md.distri {prWarp}

Mardia-Dryden distribution

Description

Create a matrix of 2D shape coordinates drawn from a Mardia-Dryden distribution (3D not implemented)

Usage

md.distri(M_ref, n, sd = 0.05)

Arguments

M_ref

a k x 2 refence matrix (usually the sample mean shape), where k is the number of 2D landmarks

n

the number of observations

sd

the standard deviation of the distribution (default = 0.02)

Value

the n x 2k matrix of shape coordinates drawn from a Mardia-Dryden distribution

Examples

# 2D landmark coordinates
library("geomorph")
data("HomoMidSag")  # dataset
n_spec <- dim(HomoMidSag)[1]  # number of specimens
k <- dim(HomoMidSag)[2] / 2  # number of landmarks
homo_ar <- arrayspecs(HomoMidSag, k, 2)  # create an array

# Procrustes registration
homo_gpa <- Morpho::procSym(homo_ar)
m_mshape <- homo_gpa$mshape  # average shape

# Mardia-Dryden distribution
Xmd <- md.distri(m_mshape, n = n_spec, sd = 0.005)
# Visualization
plot(Xmd[, 1:k], Xmd[, (k+1):(2*k)], asp = 1, las = 1, cex = 0.5,
main = "Mardia-Dryden distribution", xlab = "X", ylab = "Y")