asgr / propane

Package of stacking related functions, replacing older Rwcs and ProFound related functions.

ProPane

ProPane is a general purpose image warping and stacking package. It uses the wcslib C library for projections (all legal ones are supported) via the Rwcs package, and uses the threaded Cimg C++ library via the imager library to do image warping. Most of the CPU expensive routines are parallelised at the R level, and some aspects are threaded at the C level via OpenMP. imager in particular will need to be built with OpenMP threading support to make use of these speed-ups, e.g. for MacOS you will want to look at this page https://mac.r-project.org/openmp/.

Synopsis

Package of warping and stacking related functions. The initial version of this package involved converting older Rwcs and ProFound related functions. The non-ProPane versions of these functions will eventually be deprecated and removed, and from 06/02/2023 only the newer ProPane versions will be maintained (bug fixed etc) and enhanced.

The old to new function conversions are:

Rwcs_warp → propaneWarp

Rwcs_stack → propaneStackWarpInVar

Rwcs_rebin → propaneRebin

Rwcs_tran → propaneTran

Rwcs_tweak → propaneTweak

Rwcs_warp_promo → propaneWarpProPane

Rwcs_stack_median → propaneStackWarpMed

profoundMakeStack → propaneStackFlatInVar

profoundCombine → propaneStackFlatFunc

As of 06/02/2023 the functions above are identical to Rwcs v1.6.0 and ProFound v1.20.4, but from that point only the ProPane versions of these functions will be maintained. Much longer term (probably the next moderate version increments) these older function will be removed from Rwcs and ProFound respectively. Notably all the Rwcs equivalent functions refer to 'ProMo' class stacked objects, whereas in ProPane the new objects are 'ProPane' class stacked objects.

Installation

You will need a number of other packages for the warping and stacking functions to work, in particular Rwcs and Rfits (on asgr GitHub) and imager (on CRAN but also on asgr GitHub). Once you have these you will be able to install ProPane with the following:

# install.packages("remotes")
remotes::install_github("asgr/ProPane")

Example

Load in the target images:

image_list = Rfits_make_list(dirlist = system.file('extdata/stack/', package="ProPane"),
                             extlist = 2) #extlist=2 because these are compressed images

Define a distinct WCS to stack to (otherwise it uses the first in the image_list):

keyvalues_out = Rwcs_setkeyvalues(
    CRVAL1 = 36.8962,
    CRVAL2 = -5.1906,
    pixscale = 0.3,
    NAXIS1 = 2200,
    NAXIS2 = 2200
)

Run a parallel inverse variance weighted stack (this will take a few seconds):

stack = propaneStackWarpInVar(image_list = image_list,
                   inVar_list = 0.000468, #assume all have the same inVar
                   exp_list = 10, #all have the same exposure time of 10s
                   cores = 8,
                   keyvalues_out = keyvalues_out,
                   magzero_in = 30,
                   magzero_out = 23.9 #micro-jansky output
                   )

Make some useful plots of the output:

plot(stack$image, qdiff=TRUE)
plot(stack$inVar, magmap=FALSE)
plot(stack$weight, magmap=FALSE)
plot(stack$exp, magmap=FALSE)

For more complicated use cases please see the extended vignette at https://rpubs.com/asgr/999429.