This package is under development, but a working use case is available at frycast/kmspd.

Given a raster image or brick (i.e., multiple raster images), in tif format, the load_SAR_matrix function can load the brick into R and convert it into a matrix (called a SAR_matrix), where each image occupies one column of the matrix, and each row of the matrix represents one pixel in the image(s). The original raster attributes are stored in the SAR_matrix, so that conversion back to raster is straight forward (including with spatial geodata extent). Conversion to/from raster brick and SAR_matrix are handled by the functions brick_to_matrix and matrix_to_brick.

The function patchify_SAR_matrix is useful for reducing the resolution of the raster image(s) by aggregating (e.g., via taking the mean) within patches of specified height and width. The output and input format are both SAR_matrix, and the raster attributes are adjusted to suit the new resolution.

Installation with devtools:

devtools::install_github("frycast/rsar")
library(rsar)

The load_SAR_matrix will import a raster brick from a tif file and convert the raster brick to a SAR_matrix. Each column of a SAR_matrix corresponds to a band in the original raster brick, and each row corresponds to a pixel. So if the raster brick has dimensions A x B x C then the SAR_matrix has dimensions A*B x C. The file path specified below is to a sample dataset in the package directory.

filename <- system.file("extdata", "MG_CC_sub_norm_testclip.tif", package="rsar")
sample <- load_SAR_matrix(filename)

The sample raster brick consists of a 43x50 raster image with 29 bands.

dim(sample)

See the help files

?load_SAR_matrix
?matrix_to_brick
?brick_to_matrix
?patchify_SAR_matrix