-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathmicroscopy_data_classes.py
260 lines (215 loc) · 9.17 KB
/
microscopy_data_classes.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
import numpy as np
import cv2
import imutils
import matplotlib.pyplot as plt
class microscope_info:
"""General class that contains all the information how to treat the microscopy data
for Dataset 1 and Dataset 2, described in original Thesis text."""
extension='.tif'
folder_Z1_original = './data_Z1/tiffs/'
folder_ultra_original = './data_LightSheet/'
dtype = np.uint8
@staticmethod
def _sort_rule_Z1(name):
return int(name.split(' ')[-1].split('.')[0])
@staticmethod
def _sort_rule_light_sheet(name):
return int( name.split('.')[0].split('section')[-1] )
@staticmethod
def preprocess(image):
return image
def __init__(self):
self.name = None
self.folder = None
self.sorting_rule = None
class z1(microscope_info):
def __init__(self):
super().__init__()
self.name = 'z1'
self.folder = self.folder_Z1_original
self.sorting_rule = self._sort_rule_Z1
@staticmethod
def preprocess(image):
image = image[:,:,0]
image = image.T
return image.astype(np.uint8)
class ls(microscope_info):
def __init__(self):
self.name = 'ls'
self.folder = self.folder_ultra_original
self.sorting_rule = self._sort_rule_light_sheet
@staticmethod
def preprocess(image):
image = 255*image.astype(np.float32)/65536
return image.astype(np.uint8)
class stack3D:
"""Class that contains all the necessary functions to load,
visualize and crop the 3D image of microscopy data"""
_preview_height = 500
_preview_cmap = 'gray'
def _load_slice(self, img_name):
img_slice = cv2.imread(os.path.join(self._folder, img_name), -1)
img_slice = self._image_preprocess(img_slice)
return img_slice
def _load_example(self, idx=None):
if idx:
img_name = self.slices[idx]
print('example:', img_name)
else:
img_name = np.random.choice(self.slices)
print('random example:', img_name)
img_slice = self._load_slice(img_name)
ndims, x_y_shape, data_type = img_slice.ndim, img_slice.shape[:2], img_slice.dtype
return img_slice, ndims, x_y_shape, data_type
def preview(self, image, axes, resize=True):
if resize:
img_res = imutils.resize(image, height=self._preview_height)
else:
img_res = image
axes.imshow(img_res, cmap=self._preview_cmap)
axes.set_xlim(0, img_res.shape[1])
axes.set_ylim(img_res.shape[0], 0)
x_ticks = axes.get_xticks()
y_ticks = axes.get_yticks()
x_ticks_labels = np.linspace(0,image.shape[1],len(x_ticks), dtype=int)
y_ticks_labels = np.linspace(0,image.shape[0],len(y_ticks), dtype=int)
axes.set_xticklabels(x_ticks_labels)
axes.set_yticklabels(y_ticks_labels)
axes.set_xlabel('x', fontsize=14)
axes.set_ylabel('y', fontsize=14)
return axes
@staticmethod
def _convert_memory_values(value):
v_type = 'bits'
if value > 8:
value /= 8
v_type = 'bytes'
if value > 1024:
value /= 1024
v_type = 'kilobytes'
if value > 1024:
value /= 1024
v_type = 'megabytes'
if value > 1024:
value /= 1024
v_type = 'gigabytes'
return round(value,2), v_type
def _estimate_memory(self, size=None):
if self._loading_dtype == np.uint8:
bits = 8
elif self._loading_dtype == np.float32:
bits = 32
if size is None:
slice_memory = self._slice_shape[0]*self._slice_shape[1]*bits
full_stack_memory = len(self.slices)*slice_memory
return self._convert_memory_values(slice_memory), self._convert_memory_values(full_stack_memory)
else:
mem_bits = size*bits
return self._convert_memory_values(mem_bits)
def __init__(self, microscope):
self._folder = microscope.folder
self._image_preprocess = microscope.preprocess
self._extension = microscope.extension
self._loading_dtype = microscope.dtype
# Read information about all images available
# Sort images to left only with correct extension
# Sort images in growing order
names_list = os.listdir(self._folder)
names_list = [img for img in names_list if img.endswith(self._extension)]
names_list.sort(key = microscope.sorting_rule)
self.slices = names_list
# Read data about stack
example_slice, self._ndim, self._slice_shape, self._dtype = self._load_example()
memory_results = self._estimate_memory()
# Visualize
print('Num of slices:', len(self.slices), ', slice shape:', self._slice_shape, ', initial dtype:', self._dtype)
print('Slice memory:', memory_results[0][0],memory_results[0][1]+';',
'Brain memory:', memory_results[1][0], memory_results[1][1])
fig, axes = plt.subplots(1,1)
fig.set_figheight(6)
fig.set_figwidth(8)
axes = self.preview(example_slice, axes)
plt.show()
del example_slice
def get_sample(self, coordinates:tuple=None):
# x - along width (slice.shape[1])
# y - along height (slice.shape[0])
# z - along slices (self.slices)
if coordinates is not None:
((x_0,x_1), (y_0,y_1), (z_0,z_1)) = coordinates
else:
size = 250
x_0 = np.random.randint( self._slice_shape[1]*0.1, min(self._slice_shape[1]*0.9, self._slice_shape[1]-size) )
x_1 = x_0 + size
y_0 = np.random.randint( self._slice_shape[0]*0.1, min(self._slice_shape[0]*0.9, self._slice_shape[0]-size) )
y_1 = y_0 + size
total_slices = len(self.slices)
z_size = 5
z_0 = np.random.randint( int(total_slices*0.2), int(total_slices*0.7))
z_1 = z_0 + z_size
coordinates = ((x_0,x_1), (y_0,y_1), (z_0,z_1))
print('x:',coordinates[0], 'y:',coordinates[1], 'z:',coordinates[2])
if z_1 == z_0:
# One 2D slcie
slices_to_load = self.slices[z_0:z_0+1]
else:
# Several slices
slices_to_load = self.slices[z_0:z_1]
memory_results = self._estimate_memory( (y_1-y_0)*(x_1-x_0)*(z_1-z_0) )
if memory_results[1] in ['bits', 'bytes', 'kilobytes', 'megabytes']:
pass
else:
raise Exception('The asked sample is too large to be loaded')
sample = np.zeros((y_1-y_0, x_1-x_0, z_1-z_0), dtype = self._loading_dtype)
for z_idx, img_name in enumerate( slices_to_load ):
tmp_slice = self._load_slice(img_name)
sample[:,:,z_idx] = tmp_slice[y_0:y_1, x_0:x_1]
return sample
def projections(self, image, transpose=True, flip=False, resize=False):
self.max_projections(image, transpose=transpose, flip=flip, resize=resize)
def max_projections(self, image, transpose=True, flip=False, resize=False):
if len(image.shape) < 3:
raise Exception('Can not build projections for non-3D image')
proj_1 = np.max(image, 0)
proj_2 = np.max(image, 1)
proj_3 = np.max(image, 2)
if transpose:
proj_1 = np.transpose(proj_1, (1,0))
proj_2 = np.transpose(proj_2, (1,0))
if flip:
proj_1 = np.flip(proj_1, 0)
proj_2 = np.flip(proj_2, 0)
fig, axes = plt.subplots(2,2)
fig.set_figheight(10)
fig.set_figwidth(14)
axes[0][0] = self.preview(proj_1, axes[0][0], resize)
axes[0][0].set_ylabel('z', fontsize=14)
axes[0][1] = self.preview(proj_2, axes[0][1], resize)
axes[0][1].set_ylabel('z', fontsize=14)
axes[0][1].set_xlabel('y', fontsize=14)
axes[1][0] = self.preview(proj_3, axes[1][0], resize)
axes[1][1].axis('off')
plt.show()
def idx_projections(self, image, slice_x, slice_y, slice_z, transpose=True, flip=False, resize=False):
if len(image.shape) < 3:
raise Exception('Can not build projections for non-3D image')
proj_1 = image[slice_x,:,:]
proj_2 = image[:,slice_y,:]
proj_3 = image[:,:,slice_z]
if transpose:
proj_1 = np.transpose(proj_1, (1,0))
proj_2 = np.transpose(proj_2, (1,0))
if flip:
proj_1 = np.flip(proj_1, 0)
proj_2 = np.flip(proj_2, 0)
fig, axes = plt.subplots(2,2)
fig.set_figheight(10)
fig.set_figwidth(14)
axes[0][0] = self.preview(proj_1, axes[0][0], resize)
axes[0][0].set_ylabel('z', fontsize=14)
axes[0][1] = self.preview(proj_2, axes[0][1], resize)
axes[0][1].set_ylabel('z', fontsize=14)
axes[0][1].set_xlabel('y', fontsize=14)
axes[1][0] = self.preview(proj_3, axes[1][0], resize)
axes[1][1].axis('off')
plt.show()