AlbumentationsPerturber

class nrtk.impls.perturb_image.generic.albumentations_perturber.AlbumentationsPerturber(perturber: str, parameters: dict[str, Any] | None = None, box_alignment_mode: str = 'extent', seed: int | None = None)

AlbumentationsPerturber applies a BasicTransform from Albumentations.

Attributes:

perturber (string):

The name of the BasicTransform perturber to apply.

parameters (dict):

Keyword arguments that should be passed to the given perturber.

seed (int):

An optional seed for reproducible results.

Methods:

perturb:

Applies the specified to an input image.

__call__:

Calls the perturb method with the given input image.

get_config:

Returns the current configuration of the AlbumentationsPerturber instance.

Methods

from_config	Instantiate a new instance of this class given the configuration JSON-compliant dictionary encapsulating initialization arguments.
get_config	Returns the current configuration of the AlbumentationsPerturber instance.
get_default_config	Generate and return a default configuration dictionary for this class.
get_impls	Discover and return a set of classes that implement the calling class.
get_type_string	Returns the fully qualified type string of the PerturbImage class or its subclass.
is_usable	Checks if the required albumentations module is available.
perturb	Apply a BasicTransform from Albumentations to an image.

__init__(perturber: str, parameters: dict[str, Any] | None = None, box_alignment_mode: str = 'extent', seed: int | None = None) → None

AlbumentationsPerturber applies a BasicTransform from Albumentations.

Args:

perturber:

The name of the BasicTransform perturber to apply.

parameters:

Keyword arguments that should be passed to the given perturber.

seed (int):

An optional seed for reproducible results.

box_alignment_mode:

Mode for how to handle how bounding boxes change. Should be one of the following options:

extent: a new axis-aligned bounding box that encases the transformed misaligned box extant: a new axis-aligned bounding box that is encased inside the transformed misaligned box median: median between extent and extant

Default value is extent

Raises:

raises ValueError:

Given perturber is not available in Albumentations.

raises ValueError:

Given perturber does not inherit BasicTransform.

get_config() → dict[str, Any]

Returns the current configuration of the AlbumentationsPerturber instance.

Returns:

return dict[str, Any]:

Configuration dictionary with current settings.

classmethod is_usable() → bool

Checks if the required albumentations module is available.

Returns:

return bool:

True if albumentations is installed; False otherwise.

perturb(image: np.ndarray[Any, Any], boxes: Iterable[tuple[AxisAlignedBoundingBox, dict[Hashable, float]]] | None = None, additional_params: dict[str, Any] | None = None) → tuple[np.ndarray[Any, Any], Iterable[tuple[AxisAlignedBoundingBox, dict[Hashable, float]]] | None]

Apply a BasicTransform from Albumentations to an image.

Args:

image:

Input image as a numpy array of shape (H, W, C).

boxes:

List of bounding boxes in AxisAlignedBoundingBox format and their corresponding classes.

additional_params:

Additional parameters for perturbation.

Returns:

return tuple[np.ndarray, Iterable[tuple[AxisAlignedBoundingBox, dict[Hashable, float]]] | None]:

Image with transform applied and bounding boxes with their coordinates adjusted by the transform.