-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathcells_info.py
169 lines (162 loc) · 8.13 KB
/
cells_info.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
import os,time,shutil
import cv2
import numpy as np
import matplotlib.pyplot as plt
import copy
import basefun as bf
from tqdm import tqdm
def get_cell_nuclei_mask(img, img_org, value):
ret_, thresh = cv2.threshold(img, value, 255, cv2.THRESH_BINARY_INV)
image_, contours, hierarchy_ = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
[_h, _w] = img.shape
xo = _w/2
yo = _h/2
long_ = 200
contours_best = []
area_max = 0
_long = 0
if len(contours) == 0:
return 0, 0, 0, 0, 0, 0, 0, 0
for cnt in range(0, len(contours)):
x,y,w,h=cv2.boundingRect(contours[cnt])
xo2 = x+w/2
yo2 = y+h/2
_long = ((xo-xo2)**2+(yo-yo2)**2)**(1/2)
if _long < long_:
contours_best = contours[cnt]
long_ = _long
[w_img,h_img] = thresh.shape
if (0 in contours_best) or ((w_img-1) in contours_best) or ((h_img-1) in contours_best):
return 0, 0, 0, 0, 0, 0, 0, 0
if not type(contours_best) == np.ndarray:
return 0, 0, 0, 0, 0, 0, 0, 0
area = cv2.contourArea(contours_best)
perimeter = cv2.arcLength(contours_best,True)
hull = cv2.convexHull(contours_best)
area2 = cv2.contourArea(hull)
if area == 0:
return 0, 0, 0, 0, 0, 0, 0, 0
rule = area/area2
thresh_ = thresh*0
thresh__ = cv2.fillPoly(thresh_, [contours_best], 255)
thresh__ = cv2.polylines(thresh__, [hull], True, 255, 1)
img_b = img_org[:,:,2]*(thresh__/255)
value_1 = bf.get_img_mean_value(img_b)
return 1, thresh__, cnt+1, area, area2, perimeter, rule, value_1
def get_cell_cytoplasm_mask(img, nuclei_mask, nuclei_area, img_org, value):
ret_, thresh = cv2.threshold(img, value, 255, cv2.THRESH_BINARY_INV)
image_, contours, hierarchy_ = cv2.findContours(thresh,cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_NONE)
if len(contours) == 0:
return 0,0,0,0,0,0,0,0
area_ = 0
contours_max = []
for cnt in range(0, len(contours)):
area = cv2.contourArea(contours[cnt])
if area > area_:
contours_max = contours[cnt]
area_ = area
if not type(contours_max) == np.ndarray:
return 0,0,0,0,0,0,0,0
hull = cv2.convexHull(contours_max)
area2 = cv2.contourArea(hull)
perimeter = cv2.arcLength(contours_max,True)
if area_ == 0:
return 0,0,0,0,0,0,0,0
rule = area_/area2
thresh_ = thresh*0
thresh__ = cv2.fillPoly(thresh_, [contours_max], 255)
thresh__1 = thresh__ * (nuclei_mask/255)
_area = sum(sum(thresh__1))
if _area == 0:
return 0,0,0,0,0,0,0,0
[w_img,h_img] = thresh__.shape
if thresh__[0,0] == 255 or thresh__[w_img-1,0] == 255 or thresh__[0,h_img-1] == 255 or thresh__[w_img-1,h_img-1] == 255:
return 0,0,0,0,0,0,0,0
thresh__ = cv2.polylines(thresh__, [hull], True, 255, 1)
nuclei_mask_off = np.ones(nuclei_mask.shape)*255 - nuclei_mask
thresh___ = thresh__*(nuclei_mask_off/255)
img_r = img_org[:,:,0]*(thresh___/255)
value_1 = bf.get_img_mean_value(img_r)
img_cytoplasm = img*(thresh___/255)
cytoplasm_var = bf.get_img_var(img_cytoplasm)
return 1, thresh___, area_, area2, perimeter, rule, value_1, cytoplasm_var
def get_cell_save_sign(cellinfo):
nuclei_area = []
nuclei_hull_area = []
cytoplasm_area = []
cytoplasm_hull_area = []
nuclei_area = cellinfo['nuclei_area']
nuclei_hull_area = cellinfo['nuclei_hull_area']
cytoplasm_area = cellinfo['cytoplasm_area']
cytoplasm_hull_area = cellinfo['cytoplasm_hull_area']
nuclei_va = nuclei_area/nuclei_hull_area
cytoplasm_va = cytoplasm_area/cytoplasm_hull_area
area_diff = cytoplasm_area - nuclei_area
if nuclei_va < 0.95 or cytoplasm_va < 0.85 or area_diff <150:
sign = 0
else:
sign = 1
return sign
def main2():
dstroot = 'crop'
listcells = os.listdir(dstroot)
cellsinfo = []
for n,i in zip(listcells,tqdm(range(len(listcells)))):
cellinfo = {}
cellpath = os.path.join(dstroot, n)
img = cv2.imread(cellpath)
img_gray = cv2.imread(cellpath, 0)
img_gray = bf.get_fit_img(img_gray)
# cv2.imwrite(cellpath+'abc0.png',img_gray)
value_1,value_2 = bf.get_2value(img_gray)
sign_nuclei, cell_nuclei_mask, nuclei_cnt, nuclei_area, nuclei_hull_area, nuclei_circ, nuclei_rule, cell_nuclei_value = get_cell_nuclei_mask(img_gray, img, value_1) #获取细胞核掩码、个数、面积、凸面积、周长、核形规则度、获取细胞核深染程度
# cv2.imwrite(cellpath+'abc1.png',cell_nuclei_mask)
if sign_nuclei == 0:
continue
sign_cytoplasm, cell_cytoplasm_mask, cytoplasm_area, cytoplasm_hull_area, cytoplasm_circ, cytoplasm_rule, cell_cytoplasm_value, cytoplasm_var = get_cell_cytoplasm_mask(img_gray, cell_nuclei_mask, nuclei_area,img, value_2) #获取细胞质掩码、面积、凸面积、周长、细胞规则度、获取细胞质情况
# cv2.imwrite(cellpath+'abc2.png',cell_cytoplasm_mask)
if sign_cytoplasm == 0 or (cytoplasm_area-nuclei_area) == 0:
continue
cell_N_C = nuclei_area/(cytoplasm_area-nuclei_area) #计算核质比
cellinfo_keys = ['cellpath','nuclei_cnt','nuclei_area','nuclei_hull_area','nuclei_circ','nuclei_rule','cell_nuclei_value','cytoplasm_area','cytoplasm_hull_area','cytoplasm_circ','cytoplasm_rule','cell_cytoplasm_value','cytoplasm_var','cell_N_C']
cellinfo_values = [cellpath,nuclei_cnt,nuclei_area,nuclei_hull_area,nuclei_circ,nuclei_rule,cell_nuclei_value,cytoplasm_area,cytoplasm_hull_area,cytoplasm_circ,cytoplasm_rule,cell_cytoplasm_value,cytoplasm_var,cell_N_C]
cellinfo = dict(zip(cellinfo_keys, cellinfo_values))
cell_save_sign = get_cell_save_sign(cellinfo)
if cell_save_sign == 0:
continue
cellsinfo.append(cellinfo)
newpath = os.path.join('valid', n)
shutil.copy(cellpath, newpath)
np.save("./cells_info/cells_info.npy", cellsinfo)
if __name__ == "__main__":
dstroot = 'crop'
listcells = os.listdir(dstroot)
cellsinfo = []
for n,i in zip(listcells,tqdm(range(len(listcells)))):
cellinfo = {}
cellpath = os.path.join(dstroot, n)
# print(cellpath)
img = cv2.imread(cellpath)
img_gray = cv2.imread(cellpath, 0)
img_gray = bf.get_fit_img(img_gray)
#cv2.imwrite(cellpath+'abc0.png',img_gray)
value_1,value_2 = bf.get_2value(img_gray)
sign_nuclei, cell_nuclei_mask, nuclei_cnt, nuclei_area, nuclei_hull_area, nuclei_circ, nuclei_rule, cell_nuclei_value = get_cell_nuclei_mask(img_gray, img, value_1) #获取细胞核掩码、个数、面积、凸面积、周长、核形规则度、获取细胞核深染程度
#cv2.imwrite(cellpath+str(cell_nuclei_value)[0:5]+'abc1.png',cell_nuclei_mask)
if sign_nuclei == 0:
continue
sign_cytoplasm, cell_cytoplasm_mask, cytoplasm_area, cytoplasm_hull_area, cytoplasm_circ, cytoplasm_rule, cell_cytoplasm_value,cytoplasm_var = get_cell_cytoplasm_mask(img_gray, cell_nuclei_mask, nuclei_area,img, value_2) #获取细胞质掩码、面积、凸面积、周长、细胞规则度、获取细胞质情况
#cv2.imwrite(cellpath+'abc2.png',cell_cytoplasm_mask)
if sign_cytoplasm == 0 or (cytoplasm_area-nuclei_area) == 0:
continue
cell_N_C = nuclei_area/(cytoplasm_area-nuclei_area) #计算核质比
cellinfo_keys = ['cellpath','nuclei_cnt','nuclei_area','nuclei_hull_area','nuclei_circ','nuclei_rule','cell_nuclei_value','cytoplasm_area','cytoplasm_hull_area','cytoplasm_circ','cytoplasm_rule','cell_cytoplasm_value','cytoplasm_var','cell_N_C']
cellinfo_values = [cellpath,nuclei_cnt,nuclei_area,nuclei_hull_area,nuclei_circ,nuclei_rule,cell_nuclei_value,cytoplasm_area,cytoplasm_hull_area,cytoplasm_circ,cytoplasm_rule,cell_cytoplasm_value,cytoplasm_var,cell_N_C]
cellinfo = dict(zip(cellinfo_keys, cellinfo_values))
cell_save_sign = get_cell_save_sign(cellinfo)
if cell_save_sign == 0:
continue
cellsinfo.append(cellinfo)
newpath = os.path.join('valid', n)
shutil.copy(cellpath, newpath)
np.save("./cells_info/cells_info.npy", cellsinfo)