mr2.algorithms.image_registration.spline_registration

mr2.algorithms.image_registration.spline_registration(fixed: Tensor, moving: Tensor, *, downsampling_factor: int, window_size: int, control_point_spacing: SpatialDimension[float], regularization_weight: float, weight: Tensor | None = None, max_iterations: int = 120) → GridSamplingOp[source]

Run one B-spline registration level.

Parameters:

• fixed (Tensor) – Fixed image.

• moving (Tensor) – Moving image.

• downsampling_factor (int) – Downsampling factor.

• window_size (int) – Window size for NCC similarity measure.

• control_point_spacing (SpatialDimension[float]) – Spacing of the control points.

• regularization_weight (float) – Regularization weight for L2 regularization of the control points.

• weight (Tensor | None, default: None) – Weight/mask tensor. If None, no weighting is applied.

• max_iterations (int, default: 120) – Maximum number of iterations of the LBFGS optimizer.