surfa.Affine#

class surfa.Affine(matrix, source=None, target=None, space=None)#

N-D transform, represented by an (N+1, N+1) affine matrix, that tracks source and target geometries as well as coordinate space.

Parameters

  • matrix ((N, N) or (N, N+1) float) – 2D or 3D linear transformation matrix.

  • source (castable ImageGeometry) – Transform source (or moving) image geometry.

  • target (castable ImageGeometry) – Transform target (or fixed) image geometry.

  • space (Space) – Coordinate space of the transform.

Attributes

matrix

The (N, N) affine matrix array.

ndim

Dimensionality of the transform.

source

Source (or moving) image geometry.

space

Coordinate space of the transform.

target

Target (or fixed) image geometry.

Methods

convert(source=None, target=None, space=None, copy=True)#

Convert affine for a new coordinate space. Since original geometry and coordinate information is required for conversion, an error will be thrown if the source, target, and space have not been defined for the transform.

Parameters

  • source (ImageGeometry, Image, or Surface) – Transform source and target image geometries.

  • target (ImageGeometry, Image, or Surface) – Transform source and target image geometries.

  • space (Space) – Coordinate space of the transform.

  • copy (bool) – Return copy of object if conditions are already satisfied.

Returns

Converted affine transform.

Return type

float

copy()#

Return a copy of the affine.

decompose(degrees=True)#

Decompose the affine matrix into a set of translation, rotation, scale and shear components.

Parameters

degrees (bool) – Return rotation in degrees instead of radians.

Returns

Tuple of (translation, rotation, scale, shear) transform components.

Return type

tuple

det()#

Compute the affine matrix determinant.

Returns

Determinant.

Return type

float

inv()#

Compute the inverse linear transform, reversing source and target information.

Returns

Inverted affine transform.

Return type

Affine

new(matrix)#

Return a new instance of the affine with an updated matrix and preserved metadata.

save(filename)#

Save affine transform to file.

Parameters

filename (str) – Target filename.

to_warp(format=None)#

Convert affine transform to a dense warp field.

Parameters

format (Warp.Format, optional) – Output warp field format.

Returns

Dense transformation field.

Return type

Warp

transform(data, method='linear', rotation='corner', resample=True, fill=0, points=None)#

Apply the affine transform matrix to the input data.

Parameters

  • data ((..., N) float or Volume) – Input coordinates or image to transform.

  • method ({'linear', 'nearest'}) – Image interpolation method if resample is enabled.

  • rotation ({'corner', 'center'}) – Apply affine with rotation axis at the image corner or center.

  • resample (bool) – If enabled, voxel data will be interpolated and resampled, and geometry will be set the target. If disabled, voxel data will not be modified, and only the geometry will be updated (this is not possible if a displacement field is provided).

  • fill (scalar) – Fill value for out-of-bounds voxels.

  • points (N-D point values) – Deprecated. Use the data argument instead.

Returns

Transformed N-D point array if the input is N-D point data, or transformed image if the input is an image.

Return type

(…, N) float or Volume