Reference

Commands & recipes

Everything AstroVista Lite does with data is a command. The buttons type them for you and the terminal echoes them, so you can learn them by watching. You can also type them, repeat them with ↑, complete them with Tab, and save them as a file that runs again. This page collects worked examples; The terminal explains how a line is written and the Command reference gives every command with all its keywords.

Images and buffers

A buffer is a numbered slot holding one image. Buffers are written [1], [2], … (or B1, B2). The list on the left shows each one with how it was made.

rd 1 galaxy.fits             # read an image into buffer 1
rd 2 frame.fits[sci]         # one image of a multi-extension file, by name
rd 3 frame.fits[sci,2]       # by name and version; [1] picks by number
fitsinfo frame.fits          # what the file holds
tv 2                         # show buffer 2
copy 1 5                     # copy buffer 1 into buffer 5
tag 5 "my copy"              # a name of your own
wd 5 copy.fits               # write buffer 5 as FITS
buffers                      # list them

FITS files are recognised by their contents, gzip-compressed too, whatever they are called.

Image arithmetic

A formula is worked out pixel by pixel, and the result goes into the buffer on the left of the =. [3] = [1] - [2] means each pixel of buffer 3 is the pixel of buffer 1 at the same place minus the pixel of buffer 2 there.

Worked examples

[2] = [1] - median([1]) tag="no sky"     # subtract a constant sky
[3] = [1] / [5]                          # image / model (the model in buffer 5)
[4] = -2.5 * log10([1] / [2])            # a colour: magnitudes of two filters
[5] = [1] / sqrt(abs([1]) + 25)          # a rough signal-to-noise map
[6] = median([1], [2], [3]) tag=stack    # combine three exposures, NaN left out
[7] = ([1] - [8]) / [8] * 100            # the difference in per cent
[8] = div([1], [2], 0)                   # divide, 0 where buffer 2 is 0
set gain=2.4
[9] = [1] * gain tag=electrons           # counts to electrons

Conditions

[5] = [1] > 1000                    # 1 where brighter than 1000, else 0
[6] = where([1] > 1000, [1], 0)     # keep the bright pixels, 0 elsewhere
[7] = where([1] > 1000, nan, [1])   # hide the bright pixels (NaN)
[8] = ([1] > 10) * ([1] < 100)      # both true: multiply the conditions ("and")
[4] = [1] * mymask                  # NaN on the masked pixels
[4] = where(masked(mymask), 0, [1]) # 0 on the masked pixels

One image or several

mean median sum min max std work two ways. With several values they work pixel by pixel across the images: median([1], [2], [3]) combines exposures. With one buffer they give a single number: median([1]) is the median of all of buffer 1.

Other functions, pixel by pixel: abs sqrt exp log log10 sin cos tan arcsin arccos arctan sinh cosh tanh floor ceil round isnan isfinite nan_to_num minimum maximum clip div good masked.

Statistics and the sky

stats 1                      # mean, median, std, min, max (masked and NaN pixels left out)
stats 1 MASK=mymask,sources  # leaving out mask layers
sky 1 MASK=mymask            # a 3-sigma clipped median of what the masks leave
max 1                        # the brightest pixel and where it is
print 1 567 562 SIZE=5       # the pixel values around column 567, row 562

Masks

A mask layer marks the pixels to leave out. In files, 0 = masked and 1 = good. Layers can be drawn with the mouse, or made with commands; ERASE unmasks instead of masking.

mask new mymask 1                   # an empty mask the size of buffer 1
mask load edges edges.fits 1        # a FITS or legacy .dmask mask
mask circle mymask 812 440 12       # a circle at X, Y with radius R
mask ellipse mymask 300 300 70 22 35    # X Y A B ANGLE
mask box mymask 10 10 60 40         # X0 Y0 X1 Y1
mask polygon mymask 1 1 50 1 25 40  # corners
mask circle mymask 812 440 12 ERASE # unmask
mask nan mymask 1                   # the pixels that are not finite
mask range mymask 1 60000 1e9       # saturated pixels
mask grow mymask 2                  # 2 pixels bigger (shrink: smaller)
mask union all mymask edges         # combine layers (intersect too)
mask save mymask mymask.fits        # FITS, 1 = good (.dmask: legacy bitmap)
undo                                # and redo

Coordinates are ELLIPROF's: the first pixel spans 0 to 1, so its centre is at 0.5. help mask lists every action.

The source finder

It detects and measures the stars and galaxies in an image with SEP, which implements SExtractor's algorithms. Settings and catalogue columns keep SExtractor's names.

sources 1                                    # find them
sources 1 DETECT_THRESH=2.5 DETECT_MINAREA=8 # with your own settings
sources 1 MASK=edges TOSEG=6                 # leave out a mask; the segmentation into buffer 6
runs                                         # every run, numbered
centre 1 SOURCE=12                           # ELLIPROF's centre from source 12
mask new sources 1
mask sources sources SCALE=1.5               # mask every source but the galaxy
mask new stars 1
mask sources stars WHERE='FLUX_RADIUS < 2.5' # only the sources matching a condition
mask segments sources GROW=2                 # their pixels, grown by 2

Isophotes with ELLIPROF

elliprof 1 X0=567 Y0=562 R0=9 R1=347 NR=23 NITER=10 SKY=3246 RMSTAR MASK=u12517j,sources TOMODEL=3 TORESID=4

Image commands

In the style of the VISTA and MONSTA command languages, written for AstroVista Lite:

smooth 1 FWHM=3              # a Gaussian smoothing
zap 1 SIG=5                  # remove cosmic rays and hot pixels
rotate 1 ANGLE=90            # multiples of 90 are exact
gauss 1 X=200 Y=200 FWHM=2.5 PEAK=5000   # add an artificial star
help smooth                  # any command explained

Command files and variables

A command file is plain text, one command per line, with # for comments. File names inside it are relative to the file's own folder, so it can be shared with its data.

set xc=567 yc=562
elliprof 1 X0=xc Y0=yc R0=9 R1=347 NR=23
typ                          # show the variables
run session.txt              # run a file (or: astrovista --run session.txt)

File ▸ Save the commands writes the session so far. Programs on the PATH run in the terminal too (ls, grep, …); a line starting with ! always goes to the shell.

Recipes

Measure a galaxy's profile

Open the image ▸ Source finder ▸ Find sources ▸ ELLIPROF: Pick on the image (it snaps to the galaxy), Pick R1 where it fades ▸ Estimate the sky ▸ Mask the found sources ▸ Fit isophotes ▸ Results ▸ Profile, tick R¼ law.

See what the model misses

After a fit, tick Blink, choose the image and the residual, and press ← →. Or [9] = [1] / [3] tag="ratio", with the model in buffer 3.

Mask a bright source by hand

Choose the circle tool (M), set its radius, and click the source. Alt-click unmasks; Ctrl+Z undoes.

Find only bright, compact sources

Raise the detection threshold in the Source finder, or use Masks ▸ Mask the sources where… with MAG_AUTO < 18 and FWHM_IMAGE < 3.

Subtract the sky and save

[2] = [1] - median([1]) tag="no sky"
wd 2 nosky.fits

Combine exposures

rd 1 exp1.fits
rd 2 exp2.fits
rd 3 exp3.fits
[4] = median([1], [2], [3]) tag=stack
wd 4 stack.fits