Skip to main content

utils

Utility functions for the FPM reconstruction process.

Function: ft​

fpm_py.utils.ft(x: torch.Tensor):
    ...

Simple macro for 2D Fourier transform.

Args:​

  • x (torch.Tensor): The input image.

Returns:​

  • torch.Tensor: The Fourier transform of the input image.

Function: ift​

fpm_py.utils.ift(x: torch.Tensor):
    ...

Simple macro for 2D Inverse Fourier transform.

Args:​

  • x (torch.Tensor): The input Fourier domain image.

Returns:​

  • torch.Tensor: The inverse Fourier transform of the input image.

Function: kvector_to_x_y​

fpm_py.utils.kvector_to_x_y(
    fourier_center: tuple[int, int],
    image_size: tuple[int, int],
    du: float,
    k_vector: torch.Tensor
):
    ...

Converts k-vector to x and y coordinates in the spatial domain.

Args:​

fourier_center (tuple[int, int]): The center of the Fourier domain image. image_size (tuple[int, int]): The size of the image. du (float): The pixel size in the Fourier domain. k_vector (torch.Tensor): The k-vector associated with the image.

Returns:​

tuple[int, int]: The x and y coordinates in the spatial domain.

Function: overlap_matrices​

fpm_py.utils.overlap_matrices(
    larger: torch.Tensor,
    smaller: torch.Tensor,
    bottom: int,
    left: int
):
    ...

Adds a smaller matrix to a larger matrix at the specified position.

Args:​

  • larger (torch.Tensor): The larger matrix.
  • smaller (torch.Tensor): The smaller matrix.
  • bottom (int): The bottom row index to place the smaller matrix.
  • left (int): The left column index to place the smaller matrix.

Returns:​

  • torch.Tensor: The updated larger matrix.

Function: circle_like​

fpm_py.utils.circle_like(array: torch.Tensor):
    ...

Creates a complex-valued circular mask with the same shape as the input array.

Args:​

  • array (torch.Tensor): The input array.

Returns:​

  • torch.Tensor: The circular mask.