spherepc (version 0.1.2)

PGA: Principal geodesic analysis

Description

This function performs principal geodesic analysis.

Usage

PGA(data, col1 = "blue", col2 = "red")

Arguments

data

matrix or data frame consisting of spatial locations with two columns. Each row represents longitude and latitude.

col1

color of data.

col2

color of the resulting line.

Value

plot and a list consisting of

line

spatial locations of points in the principal geodesic line.

Details

This function performs principal geodesic analysis.

References

Fletcher, P. T., Lu, C., Pizer, S. M. and Joshi, S. (2004). Principal geodesic analysis for the study of nonlinear statistics of shape. IEEE Transactions on Medical Imaging, 23, 995-1005.

See Also

LPG.

Examples

# NOT RUN {
library(rgl)
library(sphereplot)
library(geosphere)
#### example 1: noisy half-great circle data
circle <- GenerateCircle(c(150, 60), radius = pi / 2)
half.circle <- circle[circle[, 1] < 0, , drop = FALSE]
sigma <- 2
half.circle <- half.circle + sigma * rnorm(nrow(half.circle))
PGA(half.circle)

#### example 2: noisy S-shaped data 
#### The data consists of two parts: x ~ Uniform[0, 20], y = sqrt(20 * x - x^2) + N(0, sigma^2), 
#### x ~ Uniform[-20, 0], y = -sqrt(-20 * x - x^2) + N(0, sigma^2).
n <- 500                                           
x <- 60 * runif(n)
sigma <- 2                                         
y <- (60 * x - x^2)^(1/2) + sigma * rnorm(n)
simul.S1 <- cbind(x, y)
z <- -60 * runif(n)
w <- -(-60 * z - z^2)^(1/2)+ sigma * rnorm(n)
simul.S2 <- cbind(z, w)
simul.S <- rbind(simul.S1, simul.S2)
PGA(simul.S)
# }