nvidia.dali.fn.experimental.erode(__input, /, *, anchor=[-1, -1], border_mode='constant', bytes_per_sample_hint=[0], iterations=1, mask_size=[3, 3], preserve=False, seed=-1, device=None, name=None)

Performs an erosion operation on the input image.

This operator allows sequence inputs.

Supported backends
  • ‘gpu’


__input (TensorList ('HW', 'HWC', 'FHWC', 'CHW', 'FCHW')) – Input data. Must be images in HWC or CHW layout, or a sequence of those.

Keyword Arguments:
  • anchor (int or list of int or TensorList of int, optional, default = [-1, -1]) –

    Sets the anchor point of the structuring element. Default value (-1, -1) uses the element’s center as the anchor point.

    Supports per-frame inputs.

  • border_mode (str, optional, default = ‘constant’) – Border mode to be used when accessing elements outside input image.

  • bytes_per_sample_hint (int or list of int, optional, default = [0]) –

    Output size hint, in bytes per sample.

    If specified, the operator’s outputs residing in GPU or page-locked host memory will be preallocated to accommodate a batch of samples of this size.

  • iterations (int, optional, default = 1) – Number of times to execute the operation, typically set to 1. Setting to a value higher than 1 is equivelent to increasing the mask size by (mask_width - 1, mask_height -1) for every additional iteration.

  • mask_size (int or list of int or TensorList of int, optional, default = [3, 3]) –

    Size of the structuring element.

    Supports per-frame inputs.

  • preserve (bool, optional, default = False) – Prevents the operator from being removed from the graph even if its outputs are not used.

  • seed (int, optional, default = -1) –

    Random seed.

    If not provided, it will be populated based on the global seed of the pipeline.