ngff_zarr.methods._support

Module Contents

Functions

_spatial_dims_last
_spatial_dims_last_zyx
_channel_dim_last
_dim_scale_factors	Calculate incremental scale factors to apply to previous image.
_update_previous_dim_factors
_align_chunks
_compute_sigma	Compute Gaussian kernel sigma values in pixel units for downsampling. sigma = sqrt((k^2 - 1^2)/(2sqrt(2ln(2)))^2) Ref https://discourse.itk.org/t/resampling-to-isotropic-signal-processing-theory/1403/16 https://doi.org/10.1007/978-3-319-24571-3_81 http://discovery.ucl.ac.uk/1469251/1/scale-factor-point-5.pdf
_get_block	Helper method for accessing an enumerated chunk from input
_next_scale_metadata	Compute the next scale metadata based on the previous image and scale factor.
_next_block_shape	Compute the next block shape based on the previous image and scale factor.

Data

_spatial_dims

API

ngff_zarr.methods._support._spatial_dims

None

ngff_zarr.methods._support._spatial_dims_last(

ngff_image: ngff_zarr.ngff_image.NgffImage,

) → ngff_zarr.ngff_image.NgffImage

ngff_zarr.methods._support._spatial_dims_last_zyx(

ngff_image: ngff_zarr.ngff_image.NgffImage,

) → ngff_zarr.ngff_image.NgffImage

ngff_zarr.methods._support._channel_dim_last(

ngff_image: ngff_zarr.ngff_image.NgffImage,

) → ngff_zarr.ngff_image.NgffImage

ngff_zarr.methods._support._dim_scale_factors(

dims,

scale_factor,

previous_dim_factors,

original_image=None,

previous_image=None,

)

Calculate incremental scale factors to apply to previous image.

When original_image and previous_image are provided, calculates the exact incremental factor needed to reach the target size from the previous size. This ensures we get exact 1x, 2x, 3x, 4x sizes even with incremental downsampling.

ngff_zarr.methods._support._update_previous_dim_factors(scale_factor, spatial_dims, previous_dim_factors)

ngff_zarr.methods._support._align_chunks(previous_image, default_chunks, dim_factors)

ngff_zarr.methods._support._compute_sigma(shrink_factors: List[int]) → List[float]

Compute Gaussian kernel sigma values in pixel units for downsampling. sigma = sqrt((k^2 - 1^2)/(2sqrt(2ln(2)))^2) Ref https://discourse.itk.org/t/resampling-to-isotropic-signal-processing-theory/1403/16 https://doi.org/10.1007/978-3-319-24571-3_81 http://discovery.ucl.ac.uk/1469251/1/scale-factor-point-5.pdf

Note: If input spacing / output sigma in physical units, the function would be sigma = sqrt((input_spacing^2*(k^2 - 1^2))/(2sqrt(2ln(2)))^2)

input spacings: List Input image physical spacings in xyzt order

shrink_factors: List Shrink ratio along each axis in xyzt order

result: List Standard deviation of Gaussian kernel along each axis in xyzt order

ngff_zarr.methods._support._get_block(previous_image: ngff_zarr.ngff_image.NgffImage, block_index: int)

Helper method for accessing an enumerated chunk from input

ngff_zarr.methods._support._next_scale_metadata(

previous_image: ngff_zarr.ngff_image.NgffImage,

dim_factors: dict,

spatial_dims: tuple,

) → tuple

Compute the next scale metadata based on the previous image and scale factor.

ngff_zarr.methods._support._next_block_shape(

previous_image: ngff_zarr.ngff_image.NgffImage,

dim_factors: dict,

spatial_dims: tuple,

block_input: numpy.typing.ArrayLike,

) → tuple

Compute the next block shape based on the previous image and scale factor.