RootscanR (version 0.0.1)

getNeighborCoords: Get coordinates of neighbors of specified distance

Description

getNeighborCoords - Returns coordinates of all pixels in a two dimensional raster/image/... with a specified Euclidean distance from the center, i.e., a pixel circle, either a ring or the whole area with a specified radius.

Usage

getNeighborCoords(
  center,
  radius,
  dims_px = c(1000, 1000),
  type = "ring",
  tol = "strict",
  tol_val = 0.5
)

Value

getNeighborCoords Numeric matrix with two columns containing the neighbors' coordinates (in pixels).

Arguments

center

Numeric vector of length two specifying the coordinates (in pixels) of the center point.

radius

Integer value specifying the radius (in pixels). A radius of 0 returns only the center point itself.

dims_px

Numeric vector of length two (default: 1,000 x 1,000) specifying the dimensions (in pixels). Neighbors outside are ignored.

type

Character, "ring" (default) or "area", specifying if the pixel coordinates of only the outer ring or of the whole circle area should be returned.

tol

Character, "strict" (default) or "loose", specifying how thick the ring/outer edge of the area should be, i.e., if pixels close but not exactly on the circle should be included. If set to "loose", all pixels that touch the circle are included. If set to "strict", only additional pixels whose centers have a distance of at most tol_val to the circle are included.

tol_val

Numeric value specifying the tolerance value (>=0,<1, default: 0.5), i.e., pixels are considered neighbors if the difference between their Euclidean distance to the center and the radius is less than or equal to the tolerance value. Only applies if tol is set to "strict".
A radius of 1 with a tolerance of 0.5 returns all 8 surrounding pixels and with a tolerance of 0 only the 4 orthogonally neighboring pixels.

Examples

# The neighbors with radius 1 of point (4,4) in an 8x8 grid.
# With tolerance 0.5:
test <- matrix(0, nrow = 8, ncol = 8)
test[getNeighborCoords(c(4,4), 1, c(8,8))] <- 1
test
# With tolerance 0.4:
test <- matrix(0, nrow = 8, ncol = 8)
test[getNeighborCoords(c(4,4), 1, c(8,8), tol_val = 0.4)] <- 1
test