RandomScalePerturber
- class nrtk.impls.perturb_image.generic.random_scale_perturber.RandomScalePerturber(limit: float | tuple[float, float] = 0.0, interpolation: int = 1, probability: float = 1.0, seed: int | None = None)
RandomScalePerturber applies a Scale transformation from Albumentations.
- Attributes:
- limit (float | tuple[float, float]):
Range of scaling factors. A single float value is equivalent to a tuple of (-value, value). One will then be added to each value. Therefore, a given value of (-0.2, 0.1) corresponds to scaling the image between 0.8x and 1.1x, and a given value of 0.5 corresponds to scaling the image between 0.5x and 1.5x.
- interpolation (int)
OpenCV flag indicating the interpolation algorithm.
- probability (float):
Probability of applying the scale transformation.
- seed (int):
An optional seed for reproducible results.
- Methods:
- perturb:
Applies the specified perturbation to an input image.
- __call__:
Calls the perturb method with the given input image.
- get_config:
Returns the current configuration of the RandomRotationPerturber instance.
Methods
from_configInstantiate a new instance of this class given the configuration JSON-compliant dictionary encapsulating initialization arguments.
Returns the current configuration of the RandomRotationPerturber instance.
get_default_configGenerate and return a default configuration dictionary for this class.
get_implsDiscover and return a set of classes that implement the calling class.
get_type_stringReturns the fully qualified type string of the PerturbImage class or its subclass.
is_usableChecks if the required albumentations module is available.
perturbApply a BasicTransform from Albumentations to an image.
- __init__(limit: float | tuple[float, float] = 0.0, interpolation: int = 1, probability: float = 1.0, seed: int | None = None) None
RandomScalePerturber applies a random scale perturbation to an input image.
It ensures that bounding boxes are adjusted correctly to reflect the translated image coordinates.
- Args:
- limit:
Range of scaling factors. A single float value is equivalent to a tuple of (-value, value). One will then be added to each value. Therefore, a given value of (-0.2, 0.1) corresponds to scaling the image between 0.8x and 1.1x, and a given value of 0.5 corresponds to scaling the image between 0.5x and 1.5x.
- interpolation:
OpenCV flag indicating the interpolation algorithm.
- probability:
Probability of applying the scale transformation.
- seed:
An optional seed for reproducible results.
- get_config() dict[str, Any]
Returns the current configuration of the RandomRotationPerturber instance.
- Returns:
- return dict[str, Any]:
Configuration dictionary with current settings.