ComposePerturber

class nrtk.impls.perturb_image.generic.compose_perturber.ComposePerturber(perturbers: list[PerturbImage], box_alignment_mode: str = 'extent')

A class that composes multiple image perturbations by applying a list of perturbers sequentially to an input image.

Note:

This class has not been tested with perturber factories and is not expected to work with perturber factories.

Methods

from_config	Create a ComposePerturber instance from a configuration dictionary.
get_config	Get the configuration dictionary of the ComposePerturber 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	Check whether this class is available for use.
perturb	Apply the sequence of perturbers to the input image.

__init__(perturbers: list[PerturbImage], box_alignment_mode: str = 'extent') → None

Initializes the ComposePerturber.

This has not been tested with perturber factories and is not expected to work wit perturber factories.

Parameters:

perturbers – list of perturbers to apply

classmethod from_config(config_dict: dict, merge_default: bool = True) → Self

Create a ComposePerturber instance from a configuration dictionary.

Args:

config_dict (dict): Configuration dictionary with perturber details. merge_default (bool): Whether to merge with the default configuration.

Returns:

ComposePerturber: An instance of ComposePerturber.

get_config() → dict[str, Any]

Get the configuration dictionary of the ComposePerturber instance.

Returns:

dict[str, Any]: Configuration dictionary containing perturber configurations.

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

Apply the sequence of perturbers to the input image.

Args:

image (np.ndarray): The input image to perturb. boxes (Optional[Iterable[tuple[AxisAlignedBoundingBox, dict[Hashable, float]]]): The bounding boxes for

the input image. This is the single image output from DetectImageObjects.detect_objects

additional_params (Optional[dict[str, Any]]): Additional parameters for perturbation.

Returns:

np.ndarray: The perturbed image.