/
dataEnhancetojson.py
198 lines (181 loc) · 7.63 KB
/
dataEnhancetojson.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
185
186
187
188
189
190
191
192
193
194
195
196
197
198
import os
import random
from random import shuffle
import numpy as np
import cv2
import base64
import json
import os.path as osp
import importlib
from PIL import Image
import nibabel as nib
import skimage.io as io
from nibabel.viewers import OrthoSlicer3D
import matplotlib.pyplot as plt
import imageio
path='D:/zhaojing/labelData/label/BNS/'
imagesize=256
def genJson(path_open,vecResRects,labelName):
MeserList = []
for i in range(len(vecResRects)):
listTemp1 = []
for j in range(len(vecResRects[i])):
listTemp1.append(vecResRects[i][j][0].tolist())
MeserList.append(listTemp1)
file = open(path_open, 'a') # 追加的方式写入
your_dict = {}
your_dict['version'] = "3.15.2"
your_dict['flags'] = {}
shapes_list = []
for i in range(0,len(MeserList)):
your_dict_shape = {}
your_dict_shape['label'] = labelName
your_dict_shape['line_color'] = None
your_dict_shape['fill_color'] = None
your_dict_shape['points'] = MeserList[i]
your_dict_shape['shape_type'] = "polygon"
your_dict_shape['flags'] = {}
shapes_list.append(your_dict_shape)
your_dict['shapes'] = shapes_list
json_str = json.dumps(your_dict,indent=2) # 将字典装化为json串
file.write(json_str + '\n')
def extractContour(proced_pred):
contourS=[]
for ii in range(0,proced_pred.max()):
imagei = np.zeros([imagesize,imagesize,1], dtype=np.uint8)
imagei[proced_pred==ii+1]=255
contours, hierarchy= cv2.findContours ( imagei , cv2.RETR_EXTERNAL , cv2.CHAIN_APPROX_TC89_L1 )
for i in range(0,len(contours)):
area = cv2.contourArea(contours[i])
if area<100.0: #设置目标的最小面积，可以不设置，但是我再转换的过程中发现很多area=0.0 的情况，所以最好还是设置一下
continue
contourS.append(contours[i])
if (len(contourS)<5):
return None
else:
return contourS
def sortBylen(C):
m = []
n = []
temp = []
res = []
for x in C:
pts = np.asarray(x, dtype=np.int32)
area = cv2.contourArea(pts)
m.append(area)
n.append(area)
m.sort(reverse=True)
index = 0
for i in m:
temp.append(C[n.index(i)])
n[n.index(i)] = []
for x in temp:
index += 1
if index <= 150:
res.append(x)
return res
def data_ringuf_seg_loader(mode,allpoints,img,num,name):
root_path=path+mode
img_path=root_path+'_images/'
if os.path.exists(img_path)==False:
os.mkdir(img_path)
cv2.imwrite(img_path+name+'_%d.jpg'%(num),img)
json_path=root_path+'_json/'
if os.path.exists(json_path)==False:
os.mkdir(json_path)
genJson(json_path+name+'_%d.json'%(num),allpoints,'nuclei')
ring_path=root_path+'_ring/'
if os.path.exists(ring_path)==False:
os.mkdir(ring_path)
masks_path=root_path+'_masks/'
if os.path.exists(masks_path)==False:
os.mkdir(masks_path)
imagemask = np.zeros([img.shape[0],img.shape[1]], dtype=img.dtype)
imagering = np.zeros([img.shape[0],img.shape[1]], dtype=img.dtype)
imagei = np.zeros([img.shape[0],img.shape[1]], dtype=img.dtype)
allpoints=sortBylen(allpoints)
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3, 3))#定义结构元素的形状和大小
for k in range(0, len(allpoints)):
imagei[imagei>0]=0
pts = np.asarray([allpoints[k]], dtype=np.int32)
imagei = cv2.drawContours(imagei,pts,-1,255,-1)
dst = cv2.dilate(imagei, kernel,iterations=2)#膨胀操作
contours, hierarchy= cv2.findContours ( dst , cv2.RETR_EXTERNAL , cv2.CHAIN_APPROX_TC89_L1 )
imagemask = cv2.drawContours(imagemask,pts,-1,255,-1)
imagemask = cv2.drawContours(imagemask,contours,-1,0,1)
imagering = cv2.Canny(imagemask, 50, 150)
imagering = cv2.dilate(imagering, kernel,iterations=1)#膨胀操作
cv2.imwrite(ring_path+name+'_%d.png'%(num),imagering)
cv2.imwrite(masks_path+name+'_%d.png'%(num),imagemask)
def enhance(roi,roisrc,num,mode,name):
contourS1=extractContour(roi)
if contourS1!=None:
data_ringuf_seg_loader(mode,contourS1,roisrc,num+1,name)
img_trans=cv2.transpose(roisrc)
img_flip0=cv2.flip(roisrc,0)
img_flip1=cv2.flip(roisrc,1)
img_flip_1=cv2.flip(roisrc,-1)
mimg_trans=cv2.transpose(roi)
mimg_flip0=cv2.flip(roi,0)
mimg_flip1=cv2.flip(roi,1)
mimg_flip_1=cv2.flip(roi,-1)
contourS2=extractContour(mimg_trans)
if contourS2!=None:
data_ringuf_seg_loader(mode,contourS2,img_trans,num+2,name)
contourS3=extractContour(mimg_flip0)
if contourS3!=None:
data_ringuf_seg_loader(mode,contourS3,img_flip0,num+3,name)
contourS4=extractContour(mimg_flip1)
if contourS4!=None:
data_ringuf_seg_loader(mode,contourS4,img_flip1,num+4,name)
contourS5=extractContour(mimg_flip_1)
if contourS5!=None:
data_ringuf_seg_loader(mode,contourS5,img_flip_1,num+5,name)
if __name__ == '__main__':
# print(torch.__version__())
mode='test'
i=0
pathlist='D:/zhaojing/labelData/label/BNS/test-bns/'
pathlabel='D:/zhaojing/labelData/label/BNS/test-bns-gt/'
for filename in os.listdir(pathlist):
i=i+1
grayimage = cv2.imread(pathlist + filename)
filename = filename.split('.')
image_data = nib.load(pathlabel + filename[0]+'.nii.gz')
img_fdata=image_data.dataobj
new_data1 =np.array(img_fdata)
new_data1 = cv2.flip(new_data1, 1)
rows, cols = new_data1.shape
rotate = cv2.getRotationMatrix2D((rows*0.5, cols*0.5), 90, 1)
new_data1 = cv2.warpAffine(new_data1, rotate, (cols, rows))
new_data=new_data1.copy()
for j in range(0,2):
for k in range(0,2):
roi = new_data[j*256:j*256+imagesize, k*256:k*256+imagesize]
roisrc=grayimage[j*256:j*256+imagesize, k*256:k*256+imagesize]
num=i*1000+j*100+k*10
enhance(roi, roisrc, num,mode,filename[0])
for j in range(0,2):
for k in range(0,2):
roi = new_data[j*224:j*224+300, k*224:k*224+300]
roisrc=grayimage[j*224:j*224+300, k*224:k*224+300]
roi = cv2.resize(roi, (imagesize, imagesize))
roisrc = cv2.resize(roisrc, (imagesize, imagesize))
num=(i+10)*1000+j*100+k*10
enhance(roi, roisrc, num,mode,filename[0])
for j in range(0,2):
for k in range(0,2):
roi = new_data[j*100:j*100+356, k*100:k*100+356]
roisrc=grayimage[j*100:j*100+356, k*100:k*100+356]
num=(i+20)*1000+j*100+k*10
roi = cv2.resize(roi, (imagesize, imagesize))
roisrc = cv2.resize(roisrc, (imagesize, imagesize))
enhance(roi, roisrc, num,mode,filename[0])
for j in range(0,2):
for k in range(0,2):
roi = new_data[j*224:j*224+224, k*224:k*224+224]
roisrc=grayimage[j*224:j*224+224, k*224:k*224+224]
num=(i+30)*1000+j*100+k*10
roi = cv2.resize(roi, (imagesize, imagesize))
roisrc = cv2.resize(roisrc, (imagesize, imagesize))
enhance(roi, roisrc, num,mode,filename[0])