_0p994_all_fitcir.csv')
result = pd.DataFrame(zip(file_name_list,r_mean_list,r_std_list),columns=['scan','average radius','standard deviation'])
result = result.sort_values('scan')
result.to_csv(savefolder+str(file_name)+'_30598_0p994_cir.csv')