IRAF Help page for: XOFFSET
xoffset - Performs angular cross-correlations
between pairs of
simultaneous velocity images (one being a "test"
image, the other
being a "reference" image). The output is an ASCII
table giving the
angular orientation of the test image, under
the assumption that
the orientation of the reference image is known.
NOTE: XOFFSET was designed as one component
of COPIPE, however, it
can be used as a stand-alone routine.
xoffset input reference
List of test images having unknown
camera offset angle
List of reference images having
(presumed) known camera offset
geometry = "hgeom"
Tells the code
which geometry information it should use:
If geometry = fndlmb,
the routine will read the following
keywords from each image header:
FNDLMBXC - the
x-coordinate of the solar image center
FNDLMBYC - the
y-coordinate of the solar image center
FNDLMBAN - the
position angle (in degrees)
FNDLMBMA - estimate
for the semi-major axis in pixels
FNDLMBMI - estimate
for the semi-minor axis in pixels
- the pixel length along the x-direction
- the pixel length along the y-direction
If geometry =
hgeom, the routine will use the values of C_MA
and C_MI (as determined
by HGEOM) for the semi-major and
ref_offset = 90.
The camera offset
angle (degrees) of the reference images if
not in header.
The routine will attempt to read
the value of the keyword OFFSET
from each reference
image header. If this keyword does not
exist, the angle
set by ref_offset will be used as
the angle OFFSET is given in degrees measured
clock-wise from the y-axis and
represents our best estimate of
the direction to Solar North.
direction = "clock"
Direction of positive rotation
=> counter clock-wise
nbins = 1024
Number of angular bins (must be
The image will
be devided into nannuli equal-area annuli
starting at R/Ro = rmin, out to
R/Ro = 1.0. Each annulus is
then divided into
nbins angular bins, or sectors. Because
FFT's are used to compute
the cross-correlations between the
annulus data from each pair
of input images, the number of bins
must be even (a power of 2 is
the most efficient).
fint_fact = 1
Fourier interpolate cross-correlation
One can obtain slightly higher
precision when determining the
the cross-correlation spectra by using Fourier
interpolation. This keyword allows
the user to set the amount
of interpolation used.
nannuli = 12
Number of annuli to cross-correlate
rmin = 0.7
Fractional radius at inner edge
of inner-most annulus
width = 5.
Width of the
zone to search for the correlation peak around
Once the cross-correlation between
two annuli from each pair of
images is computed, we need
to find the peak in the correlation
spectrum. To speed things
up slightly, the region over which
the code searches for the
correlation peak is restricted to +/-
width degrees around lag zero.
updhdr = no
Update input image headers with
new offset angle
output = yes
Create output file?
Output data file name.
compress = "gzip"
Compression type |none|gzip|compress|
Are the input images compressed,
and what type of compression
use_mrhiy = yes
Use mrhiy file for initial guess?
The reference images
may not have the most recently determined
offset values in their headers.
By using the latest mrhiy
file, the code
can determine compute the appropriate camera
offsets without having to actually
update the reference image
headers (which could
be time consuming if the images are
mrhiy_tbl = "mrhiy010730"
Full path & name of mrhiy
Where to find the latest mrhiy
file (full path and name)
use_pm = no
Use recent pm/drift scan data?
This switch tells the code
to check the pm_info file (see
below) for any recent PM events
at the reference site.
pm_info = "home$drift/drift.dat"
Full path & name of file with
recent drift scan results
verbose = no
Operate in verbose mode?
Data Product Code Guide |
Back to GRASP Help Page