optimised 'C++'-implementation of rectangularRange_HSV()

optimised 'C++'-implementation of rectangularRange_HSV()

Usage

rectangularRange_HSV_iteronce_cpp(
  H,
  S,
  V,
  upper_bound,
  lower_bound,
  image_width,
  check_V
)

Arguments

H

respective component of a pixel.array

S

respective component of a pixel.array

V

respective component of a pixel.array

upper_bound

EITHER:

• matrix of upper HSV-bounds, e.g. do.call(rbind,list(green = c(H_green_lower,S_green_lower,V_green_lower),drought = c(H_drought_lower,S_drought_lower,V_drought_lower)))

• single vector of length 3 declaring a set of HSV-values

lower_bound

see upper_bound

image_width

Width of pixel.array, as returned via dim(pixel.array)[1]

check_V

boolean toggle to also check the VALUE-component of an HSV-pixel

Value

A list-object with the following elements (when supplying one one pair of bounds)

• pixel.idx - pixel-locations of pixels detected between lower and upper bound.

Upon failure to find any matching pixels, an empty matrix of dimensions [0, 1:2] is returned.

Note

The use of rectangularRange_HSV() & rectangularRange_HSV_cpp() is strongly discouraged in favour of this function, due to its drastically slower execution.

Examples

if (FALSE) { # \dontrun{
library(duflor)
## load example data
file_path <- load_extdata("duflor-icon.png")
pixel.array <- load_image(file_path,F,T)
spectrums <- getOption("duflor.default_hsv_spectrums")

## convert spectrums to matrix
nlb <- do.call(rbind,spectrums$lower_bound)
nub <- do.call(rbind,spectrums$upper_bound)

## strip dimnames-attributes
dimnames(nlb) <- c()
dimnames(nub) <- c()
## extract matches
result <- rectangularRange_HSV_iteronce_cpp(H = pixel.array[,,,1],
                                            S = pixel.array[,,,2],
                                            V = pixel.array[,,,3],
                                            upper_bound = nub,
                                            lower_bound = nlb,
                                            image_width = dim(pixel.array)[1],
                                            check_V = T)
## add names to results-matrix.
names(result) <- names(spectrums$lower_bound)
} # }