forked from RandomNamer/mic_dachuang-utilities
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcuto1024.py
executable file
·173 lines (151 loc) · 5.91 KB
/
cuto1024.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
####接口
####locimg 是要切的图片路径
####locgt 是gt路径
####aimnimpath 是切好的图片发现是阴性视野图放哪的路径
####loc_IMG是切好的图片发现是阳性视野图放哪的路径
####loc_GT是切好的图片发现是阳性视野图他的json放哪的路径 ##我的的json,你们用xml的话自己转一下
####xmlpath 你们注意下 这个是视野图对应的xml的路径，自己改一下
####156行的比例0.4 你们根据自己的图像调，我用0.4卡的阈值
###注意73行
import numpy as np
import time
import os
import cv2
from glob import glob
import xml.etree.ElementTree as ET
import json
import shutil
def xml2json(xmlpath ,jsonpath):
flag = 0
res = []
pcelllist = ['abnormal','hsil','HSIL','ASC-H','asc-h']
tree = ET.parse(xmlpath)
objs = tree.findall('object')
if objs:
flag = 1
for obj in objs:
bbx = {}
bbox = obj.find('bndbox')
name = obj.find('name').text
if name in pcelllist:
x1 = float(bbox.find('xmin').text) - 1 if float(bbox.find('xmin').text) > 0 else 0
y1 = float(bbox.find('ymin').text) - 1 if float(bbox.find('ymin').text) > 0 else 0
x2 = float(bbox.find('xmax').text) - 1
y2 = float(bbox.find('ymax').text) - 1
w = x2-x1
h = y2-y1
bbx['x'] = x1
bbx['y'] = y1
bbx['w'] = w
bbx['h'] = h
bbx['class'] = name
if x1 >= x2 or y1 >= y2:
continue
else:
res.append(bbx)
# boxes = np.append(boxes, np.expand_dims([x1, y1, x2, y2], axis=0).astype(np.uint16), axis=0)
if flag:
with open(jsonpath,'w')as f:
json.dump(res,f)
return flag
def mkdir(path):
if not os.path.exists(path):
os.makedirs(path)
locimg = '.'+os.sep+'clue_images'+os.sep
locgt = '.'+os.sep+'clue_xmls'+os.sep
loc_IMG = '.'+os.sep+'pi'+os.sep#
loc_GT = '.'+os.sep+'pg'+os.sep
aimnimpath = '.'+os.sep+'ni'+os.sep
mkdir(loc_GT)
mkdir(loc_IMG)
mkdir(aimnimpath)
imgpaths = glob(os.path.join(locimg, '*.jpg'))
for imgpath in imgpaths:
## Windows:
#imgname = imgpath.split('')[-1].split('.')[0]
## Linux or Unix:
imgname = imgpath.split('/')[-1].split('.')[0]
xmlpath = locgt+imgname+'.xml'
aimpimpath = loc_IMG+imgname+'.jpg'
aimgtpath = loc_GT+imgname+'.json'
img= cv2.imread(imgpath)
(h,w,c) = img.shape
hnum = int(h/1024)
wnum = int(w/1024)
if hnum == 1 and wnum == 1:
flag = xml2json(xmlpath,aimgtpath)
print(flag)
if flag == 1:
shutil.copy(imgpath, aimpimpath)
else:
shutil.copy(imgpath, aimnimpath)
else:
print(imgname+'*'*20)
# pcelllist = ['abnormal', 'hsil', 'HSIL', 'ASC-H', 'asc-h']
tree = ET.parse(xmlpath)
objs = tree.findall('object')
res = []
if objs:
for obj in objs:
bbx = {}
bbox = obj.find('bndbox')
name = obj.find('name').text
# if name in pcelllist:
x1 = float(bbox.find('xmin').text) - 1 if float(bbox.find('xmin').text) > 0 else 0
y1 = float(bbox.find('ymin').text) - 1 if float(bbox.find('ymin').text) > 0 else 0
x2 = float(bbox.find('xmax').text) - 1
y2 = float(bbox.find('ymax').text) - 1
w = x2 - x1
h = y2 - y1
bbx['x'] = x1
bbx['y'] = y1
bbx['w'] = w
bbx['h'] = h
bbx['class'] = name
if x1 >= x2 or y1 >= y2:
continue
else:
res.append(bbx)
for i in range(wnum):
for j in range(hnum):
x1 = 1024*i
y1 = 1024*j
x2 = x1+1024
y2 = y1+1024
saveimg = img[y1:y2,x1:x2,:]
# print(saveimg.shape)
savename = imgname+'-'+str(i*hnum+j+1)
if res == []:
cv2.imwrite(aimnimpath+savename+'.jpg',saveimg)
else:
jr = []
for bbx in res:
newbbx = {}
bx1 = bbx['x']
by1 = bbx['y']
bw = bbx['w']
bh = bbx['h']
bx2 = bx1+bw
by2 = by1+bh
lu = bx1>=x1 and bx1<=x2 and by1>=y1 and by1<=y2
ru = bx2>=x1 and bx2<=x2 and by1>=y1 and by1<=y2
lb = bx1>=x1 and bx1<=x2 and by2>=y1 and by2<=y2
rb = bx2>=x1 and bx2<=x2 and by2>=y1 and by2<=y2
if lu or ru or lb or rb:
fx= max(bx1*(bx1>=x1),x1)
fy = max(by1*(by1>=y1),y1)
fw = min(x2-fx,bx2-fx)
fh = min(y2-fy,by2-fy)
newbbx['x'] = fx-x1
newbbx['y'] = fy-y1
newbbx['w'] = fw
newbbx['h'] = fh
newbbx['class'] = bbx['class']
if (newbbx['w']*newbbx['h'])/(bw*bh)>=0.4:
jr.append(newbbx)
if jr!=[]:
cv2.imwrite(loc_IMG+savename+'.jpg',saveimg)
with open(loc_GT+savename+'.json','w')as f:
json.dump(jr,f)
else:
cv2.imwrite(aimnimpath+savename+'.jpg',saveimg)