create.sim: Create Simulated Image

In AngusWright/LAMBDAR: Measure galaxy fluxes from an arbitrary FITS image using input aperture locations and dimensions

Create Simulated Image

Description

Create Simulate Photometric Image from an arbitrary FITS image. This function takes the input image and catalogue from the measure.fluxes Parameter File and uses them to generate a simulated image. Objects that are point sources are modelled with a perfect analytical gaussian, with FWHM as specified by the user or as determined empirically from the input PSF. create.sim.image is a subroutine within create.sim and is not designed to be called directly.

Usage

create.sim(par.file=NA, ObsParm=NA, noNoise=FALSE,convolveNoise=TRUE,
padGalaxies=TRUE, colourCorr=0, quiet=FALSE, confuse=FALSE)
create.sim.image(ObsParm, noNoise=FALSE, convolveNoise=TRUE,
padGals=TRUE, col.corr=0,outenv=parent.env(environment()),
confuse=FALSE, env=NULL)

Arguments

par.file	Path to, and name of, the Input Parameter file (ie. ‘/path/to/par.file.par’). This is of the same as the measure.fluxes Parameter File, however only some parameters are relevant.
ObsParm	Data frame containing Observation Parameters needed in creating simulated galaxies; Exposure Time (sec; exp=), Collecting Area (meters squared; area=), Effective wavelength of filter (Angstrom; lamEff=), and Effective width of filter (Angstrom; WEff=). For example, SDSS r-band has parameters ObsParm=data.frame(exp=53.9, area=4.91, lamEff=1111.2, Weff=6165).
noNoise	logical; if TRUE then noise is not added to the simulated image
convolveNoise	logical; if TRUE then noise is convolved with the PSF prior to addition
padGalaxies	logical; if TRUE then the simulated image will have galaxies added beyond those in the input catalogue, such that the galactic number counts follow the polynomial log(N)=-0.01476*m_r^2+1.025*m_r-11.76.
colourCorr	numeric; The colour correction need to convert SDSS r-band magnitude to magnitude in the band being analysed. This is needed when padding galaxies because our power law is defined in the r-band. Naturally, this is a 0th order correction, and should be noted when determining the accuracy of simulations that are well beyond the SDSS r-band.
confuse	logical; do you want to confuse the simulation such that the source density is greater than 1 object per FWHM^2
col.corr	same as the colourCorr argument above
padGals	same as the padGalaxies argument above
outenv	environment; parameter space to output variables to
quiet	logical; if TRUE all output to screen is suppressed
env	environment;; parameter space containing the input measure.fluxes/LAMBDAR core variables

Details

create.sim is an additional routine included in the LAMBDAR package for the purposes of testing. The LAMBDAR package is intended for use in astronomy, and other related fields, to determine photometry from input FITS photometric images of arbitrary resolution. As such, testing of the program is paramount. This function allows the user to generate simulated images from input galactic apertures and fluxes. Using the known input, the user may then test the measure.fluxes routine with all its various options permuted, thus testing the programs accuracy. Results of this type of rigorous testing are demonstrated in Wright et al. 2015 (in prep).

Value

NULL; create.sim will output a simulated image to "sim_image.fits", and a catalogue of the created objects called "SimFlux.csv".

Author(s)

Angus H Wright ICRAR [e-mail]

See Also

create.par.filelsoswrite.fitsmeasure.fluxes

Examples

#Load LAMBDAR
library(LAMBDAR)

#Create simulated image using an SDSS r-band image and observation parameters:
#Load a Sample Image and Catalogue
data("SDSS.sample",envir=environment())
data("ApCat.sample",envir=environment())
lsos(envir=environment())
#Write Sample Image to File
write.fits(file="SampleImage.fits",SDSS.sample)
#Write Sample Catalogue to File
write.csv(file="SampleCat.csv",ApCat.sample,row.names=FALSE,quote=FALSE)
#Generate Parameter File
create.par.file(file="Lambdar_Sample.par",DataMap="SampleImage.fits",
Catalogue="SampleCat.csv",IterateFluxes=1,nIterations=2,Gauss_FWHM_AS=2,
PSFConvolve=1,GainLabel='WTGAIN',SaturationLabel='SWPSATUR')

#Measure Photometry
measure.fluxes("Lambdar_Sample.par")

#Generate Simulation Parameter File
create.par.file(file="Lambdar_Simulation.par",DataMap="SampleImage.fits",
Catalogue="LAMBDAR_Results.csv",Gauss_FWHM_AS=2,SimGauss_AS=1.1,
FluxWgtColumnLabel="DFAFlux_Jy",OutputDirectory="Simulation/",PSFConvolve=1,
GainLabel='WTGAIN',SaturationLabel='SWPSATUR',FluxWgtType='flux')

#Create Simulation from Catalogued objects only
create.sim('Lambdar_Simulation.par',ObsParm=data.frame(exp=53.9,area=4.91,
lamEff=1111.2,Weff=6165),convolveNoise=TRUE,padGalaxies=TRUE)

#Read the Input Image
inp.im<-read.fits("SampleImage.fits")
#Read the Simulated Image
sim.im<-read.fits("Simulation/sim_image.fits")

#Plot the Input and Simulation Images
layout(matrix(1:4,2,2,byrow=TRUE))
#Get Object Positions
xy<-ad.to.xy(ApCat.sample$RAdeg,ApCat.sample$DECdeg,
read.astrometry("SampleImage.fits"))
aspp<-abs(read.astrometry("SampleImage.fits")$CD[1,1])*3600
#Plot Input & Aperture Ellipses
image(x=1:length(inp.im$dat[[1]][,1]),y=1:length(inp.im$dat[[1]][1,]),
inp.im$dat[[1]],col=grey.colors(1E3),zlim=quantile(inp.im$dat[[1]],
c(0,0.999)),asp=1,useRaster=TRUE,xlab='X (pix)',ylab="Y (pix)")
invisible(Vectorize(function(x,y,a,b,pa) {
   lines(ellipse(x0=x,y0=y,a=a,b=b,pa=pa),col='red',lty=2)
   points(x,y,pch=3,cex=0.5,col='red')
})(xy[,1],xy[,2],ApCat.sample$radmajasec/aspp,ApCat.sample$radminasec/aspp,
ApCat.sample$rotW2N))
#Plot Simulation & Aperture Ellipses
image(x=1:length(sim.im$dat[[1]][,1]),y=1:length(sim.im$dat[[1]][1,]),
sim.im$dat[[1]],col=grey.colors(1E3),zlim=quantile(inp.im$dat[[1]],
c(0,0.999)),asp=1,useRaster=TRUE,xlab='X (pix)',ylab="Y (pix)")
invisible(Vectorize(function(x,y,a,b,pa) {
   lines(ellipse(x0=x,y0=y,a=a,b=b,pa=pa),col='red',lty=2)
   points(x,y,pch=3,cex=0.5,col='red')
})(xy[,1],xy[,2],ApCat.sample$radmajasec/aspp,ApCat.sample$radminasec/aspp,
ApCat.sample$rotW2N))
#Plot Zoomed Sections of the Image:
#-> Input & Aperture Ellipses
image(x=1:length(inp.im$dat[[1]][,1]),y=1:length(inp.im$dat[[1]][1,]),
inp.im$dat[[1]],col=grey.colors(1E3),zlim=quantile(inp.im$dat[[1]],
c(0,0.999)),asp=1,useRaster=TRUE,xlab='X (pix)',ylab="Y (pix)",
xlim=c(800,1200),ylim=c(800,1200))
points(xy[,1],xy[,2],pch=3,cex=0.5,col='red')
#-> Simulation & Aperture Ellipses
image(x=1:length(sim.im$dat[[1]][,1]),y=1:length(sim.im$dat[[1]][1,]),
sim.im$dat[[1]],col=grey.colors(1E3),zlim=quantile(inp.im$dat[[1]],
c(0,0.999)),asp=1,useRaster=TRUE,xlab='X (pix)',ylab="Y (pix)",
xlim=c(800,1200),ylim=c(800,1200))
points(xy,pch=3,cex=0.5,col='red')

#Measure Simulation Fluxes
create.par.file(file="Lambdar_SimAnalysis.par",
DataMap="Simulation/sim_image.fits",Catalogue="Simulation/SimFlux.csv",
Gauss_FWHM_AS=2,OutputDirectory="SimResults/",PSFConvolve=1,BlankCor=1,
GainLabel='WTGAIN',SaturationLabel='SWPSATUR')
measure.fluxes('Lambdar_SimAnalysis.par')

#Read simulation results
sim.out.cat<-read.csv("SimResults/LAMBDAR_Results.csv",stringsAsFactors=FALSE)
sim.in.cat.full<-read.csv("Simulation/SimFlux.csv",stringsAsFactors=FALSE)
sim.in.cat<-sim.in.cat.full[which(sim.in.cat.full$CATAID %in% sim.out.cat$CATAID),]

#Plot Flux Comparisons
layout(c(1,2))
magplot(sim.in.cat$Sim.Flux.Jy, sim.out.cat$DFAFlux_Jy / sim.in.cat$Sim.Flux.Jy,
pch=20, log='xy', ylim=c(0.5,2),majorn=10,minorn=10,ylab='Measured / Input Flux',
xlab='Input Flux (Jy)')
#Add Uncertainties
suppressWarnings(magerr(x=sim.in.cat$Sim.Flux.Jy,
y=sim.out.cat$DFAFlux_Jy / sim.in.cat$Sim.Flux.Jy,
ylo=abs(sim.out.cat$DFAFlux_Jy / sim.in.cat$Sim.Flux.Jy)*
as.numeric(sim.out.cat$DFAErr_Jy)/sim.out.cat$DFAFlux_Jy))
abline(h=1,col='grey')
abline(h=0.96,col='blue',lty=2)
legend('topright',legend=c("Expected Fraction\n(2.5Kron w/ exponential profile)"),
col='blue',lty=2)

#Improve Flux Measurement by Iterating
create.par.file(file="Lambdar_SimAnalysis.par",
DataMap="Simulation/sim_image.fits",Catalogue="Simulation/SimFlux.csv",
Gauss_FWHM_AS=2,OutputDirectory="SimResults/",IterateFluxes=1,
PSFConvolve=1,nIterations=5,BlankCor=1,
GainLabel='WTGAIN',SaturationLabel='SWPSATUR')
measure.fluxes('Lambdar_SimAnalysis.par')

#Read simulation results
sim.out.cat<-read.csv("SimResults/LAMBDAR_Results.csv",stringsAsFactors=FALSE)
sim.in.cat<-sim.in.cat.full[which(sim.in.cat.full$CATAID %in% sim.out.cat$CATAID),]

#Plot Flux Comparisons
magplot(sim.in.cat$Sim.Flux.Jy, sim.out.cat$DFAFlux_Jy / sim.in.cat$Sim.Flux.Jy,
pch=20, log='xy', ylim=c(0.5,2),majorn=10,minorn=10,
ylab='Measured / Input Flux',xlab='Input Flux (Jy)')
#Add Uncertainties
suppressWarnings(magerr(x=sim.in.cat$Sim.Flux.Jy,
y=sim.out.cat$DFAFlux_Jy / sim.in.cat$Sim.Flux.Jy,
ylo=abs(sim.out.cat$DFAFlux_Jy / sim.in.cat$Sim.Flux.Jy)*
as.numeric(sim.out.cat$DFAErr_Jy)/sim.out.cat$DFAFlux_Jy))
abline(h=1,col='grey')
abline(h=0.96,col='blue',lty=2)
legend('topright',legend=c("Expected Fraction\n(2.5Kron w/ exponential profile)"),
col='blue',lty=2)

AngusWright/LAMBDAR documentation built on Sept. 19, 2024, 10:26 a.m.