def parse_record(raw_record, is_training=True, dtype=tf.float32):
"""Parses a record containing a training example of an image.
The input record is parsed into a label and image, and the image is passed
through preprocessing steps (cropping, flipping, and so on).
Args:
raw_record: scalar Tensor tf.string containing a serialized
Example protocol buffer.
is_training: A boolean denoting whether the input is for training.
dtype: data type to use for images/features.
Returns:
Tuple with processed image tensor and one-hot-encoded label tensor.
"""
image_buffer, label, bbox = _parse_example_proto(raw_record)
image = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
bbox=bbox,
output_height=DEFAULT_IMAGE_SIZE,
output_width=DEFAULT_IMAGE_SIZE,
num_channels=NUM_CHANNELS,
is_training=is_training
)
image = tf.cast(image, dtype)
return image, label
def parse_record(raw_record, is_training):
"""Parses a record containing a training example of an image.
The input record is parsed into a label and image, and the image is passed
through preprocessing steps (cropping, flipping, and so on).
Args:
raw_record: scalar Tensor tf.string containing a serialized
Example protocol buffer.
is_training: A boolean denoting whether the input is for training.
Returns:
Tuple with processed image tensor and one-hot-encoded label tensor.
"""
image_buffer, label, bbox = _parse_example_proto(raw_record)
image = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
bbox=bbox,
output_height=_DEFAULT_IMAGE_SIZE,
output_width=_DEFAULT_IMAGE_SIZE,
num_channels=_NUM_CHANNELS,
is_training=is_training)
label = tf.one_hot(tf.reshape(label, shape=[]), _NUM_CLASSES)
return image, label
def parse_record(
raw_record,
is_training,
dtype,
data_format="channels_last",
image_size=defaults.DEFAULT_IMAGE_SIZE,
num_channels=defaults.NUM_CHANNELS,
):
"""Parses a record containing a training example of an image.
The input record is parsed into a label and image, and the image is passed
through preprocessing steps (cropping, flipping, and so on).
Args:
raw_record: scalar Tensor tf.string containing a serialized
Example protocol buffer.
is_training: A boolean denoting whether the input is for training.
dtype: data type to use for images/features.
data_format: the axis order of the matrix, channels_last NHWC or channels_first NCHW
Returns:
Tuple with processed image tensor and one-hot-encoded label tensor.
"""
image_buffer, label = _parse_example_proto(raw_record)
image = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
output_height=image_size,
output_width=image_size,
num_channels=num_channels,
is_training=is_training,
data_format=data_format,
)
image = tf.cast(image, dtype)
return image, label
def get_sample(input_img_filename, input_xml_filename, coder=None):
if coder is None:
coder = ImageCoder()
image_buffer, height, width = _process_image(input_img_filename, coder)
boxes = ProcessXMLAnnotation(input_xml_filename)
# print(boxes[0].xmin, boxes[0].ymin, boxes[0].xmax, boxes[0].ymax)
# print(boxes[0].xmin_scaled, boxes[0].ymin_scaled, boxes[0].xmax_scaled, boxes[0].ymax_scaled)
xmin = tf.expand_dims(boxes[0].xmin_scaled, 0)
ymin = tf.expand_dims(boxes[0].ymin_scaled, 0)
xmax = tf.expand_dims(boxes[0].xmax_scaled, 0)
ymax = tf.expand_dims(boxes[0].ymax_scaled, 0)
# Note that we impose an ordering of (y, x) just to make life difficult.
bbox = tf.concat([ymin, xmin, ymax, xmax], 0)
# Force the variable number of bounding boxes into the shape
# [1, num_boxes, coords].
bbox = tf.expand_dims(bbox, 0)
bbox = tf.expand_dims(bbox, 0)
# bbox = tf.transpose(a=bbox, perm=[0, 2, 1])
image = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
bbox=bbox,
output_height=DEFAULT_IMAGE_SIZE,
output_width=DEFAULT_IMAGE_SIZE,
num_channels=NUM_CHANNELS,
is_training=False)
return image
def preprocess_image(file_name,
output_height=224,
output_width=224,
num_channels=3):
"""Run standard ImageNet preprocessing on the passed image file.
Args:
file_name: string, path to file containing a JPEG image
output_height: int, final height of image
output_width: int, final width of image
num_channels: int, depth of input image
Returns:
Float array representing processed image with shape
[output_height, output_width, num_channels]
Raises:
ValueError: if image is not a JPEG.
"""
if imghdr.what(file_name) != "jpeg":
raise ValueError("At this time, only JPEG images are supported. "
"Please try another image.")
image_buffer = tf.read_file(file_name)
normalized = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
bbox=None,
output_height=output_height,
output_width=output_width,
num_channels=num_channels,
is_training=False)
with tf.Session(config=get_gpu_config()) as sess:
result = sess.run([normalized])
#result = [normalized]
return result[0]
def preprocess_image(file_name,
output_height=RESNET_IMAGE_SIZE,
output_width=RESNET_IMAGE_SIZE,
num_channels=3):
"""Run standard ImageNet preprocessing on the passed image file.
Args:
file_name: string, path to file containing a JPEG image
output_height: int, final height of image
output_width: int, final width of image
num_channels: int, depth of input image
Returns:
Float array representing processed image with shape
[output_height, output_width, num_channels]
Raises:
ValueError: if image is not a JPEG.
"""
with tf.device('/cpu:0'):
image_buffer = tf.read_file(file_name)
normalized = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
bbox=None,
output_height=output_height,
output_width=output_width,
num_channels=num_channels,
is_training=False)
with tf.Session() as sess:
result = sess.run([normalized])
return result[0]
def parse_record(raw_record, is_training, return_label_text=True, fixed_crop=False):
"""Parses a record containing a training example of an image.
The input record is parsed into a label and image, and the image is passed
through preprocessing steps (cropping, flipping, and so on).
Args:
raw_record: scalar Tensor tf.string containing a serialized
Example protocol buffer.
is_training: A boolean denoting whether the input is for training.
Returns:
Tuple with processed image tensor and one-hot-encoded label tensor.
"""
image_buffer, label, bbox, label_text, filename = _parse_example_proto(raw_record)
image = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
bbox=bbox,
output_height=_DEFAULT_IMAGE_SIZE,
output_width=_DEFAULT_IMAGE_SIZE,
num_channels=_NUM_CHANNELS,
is_training=is_training,
fixed_crop=fixed_crop)
if return_label_text:
return image, label, label_text, filename
else:
return image, label
def parse_record(raw_record, is_training, dtype=tf.float32):
image_buffer, label = _parse_example_proto(raw_record)
image_buffer = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
output_height=224,
output_width=224,
num_channels=3,
is_training=is_training)
image = tf.cast(image_buffer, dtype)
return image, label
def _preprocess_image(image_bytes):
"""Preprocess a single raw image."""
# Bounding box around the whole image.
bbox = tf.constant([0.0, 0.0, 1.0, 1.0], dtype=dtype, shape=[1, 1, 4])
height, width, num_channels = image_shape
image = imagenet_preprocessing.preprocess_image(image_bytes,
bbox,
height,
width,
num_channels,
is_training=False)
return image
def parse_record(raw_record, is_training):
image_buffer, label, bbox = _parse_example_proto(raw_record)
image = imagenet_preprocessing.preprocess_image(
image_buffer=image_buffer,
bbox=bbox,
output_height=_DEFAULT_IMAGE_SIZE,
output_width=_DEFAULT_IMAGE_SIZE,
num_channels=_NUM_CHANNELS,
is_training=is_training)
image = tf.cast(image, tf.float32)
label = tf.one_hot(tf.reshape(label, []), _NUM_CLASSES)
return image, label
def parse_record(raw_record, is_training, dtype):
image_buffer, label, bbox = _parse_example_proto(raw_record)
image = preprocess_image(
image_buffer=image_buffer,
bbox=bbox,
output_height=DEFAULT_IMAGE_SIZE,
output_width=DEFAULT_IMAGE_SIZE,
num_channels=NUM_CHANNELS,
is_training=is_training)
image = tf.cast(image, dtype)
return image, label
