Skip to content

aguinot/metacoadd

Repository files navigation

Metacoadd

pytest

Create a coadd image at the same time as we create sheared images for the metadetection.

Installation

python -m pip install git+https://github.com/aguinot/metacoadd

Details

Steps:

  • Create sub field for the coadd region (step 1)
  • Interpolate (step 2)
  • Deconvolve the PSF (step 3)
  • Deconvolve from the pixels
  • Apply shear
  • Get larger reconv PSF
  • Reconvolve by PSF
  • Apply coadd WCS
  • Run Detection
  • Run shape measurement

I - Create subfield

The coadd size is not yet defined use 1 arcmin for testing.

Idea is to make use of the Galsim bound method to create cutout in the single exposures for the coadd area.

Question:

  1. How to handle borders?
  2. What if we want to mask?
  3. How to hadle grid difference between exposure and coadd?
  4. Does it need to be an interpolated image at this stage?

II - Interpolation

We would like to use lanczosX interpolation to get as close as possible to standar techniques used in UNIONS (through SWARP) (found a good value for X).

III - Deconvolve from PSF

License

This project is licensed under the MIT License - see the LICENSE file for details.

About

Creates coadd for metadetection for the UNIONS survey.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors