def preprocess_image_and_label(image,
label,
crop_height,
crop_width,
label_norm=True,
ignore_label=0,
is_training=True,
model_variant=None):
"""Preprocesses the image and label.
Args:
image: Input image.
label: Ground truth annotation label.
crop_height: The height value used to crop the image and label.
crop_width: The width value used to crop the image and label.
min_resize_value: Desired size of the smaller image side.
max_resize_value: Maximum allowed size of the larger image side.
resize_factor: Resized dimensions are multiple of factor plus one.
min_scale_factor: Minimum scale factor value.
max_scale_factor: Maximum scale factor value.
scale_factor_step_size: The step size from min scale factor to max scale
factor. The input is randomly scaled based on the value of
(min_scale_factor, max_scale_factor, scale_factor_step_size).
ignore_label: The label value which will be ignored for training and
evaluation.
is_training: If the preprocessing is used for training or not.
model_variant: Model variant (string) for choosing how to mean-subtract the
images. See feature_extractor.network_map for supported model variants.
Returns:
original_image: Original image (could be resized).
processed_image: Preprocessed image.
label: Preprocessed ground truth segmentation label.
Raises:
ValueError: Ground truth label not provided during training.
"""
if is_training and label is None:
raise ValueError('During training, label must be provided.')
if model_variant is None:
tf.logging.warning(
'Default mean-subtraction is performed. Please specify '
'a model_variant. See feature_extractor.network_map for '
'supported model variants.')
# Keep reference to original image.
original_image = image
processed_image = tf.cast(image, tf.float32)
###c插入 数据增强的方法，和之前的传统方式一样
processed_image = tf.image.per_image_standardization(processed_image)
if label is not None:
label = tf.cast(label, tf.int32)
if label_norm:
label = tf.div(label, 255)
original_image = tf.identity(processed_image)
processed_image.set_shape([None, None, 3])
# Pad image and label to have dimensions >= [crop_height, crop_width]
image_shape = tf.shape(processed_image)
image_height = image_shape[0]
image_width = image_shape[1]
target_height = image_height + tf.maximum(crop_height - image_height, 0)
target_width = image_width + tf.maximum(crop_width - image_width, 0)
# Pad image with mean pixel value.
mean_pixel = tf.reshape([127.5, 127.5, 127.5], [1, 1, 3])
processed_image = preprocess_utils.pad_to_bounding_box(
processed_image, 0, 0, target_height, target_width, mean_pixel)
# if label is not None:
label = preprocess_utils.pad_to_bounding_box(label, 0, 0, target_height,
target_width,
0) #ignore_label
# Randomly crop the image and label.
if is_training and label is not None:
processed_image, label = preprocess_utils.random_crop(
[processed_image, label], crop_height, crop_width)
processed_image.set_shape([crop_height, crop_width, 3])
if label is not None:
label.set_shape([crop_height, crop_width, 1])
if is_training:
# # Randomly left-right flip the image and label.
processed_image, label, _ = preprocess_utils.flip_dim(
[processed_image, label], 0.5, dim=1)
# processed_image=mean_image_subtraction(processed_image, means)
return original_image, processed_image, label
def preprocess_eval_image(image,
label,
crop_height,
crop_width,
label_norm=True,
ignore_label=0,
eval_model='center',
model_variant=None):
# if is_training and label is None:
# raise ValueError('During training, label must be provided.')
# if model_variant is None:
# tf.logging.warning('Default mean-subtraction is performed. Please specify '
# 'a model_variant. See feature_extractor.network_map for '
# 'supported model variants.')
# Keep reference to original image.
original_image = image
processed_image = tf.cast(image, tf.float32)
###c插入 数据增强的方法，和之前的传统方式一样
processed_image = tf.image.per_image_standardization(processed_image)
# if label is not None:
label = tf.cast(label, tf.int32)
if label_norm:
label = tf.div(label, 255)
original_image = tf.identity(processed_image)
processed_image.set_shape([None, None, 3])
# Pad image and label to have dimensions >= [crop_height, crop_width]
image_shape = tf.shape(processed_image)
image_height = image_shape[0]
image_width = image_shape[1]
target_height = image_height + tf.maximum(crop_height - image_height, 0)
target_width = image_width + tf.maximum(crop_width - image_width, 0)
# Pad image with mean pixel value.
mean_pixel = tf.reshape([127.5, 127.5, 127.5], [1, 1, 3])
processed_image = preprocess_utils.pad_to_bounding_box(
processed_image, 0, 0, target_height, target_width, mean_pixel)
if label is not None:
label = preprocess_utils.pad_to_bounding_box(label, 0, 0,
target_height,
target_width,
0) #ignore_label
# Randomly crop the image and label.
if eval_model == 'center':
processed_image, label = preprocess_utils.center_crop(
[processed_image, label], crop_height, crop_width)
print('eval_model: center')
elif eval_model == 'resize':
[processed_image,
label] = (preprocess_utils.resize_image(image=processed_image,
label=label,
height_size=crop_height,
width_size=crop_width,
align_corners=True))
print('eval_model: resize')
else:
raise ValueError(' eval_model: wrong !')
processed_image.set_shape([crop_height, crop_width, 3])
if label is not None:
label.set_shape([crop_height, crop_width, 1])
return original_image, processed_image, label
def preprocess_images_and_labels_consistently(images,
labels,
crop_height,
crop_width,
min_resize_value=None,
max_resize_value=None,
resize_factor=None,
min_scale_factor=1.,
max_scale_factor=1.,
scale_factor_step_size=0,
ignore_label=255,
is_training=True,
model_variant=None):
"""Preprocesses images and labels in a consistent way.
Similar to preprocess_image_and_label, but works on a list of images
and a list of labels and uses the same crop coordinates and either flips
all images and labels or none of them.
Args:
images: List of input images.
labels: List of ground truth annotation labels.
crop_height: The height value used to crop the image and label.
crop_width: The width value used to crop the image and label.
min_resize_value: Desired size of the smaller image side.
max_resize_value: Maximum allowed size of the larger image side.
resize_factor: Resized dimensions are multiple of factor plus one.
min_scale_factor: Minimum scale factor value.
max_scale_factor: Maximum scale factor value.
scale_factor_step_size: The step size from min scale factor to max scale
factor. The input is randomly scaled based on the value of
(min_scale_factor, max_scale_factor, scale_factor_step_size).
ignore_label: The label value which will be ignored for training and
evaluation.
is_training: If the preprocessing is used for training or not.
model_variant: Model variant (string) for choosing how to mean-subtract the
images. See feature_extractor.network_map for supported model variants.
Returns:
original_images: Original images (could be resized).
processed_images: Preprocessed images.
labels: Preprocessed ground truth segmentation labels.
Raises:
ValueError: Ground truth label not provided during training.
"""
if is_training and labels is None:
raise ValueError('During training, labels must be provided.')
if model_variant is None:
tf.logging.warning(
'Default mean-subtraction is performed. Please specify '
'a model_variant. See feature_extractor.network_map for '
'supported model variants.')
if labels is not None:
assert len(images) == len(labels)
num_imgs = len(images)
# Keep reference to original images.
original_images = images
processed_images = [tf.cast(image, tf.float32) for image in images]
if labels is not None:
labels = [tf.cast(label, tf.int32) for label in labels]
# Resize images and labels to the desired range.
if min_resize_value is not None or max_resize_value is not None:
processed_images, labels = zip(*[
preprocess_utils.resize_to_range(image=processed_image,
label=label,
min_size=min_resize_value,
max_size=max_resize_value,
factor=resize_factor,
align_corners=True)
for processed_image, label in zip(processed_images, labels)
])
# The `original_images` becomes the resized images.
original_images = [
tf.identity(processed_image)
for processed_image in processed_images
]
# Data augmentation by randomly scaling the inputs.
scale = get_random_scale(min_scale_factor, max_scale_factor,
scale_factor_step_size)
processed_images, labels = zip(*[
randomly_scale_image_and_label(processed_image, label, scale)
for processed_image, label in zip(processed_images, labels)
])
for processed_image in processed_images:
processed_image.set_shape([None, None, 3])
if crop_height is not None and crop_width is not None:
# Pad image and label to have dimensions >= [crop_height, crop_width].
image_shape = tf.shape(processed_images[0])
image_height = image_shape[0]
image_width = image_shape[1]
target_height = image_height + tf.maximum(crop_height - image_height,
0)
target_width = image_width + tf.maximum(crop_width - image_width, 0)
# Pad image with mean pixel value.
mean_pixel = tf.reshape(feature_extractor.mean_pixel(model_variant),
[1, 1, 3])
processed_images = [
preprocess_utils.pad_to_bounding_box(processed_image, 0, 0,
target_height, target_width,
mean_pixel)
for processed_image in processed_images
]
if labels is not None:
labels = [
preprocess_utils.pad_to_bounding_box(label, 0, 0,
target_height,
target_width,
ignore_label)
for label in labels
]
# Randomly crop the images and labels.
if is_training and labels is not None:
cropped = preprocess_utils.random_crop(processed_images + labels,
crop_height, crop_width)
assert len(cropped) == 2 * num_imgs
processed_images = cropped[:num_imgs]
labels = cropped[num_imgs:]
for processed_image in processed_images:
processed_image.set_shape([crop_height, crop_width, 3])
if labels is not None:
for label in labels:
label.set_shape([crop_height, crop_width, 1])
if is_training:
# Randomly left-right flip the image and label.
res = preprocess_utils.flip_dim(list(processed_images + labels),
_PROB_OF_FLIP,
dim=1)
maybe_flipped = res[:-1]
assert len(maybe_flipped) == 2 * num_imgs
processed_images = maybe_flipped[:num_imgs]
labels = maybe_flipped[num_imgs:]
return original_images, processed_images, labels
def preprocess_image_and_label(image,
label,
crop_height,
crop_width,
min_resize_value=None,
max_resize_value=None,
resize_factor=None,
min_scale_factor=1.,
max_scale_factor=1.,
scale_factor_step_size=0,
ignore_label=255,
is_training=True,
model_variant=None):
"""Preprocesses the image and label.
Args:
image: Input image.
label: Ground truth annotation label.
crop_height: The height value used to crop the image and label.
crop_width: The width value used to crop the image and label.
min_resize_value: Desired size of the smaller image side.
max_resize_value: Maximum allowed size of the larger image side.
resize_factor: Resized dimensions are multiple of factor plus one.
min_scale_factor: Minimum scale factor value.
max_scale_factor: Maximum scale factor value.
scale_factor_step_size: The step size from min scale factor to max scale
factor. The input is randomly scaled based on the value of
(min_scale_factor, max_scale_factor, scale_factor_step_size).
ignore_label: The label value which will be ignored for training and
evaluation.
is_training: If the preprocessing is used for training or not.
model_variant: Model variant (string) for choosing how to mean-subtract the
images. See feature_extractor.network_map for supported model variants.
Returns:
original_image: Original image (could be resized).
processed_image: Preprocessed image.
label: Preprocessed ground truth segmentation label.
Raises:
ValueError: Ground truth label not provided during training.
"""
if is_training and label is None:
raise ValueError('During training, label must be provided.')
if model_variant is None:
tf.logging.warning(
'Default mean-subtraction is performed. Please specify '
'a model_variant. See feature_extractor.network_map for '
'supported model variants.')
# Keep reference to original image.
original_image = image
processed_image = tf.cast(image, tf.float32)
if label is not None:
label = tf.cast(label, tf.int32)
# Resize image and label to the desired range.
if min_resize_value is not None or max_resize_value is not None:
[processed_image,
label] = (preprocess_utils.resize_to_range(image=processed_image,
label=label,
min_size=min_resize_value,
max_size=max_resize_value,
factor=resize_factor,
align_corners=True))
# The `original_image` becomes the resized image.
original_image = tf.identity(processed_image)
# Data augmentation by randomly scaling the inputs.
scale = get_random_scale(min_scale_factor, max_scale_factor,
scale_factor_step_size)
processed_image, label = randomly_scale_image_and_label(
processed_image, label, scale)
processed_image.set_shape([None, None, 3])
if crop_height is not None and crop_width is not None:
# Pad image and label to have dimensions >= [crop_height, crop_width].
image_shape = tf.shape(processed_image)
image_height = image_shape[0]
image_width = image_shape[1]
target_height = image_height + tf.maximum(crop_height - image_height,
0)
target_width = image_width + tf.maximum(crop_width - image_width, 0)
# Pad image with mean pixel value.
mean_pixel = tf.reshape(feature_extractor.mean_pixel(model_variant),
[1, 1, 3])
processed_image = preprocess_utils.pad_to_bounding_box(
processed_image, 0, 0, target_height, target_width, mean_pixel)
if label is not None:
label = preprocess_utils.pad_to_bounding_box(
label, 0, 0, target_height, target_width, ignore_label)
# Randomly crop the image and label.
if is_training and label is not None:
processed_image, label = preprocess_utils.random_crop(
[processed_image, label], crop_height, crop_width)
processed_image.set_shape([crop_height, crop_width, 3])
if label is not None:
label.set_shape([crop_height, crop_width, 1])
if is_training:
# Randomly left-right flip the image and label.
processed_image, label, _ = preprocess_utils.flip_dim(
[processed_image, label], _PROB_OF_FLIP, dim=1)
return original_image, processed_image, label
def aug_single_data_tf(self, image, label):
processed_image = tf.cast(image, tf.float32)
label = tf.cast(label, tf.int32)
if self.with_aug_scale_crop:
# Resize image and label to the desired range.
if self.min_resize_value or self.max_resize_value:
[processed_image, label] = (preprocess_utils.resize_to_range(
image=processed_image,
label=label,
min_size=self.min_resize_value,
max_size=self.max_resize_value,
factor=self.resize_factor))
# Data augmentation by randomly scaling the inputs.
scale = preprocess_utils.get_random_scale(
self.min_scale_factor, self.max_scale_factor,
self.scale_factor_step_size)
processed_image, label = preprocess_utils.randomly_scale_image_and_label(
processed_image, label, scale)
processed_image.set_shape([None, None, 10])
# Pad image and label to have dimensions >= [crop_height, crop_width]
image_shape = tf.shape(processed_image)
image_height = image_shape[0]
image_width = image_shape[1]
target_height = image_height + tf.maximum(
self.crop_height - image_height, 0)
target_width = image_width + tf.maximum(
self.crop_width - image_width, 0)
# offset_height = (target_height - image_height) // 2
# offset_width = (target_width - image_width) // 2
offset_height = 0
offset_width = 0
# Pad image with mean pixel value.
mean_pixel = tf.reshape(
self.mean_rgb + self.mean_hha + self.mean_dep + self.mean_xyz,
[1, 1, 10])
processed_image = preprocess_utils.pad_to_bounding_box(
processed_image, offset_height, offset_width, target_height,
target_width, mean_pixel)
label = preprocess_utils.pad_to_bounding_box(
label, offset_height, offset_width, target_height,
target_width, self.ignore_label)
# Randomly crop the image and label.
processed_image, label = preprocess_utils.random_crop(
[processed_image, label], self.crop_height, self.crop_width)
processed_image.set_shape([self.crop_height, self.crop_width, 10])
label.set_shape([self.crop_height, self.crop_width, 1])
# Randomly left-right flip the image and label.
if self.with_aug_flip:
processed_image, label, _ = preprocess_utils.random_flip(
[processed_image, label], prob=0.5)
# Noise
if self.with_aug_gaus_noise:
def tf_rand(minval=0., maxval=1., dtype=tf.float32):
return tf.reduce_sum(
tf.random.uniform(shape=[1],
minval=minval,
maxval=maxval,
dtype=dtype))
rgb, hha, dep, xyz = tf.split(axis=-1,
num_or_size_splits=[3, 3, 1, 3],
value=processed_image)
hsv = tf.image.rgb_to_hsv(rgb)
hue, sat, val = tf.split(axis=-1, num_or_size_splits=3, value=hsv)
hue = tf.keras.backend.clip(hue + tf_rand() * 70 - 35, 0, 360.)
sat = tf.keras.backend.clip(sat + tf_rand() * 0.3 - 0.15, 0, 1.)
val = tf.keras.backend.clip(val + tf_rand() * 50 - 25, 0, 255.)
hsv = tf.concat([hue, sat, val], axis=-1)
rgb = tf.image.hsv_to_rgb(hsv)
processed_image = tf.concat([rgb, hha, dep, xyz], axis=-1)
processed_image = self.normal_data(processed_image)
return processed_image, label