Fake Source Planting

Demonstration of planting fake PSFs in diff images to mimic strongly-lensed supernovae.

import os
import numpy as np
import diffimageml
from matplotlib import pyplot as plt

Out:

Created TAP+ (v1.2.1) - Connection:
        Host: gea.esac.esa.int
        Use HTTPS: True
        Port: 443
        SSL Port: 443
Created TAP+ (v1.2.1) - Connection:
        Host: geadata.esac.esa.int
        Use HTTPS: True
        Port: 443
        SSL Port: 443

Fake Source Planting Overview

From a trio of images (template, search, diff), we

1. find galaxies in the template image

2. build a PSF model from the search image

3. plant Fake PSFs in the diff image near the galaxy locations

Setup 1: Get the Data

Load in a trio of fits images from the example data dir Pull them together into a FakePlanter triplet

example_data_dict = diffimageml.get_example_data()
assert(os.path.isfile(example_data_dict['diffim1']))
assert(os.path.isfile(example_data_dict['searchim1']))
assert(os.path.isfile(example_data_dict['templateim1']))

fakeplantertrio = diffimageml.FakePlanter(
    example_data_dict['diffim1'],
    searchim_fitsfilename=example_data_dict['searchim1'],
    templateim_fitsfilename=example_data_dict['templateim1'])

print("FakePlanter Trio constructed.")
assert(fakeplantertrio.searchim.has_fakes == False)
assert(fakeplantertrio.diffim.has_fakes == False)
print("  (No fakes yet)")

Out:

/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/astropy/wcs/wcs.py:709: FITSFixedWarning: 'obsfix' made the change 'Set OBSGEO-L to   149.070647 from OBSGEO-[XYZ].
Set OBSGEO-B to   -31.272933 from OBSGEO-[XYZ].
Set OBSGEO-H to     1164.993 from OBSGEO-[XYZ]'.
  FITSFixedWarning)
FakePlanter Trio constructed.
  (No fakes yet)

Setup 2: Make the PSF Model

Measure the zero point and build the ePSF model from Gaia stars See the other example for details and plots. In practice, this code will load an existing ePSF model from the example data directory.

fakeplantertrio.searchim.fetch_gaia_sources(overwrite=False)
fakeplantertrio.searchim.do_stellar_photometry(
    fakeplantertrio.searchim.gaia_source_table)
fakeplantertrio.searchim.measure_zeropoint(showplot=False)
fakeplantertrio.searchim.build_epsf_model(
    verbose=False, save_suffix='TestEPSFModel')

Out:

Gaia catalog /Users/rodney/Dropbox/src/diffimageml/diffimageml/test_data/sky_image_1_GaiaCat.ecsv exists.
Reading without fetching.
/Users/rodney/Dropbox/src/diffimageml/diffimageml/fakeplanting.py:676: RuntimeWarning: invalid value encountered in log10
  phot['mag'] = -2.5 * np.log10( phot['aper_sum_bkgsub'] )
/Users/rodney/Dropbox/src/diffimageml/diffimageml/fakeplanting.py:752: RuntimeWarning: divide by zero encountered in true_divide
  (np.abs(star_flux/star_flux_err)>20))
/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/numpy/core/fromnumeric.py:748: UserWarning: Warning: 'partition' will ignore the 'mask' of the MaskedArray.
  a.partition(kth, axis=axis, kind=kind, order=order)
/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/numpy/core/fromnumeric.py:748: UserWarning: Warning: 'partition' will ignore the 'mask' of the MaskedArray.
  a.partition(kth, axis=axis, kind=kind, order=order)
/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/numpy/core/fromnumeric.py:748: UserWarning: Warning: 'partition' will ignore the 'mask' of the MaskedArray.
  a.partition(kth, axis=axis, kind=kind, order=order)
/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/numpy/core/fromnumeric.py:748: UserWarning: Warning: 'partition' will ignore the 'mask' of the MaskedArray.
  a.partition(kth, axis=axis, kind=kind, order=order)
/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/numpy/core/fromnumeric.py:748: UserWarning: Warning: 'partition' will ignore the 'mask' of the MaskedArray.
  a.partition(kth, axis=axis, kind=kind, order=order)

Plant Fakes

Here we plant just 10 very bright fakes

# detect sources in the template image, identify likely galaxies
fakeplantertrio.templateim.detect_sources()
hostgaltable = fakeplantertrio.templateim.detect_host_galaxies()


# Make 10 locations for random fakes (each relative to a galaxy center point)
Nfakes = 10
phi = np.random.uniform(0, 360, Nfakes)
d = np.random.uniform(0, 5, Nfakes)
fluxes = np.random.uniform(10**2, 10**4, Nfakes)

# fix the positions of the fakes in x,y coordinates on the diff image
# This returns three tables: one each for the diffim, searchim, and templateim
fake_positions_and_fluxes = fakeplantertrio.set_fake_positions_at_galaxies(
    phi, d, fluxes)

# Grab the existing ePSF model from the search image
epsfmodel = fakeplantertrio.searchim.epsf

# Plant the fakes
fakeplantertrio.plant_fakes_triplet(
    fake_positions_and_fluxes, psfmodel=epsfmodel,
    writetodisk=False, save_suffix="planted.fits")

print("Fake planting is done.")
assert(fakeplantertrio.diffim.has_fakes==True)
assert(fakeplantertrio.searchim.has_fakes==True)
print(" has_fakes is True, True!")

Out:

/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/astropy/wcs/wcs.py:709: FITSFixedWarning: 'obsfix' made the change 'Set OBSGEO-L to   149.070647 from OBSGEO-[XYZ].
Set OBSGEO-B to   -31.272933 from OBSGEO-[XYZ].
Set OBSGEO-H to     1164.993 from OBSGEO-[XYZ]'.
  FITSFixedWarning)
/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/astropy/wcs/wcs.py:709: FITSFixedWarning: 'obsfix' made the change 'Set OBSGEO-L to   149.070647 from OBSGEO-[XYZ].
Set OBSGEO-B to   -31.272933 from OBSGEO-[XYZ].
Set OBSGEO-H to     1164.993 from OBSGEO-[XYZ]'.
  FITSFixedWarning)
/usr/local/anaconda3/envs/astroconda/lib/python3.7/site-packages/astropy/wcs/wcs.py:709: FITSFixedWarning: 'obsfix' made the change 'Set OBSGEO-L to   149.070647 from OBSGEO-[XYZ].
Set OBSGEO-B to   -31.272933 from OBSGEO-[XYZ].
Set OBSGEO-H to     1164.993 from OBSGEO-[XYZ]'.
  FITSFixedWarning)
Fake planting is done.
 has_fakes is True, True!

Display Fakes

Show a few examples of fakes from the diff image, using a few random indices from the list of fakes

fakeIDs, fake_positions =  fakeplantertrio.get_fake_locations()

rng = np.random.default_rng()
fakeids_to_show = rng.choice(fakeIDs, 3)
print(fakeids_to_show)

Out:

[8 1 0]

Display the fakes in the diff image

fakeplantertrio.plot_fakes(
    fake_indices=fakeids_to_show)

TODO: Make a trio of postage stamps for each fake

Show the trio of fakes for each