Skip to contents

Prepare Photometry

Source: vignettes/preparing-psf-matched-photometry.Rmd
preparing-psf-matched-photometry.Rmd

Sagui expects a numeric array with dimensions [x, y, band], or a FITS-like list with that array in imDat. Before segmentation, every band should refer to the same sky footprint and effective angular resolution.

Required checks

  1. Register every image to one pixel grid and WCS.
  2. Match all bands to a declared target PSF.
  3. Convert fluxes and uncertainties to compatible units.
  4. Keep the band order and effective wavelengths together.
  5. Record missing-data and foreground-mask conventions.

Sagui does not perform image registration or PSF matching.

Assemble the cube 

band_files <- c("F090W.fits", "F150W.fits", "F200W.fits", "F356W.fits")
bands <- c(0.90, 1.50, 2.00, 3.56)

planes <- lapply(band_files, function(path) FITSio::readFITS(path)$imDat)
stopifnot(length(unique(lapply(planes, dim))) == 1L)
cube <- simplify2array(planes)

Apply the same geometric operations to the variance planes. Measure fluxes from the calibrated images, not display stretches.

Validate 

stopifnot(length(dim(cube)) == 3L)
stopifnot(dim(cube)[3] == length(bands))
stopifnot(all(dim(var_cube) == dim(cube)))

Next, define the pixels to segment with a support mask.