Skip to contents

Multi-Resolution Pairwise Curve Alignment

Source: R/alignment.R
elastic.align.pair.multires.Rd

Align two curves using a coarse-to-fine multi-resolution strategy. The curves are first coarsened, aligned at low resolution, and then the alignment is progressively refined to the original grid resolution. This can improve robustness and speed for highly misaligned curves.

Usage

elastic.align.pair.multires(
  f1,
  f2,
  argvals,
  coarsen.factor = 4,
  n.refine = 10,
  step.size = 0.01,
  lambda = 0
)

Arguments

f1

Numeric vector of the first curve's values.

f2

Numeric vector of the second curve's values.

argvals

Numeric vector of the evaluation grid (common to both curves).

coarsen.factor

Coarsening factor for the initial resolution (default 4).

n.refine

Number of refinement steps (default 10).

step.size

Step size for gradient-based refinement (default 0.01).

lambda

Regularisation parameter controlling warping smoothness (default 0).

Value

A list with components:

gamma

numeric vector of the estimated warping function

f2.aligned

numeric vector of the aligned second curve

distance

elastic distance after alignment

See also

elastic.align

Examples

# \donttest{
t <- seq(0, 1, length.out = 200)
f1 <- sin(2 * pi * t)
f2 <- sin(2 * pi * (t^1.5))
res <- elastic.align.pair.multires(f1, f2, t)
# }