Python inception_preprocessing.preprocess_image方法代码示例

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def get_inception_preprocess_patches(is_training, resize_size, num_patches):

  def _inception_preprocess_patches(data):
    patches = []
    for _ in range(num_patches):
      patches.append(
          inception_pp.preprocess_image(
              data["image"],
              resize_size[0],
              resize_size[1],
              is_training,
              add_image_summaries=False))
    patches = tf.stack(patches)
    data["image"] = patches
    return data

  return _inception_preprocess_patches 

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def get_inception_preprocess_patches(is_training, resize_size, num_of_patches):

  def _inception_preprocess_patches(data):
    patches = []
    for _ in range(num_of_patches):
      patches.append(
          inception_preprocessing.preprocess_image(
              data["image"],
              resize_size[0],
              resize_size[1],
              is_training,
              add_image_summaries=False))
    patches = tf.stack(patches)
    data["image"] = patches
    return data

  return _inception_preprocess_patches 

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def preprocess_raw_bytes(image_bytes, is_training=False, bbox=None):
  """Preprocesses a raw JPEG image.

  This implementation is shared in common between train/eval pipelines,
  and when serving the model.

  Args:
    image_bytes: A string Tensor, containing the encoded JPEG.
    is_training: Whether or not to preprocess for training.
    bbox:        In inception preprocessing, this bbox can be used for cropping.

  Returns:
    A 3-Tensor [height, width, RGB channels] of type float32.
  """

  image = tf.image.decode_jpeg(image_bytes, channels=3)
  image = tf.image.convert_image_dtype(image, dtype=tf.float32)

  if FLAGS.preprocessing == 'vgg':
    image = vgg_preprocessing.preprocess_image(
        image=image,
        output_height=FLAGS.height,
        output_width=FLAGS.width,
        is_training=is_training,
        resize_side_min=_RESIZE_SIDE_MIN,
        resize_side_max=_RESIZE_SIDE_MAX)
  elif FLAGS.preprocessing == 'inception':
    image = inception_preprocessing.preprocess_image(
        image=image,
        output_height=FLAGS.height,
        output_width=FLAGS.width,
        is_training=is_training,
        bbox=bbox)
  else:
    assert False, 'Unknown preprocessing type: %s' % FLAGS.preprocessing
  return image 

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def get_inception_preprocess(is_training, im_size):
  def _inception_preprocess(data):
    data["image"] = inception_pp.preprocess_image(
        data["image"], im_size[0], im_size[1], is_training,
        add_image_summaries=False)
    return data
  return _inception_preprocess 

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def get_inception_preprocess(is_training, im_size):
  def _inception_preprocess(data):
    data["image"] = inception_preprocessing.preprocess_image(
        data["image"], im_size[0], im_size[1], is_training,
        add_image_summaries=False)
    return data
  return _inception_preprocess 

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def _dataset_parser(self, serialized_proto):
    """Parse an Imagenet record from value."""
    keys_to_features = {
        'image/encoded':
            tf.FixedLenFeature((), tf.string, default_value=''),
        'image/format':
            tf.FixedLenFeature((), tf.string, default_value='jpeg'),
        'image/class/label':
            tf.FixedLenFeature([], dtype=tf.int64, default_value=-1),
        'image/class/text':
            tf.FixedLenFeature([], dtype=tf.string, default_value=''),
        'image/object/bbox/xmin':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/ymin':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/xmax':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/ymax':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/class/label':
            tf.VarLenFeature(dtype=tf.int64),
    }

    features = tf.parse_single_example(serialized_proto, keys_to_features)

    bbox = None

    image = features['image/encoded']
    image = tf.image.decode_jpeg(image, channels=3)
    image = tf.image.convert_image_dtype(image, dtype=tf.float32)

    image = inception_preprocessing.preprocess_image(
        image=image,
        output_height=self.hparams.image_size,
        output_width=self.hparams.image_size,
        is_training=self.is_training,
        # If eval_from_hub, do not scale the images during preprocessing.
        scaled_images=not self.eval_from_hub,
        bbox=bbox)

    label = tf.cast(
        tf.reshape(features['image/class/label'], shape=[]), dtype=tf.int32)

    return image, label 

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def _rgb_preprocessing(
            self, rgb_image_op,
            image_augmentation=None,
            imagenet_preprocessing=None,
            resize=None,
            resize_height=None,
            resize_width=None,
            mode='tf'):
        """Preprocess an rgb image into a float image, applying image augmentation and imagenet mean subtraction if desired.

        Please note that cropped images are generated in `_image_decode()` and given separate feature names.
        Also please be very careful about resizing the rgb image

        # Arguments

            mode: One of "caffe", "tf" or "torch".
                - caffe: will convert the images from RGB to BGR,
                    then will zero-center each color channel with
                    respect to the ImageNet dataset,
                    without scaling.
                - tf: will scale pixels between -1 and 1,
                    sample-wise.
                - torch: will scale pixels between 0 and 1 and then
                    will normalize each channel with respect to the
                    ImageNet dataset.
        """
        with tf.name_scope('rgb_preprocessing'):
            if image_augmentation is None:
                image_augmentation = FLAGS.image_augmentation
            if imagenet_preprocessing is None:
                imagenet_preprocessing = FLAGS.imagenet_preprocessing
            if resize is None:
                resize = FLAGS.resize
            if resize_height is None:
                resize_height = FLAGS.resize_height
            if resize_width is None:
                resize_width = FLAGS.resize_width
            # make sure the shape is correct
            rgb_image_op = tf.squeeze(rgb_image_op)
            # apply image augmentation and imagenet preprocessing steps adapted from keras
            if resize:
                rgb_image_op = tf.image.resize_images(rgb_image_op,
                                                      tf.constant([resize_height, resize_width],
                                                                  name='resize_height_width'))
            if imagenet_preprocessing:
                data_format = K.image_data_format()
                # TODO(ahundt) add scaling to augmentation and use that to augment delta depth parameters
                # TODO(ahundt) possibly subtract imagenet mean if using pretrained weights, also simply divide channels by 255, and  see https://github.com/tensorflow/tensorflow/issues/15722
                rgb_image_op = inception_preprocessing.preprocess_image(
                    rgb_image_op,
                    is_training=image_augmentation,
                    fast_mode=False,
                    mode=mode, data_format=data_format)
            else:
                rgb_image_op = tf.cast(rgb_image_op, tf.float32)

            return rgb_image_op 

# 需要导入模块: import inception_preprocessing [as 别名]
# 或者: from inception_preprocessing import preprocess_image [as 别名]
def _dataset_parser(self, serialized_proto):
    """Parse an Imagenet record from value."""
    keys_to_features = {
        'image/encoded':
            tf.FixedLenFeature((), tf.string, default_value=''),
        'image/format':
            tf.FixedLenFeature((), tf.string, default_value='jpeg'),
        'image/class/label':
            tf.FixedLenFeature([], dtype=tf.int64, default_value=-1),
        'image/class/text':
            tf.FixedLenFeature([], dtype=tf.string, default_value=''),
        'image/object/bbox/xmin':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/ymin':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/xmax':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/ymax':
            tf.VarLenFeature(dtype=tf.float32),
        'image/object/class/label':
            tf.VarLenFeature(dtype=tf.int64),
    }

    features = tf.parse_single_example(serialized_proto, keys_to_features)

    bbox = None

    image = features['image/encoded']
    image = tf.image.decode_jpeg(image, channels=3)
    image = tf.image.convert_image_dtype(image, dtype=tf.float32)

    image = inception_preprocessing.preprocess_image(
        image=image,
        output_height=self.hparams.image_size,
        output_width=self.hparams.image_size,
        is_training=self.is_training,
        bbox=bbox)

    label = tf.cast(
        tf.reshape(features['image/class/label'], shape=[]), dtype=tf.int32)

    return image, label