R/patternize.R

#' patternize - An R package for quantifying color pattern variation.
#'
#' Quantifying variation in color patterns to study and compare the consistency of their expression
#' necessitates the homologous alignment and color-based segmentation of images. Patternize is an R
#' package that quantifies variation in color patterns as obtained from image data. Patternize
#' defines homology between pattern positions across specimens either through fixed landmarks or
#' image registration. Pattern identification is performed by categorizing the distribution of
#' colors using either an RGB threshold or an unsupervised image segmentation. The quantification
#' of the color patterns can be visualized as heat maps and compared between sets of samples.
#'
#' @author Steven M. Van Belleghem
#'
#' @section patternize main functions:
#'
#' The package has six main functions depending on how you want the alignment of the iamges and
#' the color extraction to be performed.
#'
#' \code{patLanRGB} \cr
#'    Aligns images by transformations obtained from fixed landmarks and extracts colors using
#'    a predefined RGB values and cutoff value.
#'
#' \code{patLanK} \cr
#'    Aligns images by transformations obtained from fixed landmarks and extracts colors using
#'    k-means clustering.
#'
#' \code{patLanW} \cr
#'    Aligns images by transformations obtained from fixed landmarks and extracts color
#'    patterns by watershed segmentation using \code{\link[imager]{imager}} utilities.
#'
#' \code{patRegRGB} \cr
#'    Aligns images using \code{\link[RNiftyReg]{niftyreg}} utilities for automated image
#'    registration and extracts colors using a predefined RGB values and cutoff value.
#'
#' \code{patRegK} \cr
#'    Aligns images using \code{\link[RNiftyReg]{niftyreg}} utilities for automated image
#'    registration and extracts colors using k-means clustering.
#'
#' \code{patRegW} \cr
#'    Aligns images using \code{\link[RNiftyReg]{niftyreg}} utilities for automated image
#'    registration and extracts color patterns by watershed segmentation using
#'    \code{\link[imager]{imager}} utilities.
#'
#'
#' @section patternize preprocessing functions:
#'
#' The input for the main patternize functions are \code{RasterStack} objects and when landmark
#' transformation is used, landmark arrays.
#'
#' \code{makeList} \cr
#'    This function returns a list of RasterStacks or a list of landmarks depending on the input
#'    provided.
#'
#' \code{sampleLandmarks} \cr
#'    Sample landmarks in an image.
#'
#' \code{lanArray} \cr
#'    This function creates a landmark array as used by \code{\link[Morpho]{procSym}} in the
#'    package \code{Morpho}.
#'
#'
#'
#' @section patternize postprocessing functions:
#'
#' \code{sumRaster} \cr
#'    This function sums the individual color pattern rasters as obtained by the main patternize
#'    functions.
#'
#' \code{plotHeat} \cr
#'    Plots the color pattern heatmaps from \code{sumRaster} output.
#'
#' \code{patPCA} \cr
#'    This function transforms the individual color pattern rasters as obtained by the main
#'    patternize functions to a dataframe of 0 and 1 values that can be used for Principal
#'    Component Analysis (\code{\link[stats]{prcomp}}). This function also allows to plot the
#'    analysis including a visualization of the shape changes along the axis.
#'
#' \code{patRDA} \cr
#'    This function transforms the individual color pattern rasters as obtained by the main
#'    patternize functions to a dataframe of 0 and 1 values that can be used for constrained
#'    Redundancy Analysis (\code{\link[vegan]{rda}}). This function also allows to plot the
#'    analysis including a visualization of the shape changes along the axis.
#'
#' \code{patArea} \cr
#'    This fucntion calculates the area in which the color pattern is expressed in each sample
#'    as the relative proportion using the provided outline of the considered trait or structure.
#'
#'
#'
#' @section patternize miscellaneous functions:
#'
#' \code{redRes} \cr
#'    Reduces the resolution of the \code{RasterStack} objects to speed up analysis.
#'
#' \code{kImage} \cr
#'    Performs k-means clustering of images.
#'
#' \code{sampleRGB} \cr
#'    Interactive function to sample RGB value from pixel or area in an image.
#'
#' \code{createTarget} \cr
#'    Creates an artificial target images using a provided outline that can be used for image
#'    registration (experimantal).
#'
#' \code{maskOutline} \cr
#'    Intersects a RasterStack with an outline. Everything outside of the outline will be removed
#'    from the raster.
#'
#' \code{colorChecker} \cr
#'    Calibrate images using ColorChecker.
#'
#'
#' @seealso
#' \code{\link[raster]{raster}},
#' \code{\link[raster]{stack}},
#' \code{\link[Morpho]{procSym}},
#' \code{\link[Morpho]{computeTransform}},
#' \code{\link[RNiftyReg]{niftyreg}}
#' \code{\link[imager]{imager}}
#'
#' \cite{Jon Clayden, Marc Modat, Benoit Presles, Thanasis Anthopoulos and Pankaj Daga (2017).
#' RNiftyReg: Image Registration Using the 'NiftyReg' Library. R package version 2.5.0.
#' https://CRAN.R-project.org/package=RNiftyReg} \cr
#'
#' \cite{Stefan Schlager (2016). Morpho: Calculations and Visualisations Related to Geometric
#' Morphometrics. R package version 2.4.1.1. https://github.com/zarquon42b/Morpho} \cr
#'
#' \cite{Simon Barthelmé (2017). imager: Image processing library based on ‘CImg’. R package
#' version 0.40.2. https://CRAN.R-project.org/package=imager} \cr
#'
#' @docType package
#' @name patternize
NULL

Try the patternize package in your browser

Any scripts or data that you put into this service are public.

patternize documentation built on May 2, 2019, 5:47 a.m.