+
    åûj  ã                   ó   € R t ^RI5 R# )a€  Dilate, erode, open and close images by a structuring element, and reconstruct them from a seed.

Every function takes floating-point images ``(B, C, H, W)`` and applies one 2D ``kernel`` to every channel. The
non-zero cells of the ``(k_h, k_w)`` kernel form the neighborhood, and a ``structuring_element`` of the same shape
makes the operation non-flat. :func:`dilation` reflects the kernel and :func:`erosion` does not, as in
``scipy.ndimage``; the composite operations inherit both. :func:`reconstruction` takes a seed and a mask instead, an
optional ``kernel`` (``3 x 3`` by default) and no structuring element.

Which function to use:

- Grow bright regions, the maximum over the neighborhood: :func:`dilation`; shrink them, the minimum:
  :func:`erosion`.
- Remove the bright details the kernel does not fit in: :func:`opening`; fill the dark ones: :func:`closing`.
- Those details alone: the bright ones, :func:`top_hat`; the dark ones, :func:`bottom_hat`.
- Edge strength, the dilation minus the erosion: :func:`gradient`.
- Grow a seed image inside a mask until it stops changing, to fill holes or keep whole shapes:
  :func:`reconstruction`.
)Ú*N)Ú__doc__Ú
morphology© ó    ÚVC:\Ai_Coding\apps\studio\ComfyUI\.venv\Lib\site-packages\kornia\morphology\__init__.pyÚ<module>r      s   ðñ$ö& r   