def unet2D_bn_modified(images, training, nlabels):
images_padded = tf.pad(images, [[0,0], [92, 92], [92, 92], [0,0]], 'CONSTANT')
conv1_1 = layers.conv2D_layer_bn(images_padded, 'conv1_1', num_filters=64, training=training, padding='VALID')
conv1_2 = layers.conv2D_layer_bn(conv1_1, 'conv1_2', num_filters=64, training=training, padding='VALID')
pool1 = layers.max_pool_layer2d(conv1_2)
conv2_1 = layers.conv2D_layer_bn(pool1, 'conv2_1', num_filters=128, training=training, padding='VALID')
conv2_2 = layers.conv2D_layer_bn(conv2_1, 'conv2_2', num_filters=128, training=training, padding='VALID')
pool2 = layers.max_pool_layer2d(conv2_2)
conv3_1 = layers.conv2D_layer_bn(pool2, 'conv3_1', num_filters=256, training=training, padding='VALID')
conv3_2 = layers.conv2D_layer_bn(conv3_1, 'conv3_2', num_filters=256, training=training, padding='VALID')
pool3 = layers.max_pool_layer2d(conv3_2)
conv4_1 = layers.conv2D_layer_bn(pool3, 'conv4_1', num_filters=512, training=training, padding='VALID')
conv4_2 = layers.conv2D_layer_bn(conv4_1, 'conv4_2', num_filters=512, training=training, padding='VALID')
pool4 = layers.max_pool_layer2d(conv4_2)
conv5_1 = layers.conv2D_layer_bn(pool4, 'conv5_1', num_filters=1024, training=training, padding='VALID')
conv5_2 = layers.conv2D_layer_bn(conv5_1, 'conv5_2', num_filters=1024, training=training, padding='VALID')
upconv4 = layers.deconv2D_layer_bn(conv5_2, name='upconv4', kernel_size=(4, 4), strides=(2, 2), num_filters=nlabels, training=training)
concat4 = layers.crop_and_concat_layer([upconv4, conv4_2], axis=3)
conv6_1 = layers.conv2D_layer_bn(concat4, 'conv6_1', num_filters=512, training=training, padding='VALID')
conv6_2 = layers.conv2D_layer_bn(conv6_1, 'conv6_2', num_filters=512, training=training, padding='VALID')
upconv3 = layers.deconv2D_layer_bn(conv6_2, name='upconv3', kernel_size=(4, 4), strides=(2, 2), num_filters=nlabels, training=training)
concat3 = layers.crop_and_concat_layer([upconv3, conv3_2], axis=3)
conv7_1 = layers.conv2D_layer_bn(concat3, 'conv7_1', num_filters=256, training=training, padding='VALID')
conv7_2 = layers.conv2D_layer_bn(conv7_1, 'conv7_2', num_filters=256, training=training, padding='VALID')
upconv2 = layers.deconv2D_layer_bn(conv7_2, name='upconv2', kernel_size=(4, 4), strides=(2, 2), num_filters=nlabels, training=training)
concat2 = layers.crop_and_concat_layer([upconv2, conv2_2], axis=3)
conv8_1 = layers.conv2D_layer_bn(concat2, 'conv8_1', num_filters=128, training=training, padding='VALID')
conv8_2 = layers.conv2D_layer_bn(conv8_1, 'conv8_2', num_filters=128, training=training, padding='VALID')
upconv1 = layers.deconv2D_layer_bn(conv8_2, name='upconv1', kernel_size=(4, 4), strides=(2, 2), num_filters=nlabels, training=training)
concat1 = layers.crop_and_concat_layer([upconv1, conv1_2], axis=3)
conv9_1 = layers.conv2D_layer_bn(concat1, 'conv9_1', num_filters=64, training=training, padding='VALID')
conv9_2 = layers.conv2D_layer_bn(conv9_1, 'conv9_2', num_filters=64, training=training, padding='VALID')
pred = layers.conv2D_layer_bn(conv9_2, 'pred', num_filters=nlabels, kernel_size=(1,1), activation=tf.identity, training=training, padding='VALID')
return pred
def unet3D_bn(images, training, nlabels):
images_padded = tf.pad(images, [[0, 0], [44, 44], [44, 44],[44, 44], [0, 0]], 'CONSTANT')
conv1_1 = layers.conv3D_layer_bn(images_padded, 'conv1_1', num_filters=32, kernel_size=(3,3,3), training=training, padding='VALID')
conv1_2 = layers.conv3D_layer_bn(conv1_1, 'conv1_2', num_filters=64, kernel_size=(3,3,3), training=training, padding='VALID')
pool1 = layers.max_pool_layer3d(conv1_2, kernel_size=(2,2,2), strides=(2,2,2))
conv2_1 = layers.conv3D_layer_bn(pool1, 'conv2_1', num_filters=64, kernel_size=(3,3,3), training=training, padding='VALID')
conv2_2 = layers.conv3D_layer_bn(conv2_1, 'conv2_2', num_filters=128, kernel_size=(3,3,3), training=training, padding='VALID')
pool2 = layers.max_pool_layer3d(conv2_2, kernel_size=(2,2,2), strides=(2,2,2))
conv3_1 = layers.conv3D_layer_bn(pool2, 'conv3_1', num_filters=128, kernel_size=(3,3,3), training=training, padding='VALID')
conv3_2 = layers.conv3D_layer_bn(conv3_1, 'conv3_2', num_filters=256, kernel_size=(3,3,3), training=training, padding='VALID')
pool3 = layers.max_pool_layer3d(conv3_2, kernel_size=(2,2,2), strides=(2,2,2))
conv4_1 = layers.conv3D_layer_bn(pool3, 'conv4_1', num_filters=256, kernel_size=(3,3,3), training=training, padding='VALID')
conv4_2 = layers.conv3D_layer_bn(conv4_1, 'conv4_2', num_filters=512, kernel_size=(3,3,3), training=training, padding='VALID')
upconv3 = layers.deconv3D_layer_bn(conv4_2, name='upconv3', kernel_size=(4, 4, 4), strides=(2, 2, 2), num_filters=512, training=training)
concat3 = layers.crop_and_concat_layer([upconv3, conv3_2], axis=4)
conv5_1 = layers.conv3D_layer_bn(concat3, 'conv5_1', num_filters=256, kernel_size=(3,3,3), training=training, padding='VALID')
conv5_2 = layers.conv3D_layer_bn(conv5_1, 'conv5_2', num_filters=256, kernel_size=(3,3,3), training=training, padding='VALID')
upconv2 = layers.deconv3D_layer_bn(conv5_2, name='upconv2', kernel_size=(4, 4, 4), strides=(2, 2, 2), num_filters=256, training=training)
concat2 = layers.crop_and_concat_layer([upconv2, conv2_2], axis=4)
conv6_1 = layers.conv3D_layer_bn(concat2, 'conv6_1', num_filters=128, kernel_size=(3,3,3), training=training, padding='VALID')
conv6_2 = layers.conv3D_layer_bn(conv6_1, 'conv6_2', num_filters=128, kernel_size=(3,3,3), training=training, padding='VALID')
upconv1 = layers.deconv3D_layer_bn(conv6_2, name='upconv1', kernel_size=(4, 4, 2), strides=(2, 2, 2), num_filters=128, training=training)
concat1 = layers.crop_and_concat_layer([upconv1, conv1_2], axis=4)
conv8_1 = layers.conv3D_layer_bn(concat1, 'conv8_1', num_filters=64, kernel_size=(3,3,3), training=training, padding='VALID')
conv8_2 = layers.conv3D_layer_bn(conv8_1, 'conv8_2', num_filters=64, kernel_size=(3,3,3), training=training, padding='VALID')
pred = layers.conv3D_layer_bn(conv8_2, 'pred', num_filters=nlabels, kernel_size=(1,1,1), activation=tf.identity, training=training, padding='VALID')
return pred
def forward(images, training, nlabels):
conv1_1 = layers.conv2D_layer_bn(images,
'conv1_1',
num_filters=64,
training=training,
padding='SAME')
conv1_2 = layers.conv2D_layer_bn(conv1_1,
'conv1_2',
num_filters=64,
training=training,
padding='SAME')
pool1 = layers.max_pool_layer2d(conv1_2, 'pool_1')
conv2_1 = layers.conv2D_layer_bn(pool1,
'conv2_1',
num_filters=128,
training=training,
padding='SAME')
conv2_2 = layers.conv2D_layer_bn(conv2_1,
'conv2_2',
num_filters=128,
training=training,
padding='SAME')
pool2 = layers.max_pool_layer2d(conv2_2, 'pool_2')
dout2 = layers.dropout_layer(pool2, 'dropout_2', training)
conv3_1 = layers.conv2D_layer_bn(dout2,
'conv3_1',
num_filters=256,
training=training,
padding='SAME')
conv3_2 = layers.conv2D_layer_bn(conv3_1,
'conv3_2',
num_filters=256,
training=training,
padding='SAME')
pool3 = layers.max_pool_layer2d(conv3_2, 'pool_3')
dout3 = layers.dropout_layer(pool3, 'dropout_3', training)
conv4_1 = layers.conv2D_layer_bn(dout3,
'conv4_1',
num_filters=512,
training=training,
padding='SAME')
conv4_2 = layers.conv2D_layer_bn(conv4_1,
'conv4_2',
num_filters=512,
training=training,
padding='SAME')
pool4 = layers.max_pool_layer2d(conv4_2, 'pool_4')
dout4 = layers.dropout_layer(pool4, 'dropout_4', training)
conv5_1 = layers.conv2D_layer_bn(dout4,
'conv5_1',
num_filters=1024,
training=training,
padding='SAME')
conv5_2 = layers.conv2D_layer_bn(conv5_1,
'conv5_2',
num_filters=1024,
training=training,
padding='SAME')
upconv4 = layers.deconv2D_layer_bn(conv5_2,
name='upconv4',
kernel_size=(4, 4),
strides=(2, 2),
num_filters=nlabels,
training=training)
concat4 = layers.crop_and_concat_layer([upconv4, conv4_2],
'crop_concat_4',
axis=3)
dout5 = layers.dropout_layer(concat4, 'dropout_5', training)
conv6_1 = layers.conv2D_layer_bn(dout5,
'conv6_1',
num_filters=512,
training=training,
padding='SAME')
conv6_2 = layers.conv2D_layer_bn(conv6_1,
'conv6_2',
num_filters=512,
training=training,
padding='SAME')
upconv3 = layers.deconv2D_layer_bn(conv6_2,
name='upconv3',
kernel_size=(4, 4),
strides=(2, 2),
num_filters=nlabels,
training=training)
concat3 = layers.crop_and_concat_layer([upconv3, conv3_2],
'crop_concat_3',
axis=3)
dout6 = layers.dropout_layer(concat3, 'dropout_6', training)
conv7_1 = layers.conv2D_layer_bn(dout6,
'conv7_1',
num_filters=256,
training=training,
padding='SAME')
conv7_2 = layers.conv2D_layer_bn(conv7_1,
'conv7_2',
num_filters=256,
training=training,
padding='SAME')
upconv2 = layers.deconv2D_layer_bn(conv7_2,
name='upconv2',
kernel_size=(4, 4),
strides=(2, 2),
num_filters=nlabels,
training=training)
concat2 = layers.crop_and_concat_layer([upconv2, conv2_2],
'crop_concat_2',
axis=3)
dout7 = layers.dropout_layer(concat2, 'dropout_7', training)
conv8_1 = layers.conv2D_layer_bn(dout7,
'conv8_1',
num_filters=128,
training=training,
padding='SAME')
conv8_2 = layers.conv2D_layer_bn(conv8_1,
'conv8_2',
num_filters=128,
training=training,
padding='SAME')
upconv1 = layers.deconv2D_layer_bn(conv8_2,
name='upconv1',
kernel_size=(4, 4),
strides=(2, 2),
num_filters=nlabels,
training=training)
concat1 = layers.crop_and_concat_layer([upconv1, conv1_2],
'crop_concat_1',
axis=3)
conv9_1 = layers.conv2D_layer_bn(concat1,
'conv9_1',
num_filters=64,
training=training,
padding='SAME')
conv9_2 = layers.conv2D_layer_bn(conv9_1,
'conv9_2',
num_filters=64,
training=training,
padding='SAME')
pred_1 = layers.conv2D_layer_bn(conv9_2,
'pred',
num_filters=nlabels,
kernel_size=(1, 1),
activation=tf.identity,
training=training,
padding='SAME')
# Deep supervision
ds1_1 = layers.conv2D_layer(conv7_2,
'ds_1',
num_filters=nlabels,
kernel_size=(1, 1),
activation=tf.identity,
padding='SAME')
ds1_2 = layers.deconv2D_layer(ds1_1,
'ds_2',
kernel_size=(4, 4),
strides=(2, 2),
num_filters=nlabels,
padding='SAME')
ds2_1 = layers.conv2D_layer(conv8_2,
'ds_3',
num_filters=nlabels,
kernel_size=(1, 1),
activation=tf.identity,
padding='SAME')
ds1_ds2 = tf.add(ds1_2, ds2_1)
ds = layers.deconv2D_layer(ds1_ds2,
'ds_4',
kernel_size=(4, 4),
strides=(2, 2),
num_filters=nlabels,
padding='SAME')
pred_2 = tf.add(pred_1, ds)
return pred_2
def unet_16_2D_bn(x, training, scope_name='generator'):
n_ch_0 = 16
with tf.variable_scope(scope_name):
conv1_1 = layers.conv2D_layer_bn(x,
'conv1_1',
num_filters=n_ch_0,
training=training)
conv1_2 = layers.conv2D_layer_bn(conv1_1,
'conv1_2',
num_filters=n_ch_0,
training=training)
pool1 = layers.maxpool2D_layer(conv1_2)
conv2_1 = layers.conv2D_layer_bn(pool1,
'conv2_1',
num_filters=n_ch_0 * 2,
training=training)
conv2_2 = layers.conv2D_layer_bn(conv2_1,
'conv2_2',
num_filters=n_ch_0 * 2,
training=training)
pool2 = layers.maxpool2D_layer(conv2_2)
conv3_1 = layers.conv2D_layer_bn(pool2,
'conv3_1',
num_filters=n_ch_0 * 4,
training=training)
conv3_2 = layers.conv2D_layer_bn(conv3_1,
'conv3_2',
num_filters=n_ch_0 * 4,
training=training)
pool3 = layers.maxpool2D_layer(conv3_2)
conv4_1 = layers.conv2D_layer_bn(pool3,
'conv4_1',
num_filters=n_ch_0 * 8,
training=training)
conv4_2 = layers.conv2D_layer_bn(conv4_1,
'conv4_2',
num_filters=n_ch_0 * 8,
training=training)
upconv3 = layers.deconv2D_layer_bn(conv4_2,
name='upconv3',
num_filters=n_ch_0,
training=training)
concat3 = layers.crop_and_concat_layer([upconv3, conv3_2], axis=-1)
conv5_1 = layers.conv2D_layer_bn(concat3,
'conv5_1',
num_filters=n_ch_0 * 4,
training=training)
conv5_2 = layers.conv2D_layer_bn(conv5_1,
'conv5_2',
num_filters=n_ch_0 * 4,
training=training)
upconv2 = layers.deconv2D_layer_bn(conv5_2,
name='upconv2',
num_filters=n_ch_0,
training=training)
concat2 = layers.crop_and_concat_layer([upconv2, conv2_2], axis=-1)
conv6_1 = layers.conv2D_layer_bn(concat2,
'conv6_1',
num_filters=n_ch_0 * 2,
training=training)
conv6_2 = layers.conv2D_layer_bn(conv6_1,
'conv6_2',
num_filters=n_ch_0 * 2,
training=training)
upconv1 = layers.deconv2D_layer_bn(conv6_2,
name='upconv1',
num_filters=n_ch_0,
training=training)
concat1 = layers.crop_and_concat_layer([upconv1, conv1_2], axis=-1)
conv8_1 = layers.conv2D_layer_bn(concat1,
'conv8_1',
num_filters=n_ch_0,
training=training)
conv8_2 = layers.conv2D_layer(conv8_1,
'conv8_2',
num_filters=1,
activation=tf.identity)
return conv8_2