-
Notifications
You must be signed in to change notification settings - Fork 7
/
make_deep_resunet.py
157 lines (132 loc) · 6.77 KB
/
make_deep_resunet.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
import matplotlib
import matplotlib.pyplot as plt
from PIL import Image
import numpy as np
from keras import layers
from keras.layers import Input, Dense, Activation, Cropping2D, ZeroPadding2D, BatchNormalization, Flatten, Conv2D, Concatenate
from keras.layers import AveragePooling2D, MaxPooling2D, Dropout, GlobalMaxPooling2D, GlobalAveragePooling2D, UpSampling2D, Add
from keras.models import Model
from keras.preprocessing import image
from keras.utils import layer_utils
from keras.utils.data_utils import get_file
from keras.applications.imagenet_utils import preprocess_input
from keras.utils.vis_utils import model_to_dot
from keras.utils import plot_model, to_categorical
from keras.optimizers import Adam
#from FCN_utils import *
from keras.metrics import categorical_accuracy
#from sklearn.metrics import confusion_matrix
import keras.backend as K
K.set_image_data_format('channels_last')
import matplotlib.pyplot as plt
import json
N_CLASSES = 2
def RESUNET(input_shape):
inputs = Input(input_shape)
###encoding block 1
iden1 = Conv2D(64, 1, activation = None, padding='same', kernel_initializer = 'he_normal')(inputs)
conv1 = Conv2D(64, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(inputs) #200
conv1 = BatchNormalization()(conv1)
conv1 = Activation('relu')(conv1)
conv1 = Conv2D(64, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv1) #200
add1 = Add()([iden1,conv1])
pool1 = MaxPooling2D()(add1) #200 -> 100
print(add1)
###encoding block2
iden2 = Conv2D(128, 1, activation = None, padding='same', kernel_initializer = 'he_normal')(pool1)
conv2 = BatchNormalization()(pool1)
conv2 = Activation('relu')(conv2)
conv2 = Conv2D(128, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv2) #200
conv2 = BatchNormalization()(conv2)
conv2 = Activation('relu')(conv2)
conv2 = Conv2D(128, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv2) #200
add2 = Add()([iden2,conv2])
pool2 = MaxPooling2D()(add2) #100 ->50
print (add2)
###encoding block3
iden3 = Conv2D(256, 1, activation = None, padding='same', kernel_initializer = 'he_normal')(pool2)
conv3 = BatchNormalization()(pool2)
conv3 = Activation('relu')(conv3)
conv3 = Conv2D(256, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv3) #200
conv3 = BatchNormalization()(conv3)
conv3 = Activation('relu')(conv3)
conv3 = Conv2D(256, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv3) #200
add3 = Add()([iden3,conv3])
pool3= MaxPooling2D()(add3) #50->25
print (add3)
###encoding block4
iden4 = Conv2D(512, 1, activation=None, padding='same', kernel_initializer = 'he_normal')(pool3)
conv4 = BatchNormalization()(pool3)
conv4 = Activation('relu')(conv4)
conv4 = Conv2D(512, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv4) #200
conv4 = BatchNormalization()(conv4)
conv4 = Activation('relu')(conv4)
conv4 = Conv2D(512, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv4) #200
add4 = Add()([iden4,conv4])
drop4 = Dropout(0.5)(add4)
pool4 = MaxPooling2D()(drop4) #25->12
print (pool4)
###bridge
conv5 = BatchNormalization()(pool4)
conv5 = Activation('relu')(conv5)
conv5 = Conv2D(1014, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv5) #200
conv5 = BatchNormalization()(conv5)
conv5 = Activation('relu')(conv5)
conv5 = Conv2D(1024, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv5) #200
drop5 = Dropout(0.5)(conv5)
print (conv5)
###decoding block1
up6 = UpSampling2D()(drop5) #12->24
up6 = ZeroPadding2D(((1,0),(1,0)))(up6) #24->25
concat6 = Concatenate(axis=3)([up6,add4])
iden6 = Conv2D(512, 1, activation=None, padding='same', kernel_initializer = 'he_normal')(concat6)
conv6 = BatchNormalization()(concat6)
conv6 = Activation('relu')(conv6)
conv6 = Conv2D(512, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv6) #200
conv6 = BatchNormalization()(conv6)
conv6 = Activation('relu')(conv6)
conv6 = Conv2D(512, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv6) #200
add6 = Add()([iden6,conv6])
###decoding block2
up7 = UpSampling2D()(add6) #25->50
#up7 = ZeroPadding2D(((1,0),(1,0)))(up7) 24->25
concat7 = Concatenate(axis=3)([up7,add3])
iden7 = Conv2D(256, 1, activation=None, padding='same', kernel_initializer = 'he_normal')(concat7)
conv7 = BatchNormalization()(concat7)
conv7 = Activation('relu')(conv7)
conv7 = Conv2D(256, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv7) #200
conv7 = BatchNormalization()(conv7)
conv7 = Activation('relu')(conv7)
conv7 = Conv2D(256, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv7) #200
add7 = Add()([iden7,conv7])
###decoding block3
up8 = UpSampling2D()(add7) #50->100
concat8 = Concatenate(axis=3)([up8,add2])
iden8 = Conv2D(128, 1, activation=None, padding='same', kernel_initializer = 'he_normal')(concat8)
conv8 = BatchNormalization()(concat8)
conv8 = Activation('relu')(conv8)
conv8 = Conv2D(128, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv8) #200
conv8 = BatchNormalization()(conv8)
conv8 = Activation('relu')(conv8)
conv8 = Conv2D(128, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv8) #200
add8 = Add()([iden8,conv8])
###decoding block4
up9 = UpSampling2D()(add8) #100->200
#up7 = ZeroPadding2D(((1,0),(1,0)))(up7) 24->25
concat9 = Concatenate(axis=3)([up9,add1])
iden9 = Conv2D(64,1,activation=None, padding='same', kernel_initializer = 'he_normal')(concat9)
conv9 = BatchNormalization()(concat9)
conv9 = Activation('relu')(conv9)
conv9 = Conv2D(64, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv9) #200
conv9 = BatchNormalization()(conv9)
conv9 = Activation('relu')(conv9)
conv9 = Conv2D(64, 3, activation = None, padding = 'same', kernel_initializer = 'he_normal')(conv9) #200
add9 = Add()([iden9,conv9])
conv10 = Conv2D(N_CLASSES, 3, activation ='sigmoid', padding = 'same', kernel_initializer = 'he_normal')(add9)
#sigmoid probably too strong an activation
model = Model(input = inputs, output = conv10)
return model
resunet_model = RESUNET((200,200,14))
print (resunet_model.summary())
with open('model_resunet.json', 'w') as outfile:
outfile.write(json.dumps(json.loads(resunet_model.to_json()), indent=2))