8.1.2.9. blueoil.tfds_augmentor

8.1.2.9.1. Module Contents

8.1.2.9.1.1. Classes

TFDSPad

Add padding to images.

TFDSCrop

Randomly crop an image.

TFDSFlipLeftRight

Flip left right with a probability 0.5.

TFDSBrightness

Adjust the brightness of images by a random factor.

TFDSHue

Randomly change image hue.

TFDSSaturation

Randomly adjust the saturation of a RGB image.

TFDSContrast

Randomly adjust the contrast of an image.

8.1.2.9.1.2. Functions

_random_flip_left_right_bounding_box(image, gt_boxes, seed)

Flip left right only bounding box.

class blueoil.tfds_augmentor.TFDSPad(value, fill=0)

Bases: blueoil.data_processor.Processor

Add padding to images.

Parameters
  • value (int or tuple) – Padding on each border. If a single int is provided this is used to pad all borders. If tuple of length 2 is provided this is the padding on left/right and top/bottom respectively. If a tuple of length 4 is provided this is the padding for the left, top, right and bottom borders respectively.

  • fill (int) – Pixel fill value. Default is 0.

__call__(self, image, **kwargs)

Call processor method for each a element of data.

Return image and labels etc.

class blueoil.tfds_augmentor.TFDSCrop(size, seed=0)

Bases: blueoil.data_processor.Processor

Randomly crop an image.

Parameters
  • size (tuple | list) – the size to crop. i.e. [crop_height, crop_width]

  • seed (int) – seed of a random factor

__call__(self, image, **kwargs)

Call processor method for each a element of data.

Return image and labels etc.

blueoil.tfds_augmentor._random_flip_left_right_bounding_box(image, gt_boxes, seed)

Flip left right only bounding box.

Parameters
  • image (tf.Tensor) – image

  • gt_boxes (tf.Tensor) – bounding boxes. shape is [num_boxes, 5(x, y, w, h, class_id)]

  • seed (int) – seed of a random factor

class blueoil.tfds_augmentor.TFDSFlipLeftRight(seed=0)

Bases: blueoil.data_processor.Processor

Flip left right with a probability 0.5.

Parameters

seed (int) – seed of a random factor

__call__(self, image, gt_boxes=None, **kwargs)

Call processor method for each a element of data.

Return image and labels etc.

class blueoil.tfds_augmentor.TFDSBrightness(delta=0.25, seed=0)

Bases: blueoil.data_processor.Processor

Adjust the brightness of images by a random factor.

(picked from uniform distribution [-delta, delta) )

Parameters
  • delta (float) – max delta for distribution. must be positive

  • seed (int) – seed of a random factor

__call__(self, image, **kwargs)

Call processor method for each a element of data.

Return image and labels etc.

class blueoil.tfds_augmentor.TFDSHue(delta=10.0 / 255, seed=0)

Bases: blueoil.data_processor.Processor

Randomly change image hue.

Parameters
  • delta (float) – max delta for distribution. must be in [0, 0.5]

  • seed (int) – seed of a random factor

__call__(self, image, **kwargs)

Call processor method for each a element of data.

Return image and labels etc.

class blueoil.tfds_augmentor.TFDSSaturation(value=(0.75, 1.25), seed=0)

Bases: blueoil.data_processor.Processor

Randomly adjust the saturation of a RGB image.

Random factor is picked from [lower, upper] 0 <= lower < upper must be satisfied.

Parameters
  • value (float|tuple|list) – Range for random factor is taken as [1 - value, 1 + value] if value is float value and [1 - value[0], 1 + value[1]] if tuple or list with 2 elements

  • seed (int) – seed of a random factor

__call__(self, image, **kwargs)

Call processor method for each a element of data.

Return image and labels etc.

class blueoil.tfds_augmentor.TFDSContrast(value=(0.75, 1.25), seed=0)

Bases: blueoil.data_processor.Processor

Randomly adjust the contrast of an image.

Random factor is picked from [lower, upper] 0 <= lower < upper must be satisfied.

Parameters
  • value (float|tuple|list) – Range for random factor is taken as [1 - value, 1 + value] if value is float value and [1 - value[0], 1 + value[1]] if tuple or list with 2 elements

  • seed (int) – seed of a random factor

__call__(self, image, **kwargs)

Call processor method for each a element of data.

Return image and labels etc.