Imstat
From CASA Guides
Help on imstat task:
Displays statistical information from an image or image region
Many parameters are determined from the specified region of an image.
For this version, the region can be specified by a set of rectangular
pixel coordinates, the channel ranges and the Stokes.
For directed output, run as
myoutput = imstat()
Keyword arguments:
imagename -- Name of input image
Default: none; Example: imagename='ngc5921_task.im'
region -- File path to an ImageRegion file or name.
Use the viewer, then region manager to select regions of
the image to process. Similar to box, but graphical
Or the name of a region stored with the image,
use rg.namesintable()
to retrieve the list of names.
Default: none
Example: region='myimage.im.rgn'
region='region1'
box -- A box region on the directional plane
Only pixel values acceptable at this time.
Default: none (whole 2-D plane);
Example: box='10,10,50,50'
box = '10,10,30,30,35,35,50,50' (two boxes)
chans -- channel numbers
Range of channel numbers to include in statistics
All spectral windows are included
Default:''= all; Example: chans='3~20'
stokes -- Stokes parameters to analyze.
Default: none (all); Example: stokes='IQUV';
Example:stokes='I,Q'
Options: 'I','Q','U','V','RR','RL','LR','LL','XX','YX','XY','YY', ...
General procedure:
1. Specify inputs, then
2. myoutput = imstat()
or specify inputs directly in calling sequence to task
myoutput = imstat(imagename='image.im', etc)
3. myoutput['KEYS'] will contain the result associated with any
of the keys given below
KEYS CURRENTLY AVAILABLE
blc - absolute PIXEL coordinate of the bottom left corner of
the bounding box surrounding the selected region
blcf - Same as blc, but uses WORLD coordinates instead of pixels
trc - the absolute PIXEL coordinate of the top right corner
of the bounding box surrounding the selected region
trcf - Same as trc, but uses WORLD coordinates instead of pixels
flux - the integrated flux density if the beam is defined and
the if brightness units are $Jy/beam$
npts - the number of unmasked points used
max - the maximum pixel value
min - minimum pixel value
maxpos - absolute PIXEL coordinate of maximum pixel value
maxposf - Same as maxpos, but uses WORLD coordinates instead of pixels
minpos - absolute pixel coordinate of minimum pixel value
minposf - Same as minpos, but uses WORLD coordinates instead of pixels
sum - the sum of the pixel values: $\sum I_i$
sumsq - the sum of the squares of the pixel values: $\sum I_i^2$
mean - the mean of pixel values:
ar{I} = \sum I_i / n$
sigma - the standard deviation about the mean:
$\sigma^2 = (\sum I_i -ar{I})^2 / (n-1)$
rms - the root mean square:
$\sqrt {\sum I_i^2 / n}$
median - the median pixel value (if robust=T)
medabsdevmed - the median of the absolute deviations from the
median (if robust=T)
quartile - the inter-quartile range (if robust=T). Find the points
which are 25% largest and 75% largest (the median is
50% largest), find their difference and divide that
difference by 2.
Additional Examples
# Selected two box region
# box 1, bottom-left coord is 2,3 and top-right coord is 14,15
# box 2, bottom-left coord is 30,31 and top-right coord is 42,43
imstat( 'myImage', box='2,3,14,15;30,31,42,43' )
# Select the same two box regions but only channels 4 and 5
imstat( 'myImage', box='2,3,14,15;30,31,42,43', chan='4~5' )
# Select all channels greater the 20 as well as channel 0.
# Then the mean and standard deviation are printed
results = imstat( 'myImage', chans='>20;0' )
print "Mean is: ", results['mean'], " s.d. ", results['sigma']
max - the maximum pixel value
min - minimum pixel value
maxpos - absolute PIXEL coordinate of maximum pixel value
maxposf - Same as maxpos, but uses WORLD coordinates instead of pixels
minpos - absolute pixel coordinate of minimum pixel value
minposf - Same as minpos, but uses WORLD coordinates instead of pixels
sum - the sum of the pixel values: $\sum I_i$
sumsq - the sum of the squares of the pixel values: $\sum I_i^2$
mean - the mean of pixel values:
ar{I} = \sum I_i / n$
sigma - the standard deviation about the mean:
$\sigma^2 = (\sum I_i -ar{I})^2 / (n-1)$
rms - the root mean square:
$\sqrt {\sum I_i^2 / n}$
median - the median pixel value (if robust=T)
medabsdevmed - the median of the absolute deviations from the
median (if robust=T)
quartile - the inter-quartile range (if robust=T). Find the points
which are 25% largest and 75% largest (the median is
50% largest), find their difference and divide that
difference by 2.
Additional Examples
# Selected two box region
# box 1, bottom-left coord is 2,3 and top-right coord is 14,15
# box 2, bottom-left coord is 30,31 and top-right coord is 42,43
imstat( 'myImage', box='2,3,14,15;30,31,42,43' )
# Select the same two box regions but only channels 4 and 5
imstat( 'myImage', box='2,3,14,15;30,31,42,43', chan='4~5' )
# Select all channels greater the 20 as well as channel 0.
# Then the mean and standard deviation are printed
results = imstat( 'myImage', chans='>20;0' )
print "Mean is: ", results['mean'], " s.d. ", results['sigma']
# Find statistical information for the Q stokes value only
# then the I stokes values only, and printing out the statistical
# values that we are interested in.
s1 = imstat( 'myimage', stokes='Q' )
s2 = imstat( 'myimage', stokes='I' )
print " | MIN | MAX | MEAN"
print " Q | ",s1['min'][0]," | ",s1['max'][0]," | ",," | ",s1['mean'][0]
print " I | ",s2['min'][0]," | ",s2['max'][0]," | ",," | ",s2['mean'][0]
