/
MergerMaker.py
381 lines (324 loc) · 13.2 KB
/
MergerMaker.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
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
"""
拼合多个stl的label制作！多线程版！
"""
import os
import vtk
import time
import argparse
from vtk.util.numpy_support import vtk_to_numpy
from vtk.util.numpy_support import numpy_to_vtk
import numpy as np
import skimage.io as io
from PIL import Image
import shutil
from concurrent.futures import ProcessPoolExecutor
# 融合两张图片
def blend_two_images(ori, label, savepath):
rgb = np.zeros((ori.shape[0], ori.shape[1], 3), dtype=np.uint8)
rgb[:, :, 0] = ori
rgb[:, :, 1] = ori
rgb[:, :, 2] = ori
img1 = Image.fromarray(rgb, "RGB")
img1 = img1.convert('RGBA')
img2 = Image.fromarray(label, "RGB")
img2 = img2.convert('RGBA')
img = Image.blend(img1, img2, 0.3)
img.save(savepath)
# 灰度三通道变rgb三通道
def grey_to_red(train):
rgb = np.zeros((train.shape[0], train.shape[1], 3), dtype=np.uint8)
rgb[:, :, 0] = train*1.5
return rgb
# 变换窗宽窗位
def winchange(pix, wl, ww):
window_center = wl # 窗位
window_width = ww # 窗宽
win_min = (2 * window_center - window_width) / 2.0 + 0.5
win_max = (2 * window_center + window_width) / 2.0 + 0.5
# 计算pix为ct值
ct = pix # * slope + intercept
# 计算比例后乘以255的单位矩阵
ct = ((ct - window_center) / (win_max - win_min) + 0.5) * 255
# 消除大于最大ct值的数值，设为窗口最大值
ct = np.where(ct > 255, 255, ct)
# 消除小于最小ct值的数值，设为窗口最小值
ct = np.where(ct < 0, 0, ct)
ct = ct.astype(np.uint8)
return ct
# 制作Label数据
def MakeLabel(dicomPath, stlfilepath, savepath, orderdir, wl, ww):
MakeLabel(dicomPath, stlfilepath, savepath, orderdir, wl, ww, 'S')
# 制作Label数据
def MakeLabel(dicomPath, stlfilepaths, savepath, orderdir, wl, ww, face):
# 定义保存路径
label_path, ori_path, train_path = getSavepath(savepath)
# 读取dicom文件
dicomImageData, m_Origin = ReadDicomFile(dicomPath)
# 读取stl
stlmapImageDatas = processStlToImageData(dicomImageData, m_Origin, stlfilepaths)
# 处理文件夹
checkAndMakeDir(label_path)
checkAndMakeDir(ori_path)
checkAndMakeDir(train_path)
# 获取stl和dicom的array数据
count, dicomArray, stlmapArray = getResharpedAryFromImageData(dicomImageData, stlmapImageDatas)
# count等于0代表读取image文件存在问题
if count == 0:
print("folder " + orderdir + " image size error!")
return
dicomshape = dicomArray.shape
# 横断面
if face == 'S':
# 操作numpy数据
for i in range(dicomshape[0]):
if i % 2 > 0:
train = stlmapArray[i, :, :]
# 上下镜像翻转
train = train[::-1]
train = np.where(train > 135, 135, 0)
io.imsave(train_path + orderdir + '-' + addPadding(str(i)) + '.png', train)
ori = dicomArray[i, :, :]
ori = ori[::-1]
ori = winchange(ori, wl, ww)
io.imsave(ori_path + orderdir + '-' + addPadding(str(i)) + '.jpg', ori)
blend_two_images(ori, grey_to_red(train), label_path + orderdir + '-' + addPadding(str(i)) + '.png')
print("folder " + orderdir + " done!")
print("train data has save to path: " + train_path)
print("original data has save to path: " + ori_path)
print("label data has save to path: " + label_path)
# 冠状面
if face == 'A':
# 操作numpy数据
for i in range(dicomshape[1]):
if i % 2 > 0:
train = stlmapArray[:, i, :]
train = np.where(train > 135, 135, 0)
io.imsave(train_path + orderdir + '-' + addPadding(str(i)) + '.png', train)
ori = dicomArray[:, i, :]
ori = winchange(ori, wl, ww)
io.imsave(ori_path + orderdir + '-' + addPadding(str(i)) + '.jpg', ori)
blend_two_images(ori, grey_to_red(train), label_path + orderdir + '-' + addPadding(str(i)) + '.png')
print("folder " + orderdir + " done!")
print("train data has save to path: " + train_path)
print("original data has save to path: " + ori_path)
print("label data has save to path: " + label_path)
# 矢状面
if face == 'R':
# 操作numpy数据
for i in range(dicomshape[2]):
if i % 2 > 0:
train = stlmapArray[:, :, i]
train = np.where(train > 135, 135, 0)
io.imsave(train_path + orderdir + '-' + addPadding(str(i)) + '.png', train)
ori = dicomArray[:, :, i]
ori = winchange(ori, wl, ww)
io.imsave(ori_path + orderdir + '-' + addPadding(str(i)) + '.jpg', ori)
blend_two_images(ori, grey_to_red(train), label_path + orderdir + '-' + addPadding(str(i)) + '.png')
print("folder " + orderdir + " done!")
print("train data has save to path: " + train_path)
print("original data has save to path: " + ori_path)
print("label data has save to path: " + label_path)
def getResharpedAryFromImageData(dicomImageData, stlmapImageDatas):
stlmapArray = None
# 保存label
# 转化标量位一维数组
for stlmapImageData in stlmapImageDatas:
if stlmapArray is None:
stlmapArray = vtk_to_numpy(stlmapImageData.GetPointData().GetScalars())
else:
stlmapArray = stlmapArray + vtk_to_numpy(stlmapImageData.GetPointData().GetScalars())
dicomArray = vtk_to_numpy(dicomImageData.GetPointData().GetScalars())
if stlmapArray.size % (512*512) > 0:
print('image size error!')
return 0, None, None
try:
# 转置一维数组为三维数据，首维为图片数量
count = int(stlmapArray.size / (512 * 512))
stlmapArray = stlmapArray.reshape((count, 512, 512))
dicomArray = dicomArray.reshape((count, 512, 512))
return count, dicomArray, stlmapArray
except Exception :
print(Exception)
return 0, None, None
def processStlToImageData(dicomImageData, m_Origin, stlfilepaths):
resultary=None
for stlfilepath in stlfilepaths:
if not os.path.exists(stlfilepath):
continue
polyData = readStlFile(stlfilepath)
orginlData = dicomImageData
spacing = orginlData.GetSpacing()
outval = 0
whiteData = vtk.vtkImageData()
whiteData.DeepCopy(orginlData)
pointdata = whiteData.GetPointData()
pointscalars = pointdata.GetScalars()
# 通过矩阵计算将whiteData中点的颜色全部设置成白色
sc = vtk_to_numpy(pointscalars)
sc = np.where(sc < 255, 255, 255)
newscalars = numpy_to_vtk(sc)
pointdata.SetScalars(newscalars)
whiteData.Modified()
pol2stenc = vtk.vtkPolyDataToImageStencil()
pol2stenc.SetInputData(polyData)
pol2stenc.SetOutputOrigin(m_Origin)
pol2stenc.SetOutputSpacing(spacing)
pol2stenc.SetOutputWholeExtent(orginlData.GetExtent())
pol2stenc.Update()
imgstenc = vtk.vtkImageStencil()
imgstenc.SetInputData(whiteData)
imgstenc.SetStencilConnection(pol2stenc.GetOutputPort())
imgstenc.ReverseStencilOff()
imgstenc.SetBackgroundValue(outval)
imgstenc.Update()
flip = vtk.vtkImageFlip()
flip.SetInputData(imgstenc.GetOutput())
flip.SetFilteredAxes(1)
flip.Update()
flip2 = vtk.vtkImageFlip()
flip2.SetInputData(flip.GetOutput())
flip2.SetFilteredAxes(2)
flip2.Update()
if resultary is None:
resultary = [flip2.GetOutput()]
else:
resultary.append(flip2.GetOutput())
return resultary
def processDICOMToImageX(dicomImageData, origin, extent):
center =[0,0,0]
centers = [0,0,0]
center[0] = origin[0] + spacing[0] * 0.5 * (extent[0] + extent[1])
center[1] = origin[1] + spacing[1] * 0.5 * (extent[2] + extent[3])
center[2] = origin[2] + spacing[2] * 0.5 * (extent[4] + extent[5])
centers[0] = center[0]
centers[1] = center[1]
centers[2] = center[2]
imagecast = vtk.vtkImageCast()
imagecast.SetInputConnection(dicomImageData)
imagecast.SetOutputScalarTypeToChar()
imagecast.ClampOverflowOn()
imagecast.Update()
imagecast.SetUpdateExtentToWholeExtent()
pImageResliceY = vtk.vtkImageReslice()
pImageResliceY.SetInputConnection(imagecast.GetOutputPort())
pImageResliceY.SetOutputDimensionality(2)
pImageResliceY.SetResliceAxesDirectionCosines(coronalX, coronalY, coronalZ)
pImageResliceY.SetResliceAxesOrigin(center)
pImageResliceY.SetInterpolationModeToLinear()
pImageResliceY.Update()
pImageResliceY.GetOutput()
def readStlFile(stlfilepath):
stlreader = vtk.vtkSTLReader()
stlreader.SetFileName(stlfilepath)
stlreader.Update()
# 处理stl
polyData = stlreader.GetOutput()
return polyData
def ReadDicomFile(dicomPath):
# 读取dicom
reader = vtk.vtkDICOMImageReader()
reader.SetDirectoryName(dicomPath)
reader.Update()
# intercept =reader.GetRescaleOffset()
# slope = reader.GetRescaleSlope()
dicomImageData = reader.GetOutput()
m_Origin = [reader.GetImagePositionPatient()[0],
reader.GetImagePositionPatient()[1],
reader.GetImagePositionPatient()[2]]
return dicomImageData, m_Origin
def getSavepath(savepath):
train_path = savepath + '/train/'
ori_path = savepath + '/ori/'
label_path = savepath + '/label/'
return label_path, ori_path, train_path
def checkAndMakeDir(path):
if not os.path.exists(path):
os.mkdir(path)
def addPadding(number):
newnumber = '0000'+number
length = len(newnumber)
return newnumber[length-4:length]
def getCurrentTime():
return str(time.strftime("%a %b %d %Y %H:%M:%S ", time.localtime()))+':'
def _process(path, orderDir, args):
# 消除warning
import warnings
warnings.filterwarnings("ignore")
# 判断文件夹路径是否正确
if os.path.isdir(path):
print(getCurrentTime() + "switch work dir to path:" + path)
dicomdir = path + '/' + args.dicom_path
stlpaths = []
realStlPaths = []
i = 0
for stl in args.stl_path.split(','):
stlpaths.append(path + '/' + stl)
realStlPaths.append(path + '/' + os.path.dirname(stl) + '/a' + str(i) + 'a.stl')
i = i + 1
if not os.path.exists(dicomdir):
print(getCurrentTime() + "DICOM dir can't find!")
return
for j in range(len(stlpaths)):
if os.path.exists(realStlPaths[j]):
os.remove(realStlPaths[j])
print('remove unkown stl at path:' + realStlPaths[j])
if not os.path.exists(stlpaths[j]):
continue
print('copying stl file to order dir path!')
shutil.copy(stlpaths[j], realStlPaths[j])
savedir = args.save_dir + '/'
checkAndMakeDir(savedir)
orderFolder = savedir + orderDir + '/'
checkAndMakeDir(orderFolder)
partFolder = orderFolder + args.part + '/'
checkAndMakeDir(partFolder)
try:
MakeLabel(dicomdir, realStlPaths, partFolder, orderDir, args.wl, args.ww, args.face)
finally:
for j in range(len(stlpaths)):
if os.path.exists(realStlPaths[j]):
os.remove(realStlPaths[j])
print('remove stl file at:' + realStlPaths[j])
if __name__ == "__main__":
# 消除warning
import warnings
warnings.filterwarnings("ignore")
# 读取命令行参数
parser = argparse.ArgumentParser(description='manual to this script')
parser.add_argument('--mode', type=str, default="s")
parser.add_argument('--part', type=str, default=None)
parser.add_argument('--face', type=str, default='S')
parser.add_argument('--root-dir', type=str, default=None)
parser.add_argument('--save-dir', type=str, default=None)
parser.add_argument('--dicom-path', type=str, default=None)
parser.add_argument('--stl-path', type=str, default=None)
parser.add_argument('--wl', type=int, default=0)
parser.add_argument('--ww', type=int, default=200)
# parser.add_argument('--parallel', type=str, default='f')
args = parser.parse_args()
# 运行模式s为单文件夹模式，m为里面文件夹下所有子文件夹的多文件夹模式
print('mode:' + args.mode)
# 多线程模式
# print('parallelmode:' + args.parallel)
# 部位
print('part:' + args.part)
# 部位
print('face:' + args.face)
# 操作根目录
print('rootdir:' + args.root_dir)
# 保存数据的目录，需要已经创建
print('savedir:' + args.save_dir)
# dicom文件的子路径，例如DICOM/A
print('dicom-path:' + args.dicom_path)
# stl文件的子路径，例如 STL/肝脏.stl
print('stl-path:' + args.stl_path)
# 窗位 默认为0
print('wl:' + str(args.wl))
# 窗宽 默认200
print('ww:' + str(args.ww))
# executor = ProcessPoolExecutor(max_workers=3)
dirlist = os.listdir(args.root_dir)
for orderDir in dirlist:
root = args.root_dir
spath = args.root_dir+"/"+orderDir
_process(spath, orderDir, args)