Next: New synonyms
Up: Release Notes V0.9
Previous: Global changes
- Creates a text file which describes
selected regions of a previously displayed image. The file is
in the ARD format. There are eleven shapes, and 5 logical
operators for combining regions. There are options, amongst others,
for listing and deleting regions .
- Uses an ARD file, such as created
by ARDGEN, to define regions of an NDF to mask. It therefore
allows regions to be excluded from subsequent data processing.
- Sets a new LABEL component for an
- Sets a new UNITS component for an
- Evaluates a Fortran-like mathematical
expression. This is particularly useful for non-integer
arithmetic in C-shell scripts. Parameters and sub-expressions are
- Calculates various polarisation
parameters from four NDFs containing intensity arrays analysed at
, and 135
from a reference direction.
The resultant parameters are stored in NDFs, of which there are
- Convolves a one- or two-dimensional
NDF using a function given by a second NDF. The latter is
normally a point-spread function, and need not have the same
dimensions as the convolved NDF.
- Creates a radial or azimuthal
profile of a two-dimensional NDF. It bins the image into elliptical
annuli or a fan-shaped region of adjacent sectors. The mean
values in each bin are stored in a one-dimensional output NDF.
There are options to restrict the range of radial distance and
- Removes pixels with errors larger
than some limit from an NDF by flagging them with the bad value in
a copy NDF.
- Removes regions of bad values
from a two-dimensional NDF using a smooth function which matches the
- Moves ancillary data within an
NDF from arbitrary extensions to the FITS extension. It uses a
keyword translation table held in a text file.
- Creates an NDF FITS extension
from a text file. There is limited validation of the headers.
- Fits a polynomial surface
to two-dimensional NDF. It does not bin the data and does not reject
outliers to the fit. It stores the result within a POLYNOMIAL data
structure called SURFACEFIT within the fitted NDF.
- Adds commentary to the history
records of an NDF. The text can be supplied through a parameter, or
a text file. There is optional control of paragraph wrapping.
- Lists all the history records in
an NDF. The output comprises the date, time, and application name, and
optionally the history text.
- Sets the history update mode for
an NDF. History recording disabled, or set to record at three
levels of verbosity. Existing history records can also be deleted.
- Performs a Lucy-Richardson deconvolution
of a one- or two-dimensional NDF with a point-spread function. It uses an
iterative process to provide new estimates of the restored,
higher-resolution array. The closeness of the fit is determined
per pixel. The algorithm terminates
after a specified number of iterations or when a certain
has been reached. The Synder correction may be applied
using the data variance.
- Creates a two-dimensional NDF
from the coefficients of a polynomial surface stored in a SURFACEFIT
extension (as written by FITSURFACE). The size and extent of the
output NDF may be defined by a template NDF.
- Obtains the value or values of an
application parameter. This offers an easier way to pass parameter
values between tasks in shell scripts.
- Pastes a series of NDFs upon each
other. Bad values may be transparent or opaque. The NDFs need not
be congruent or even have the same number of dimensions.
- Uses a cursor to select the current
picture and to report the co-ordinates of points. Some of the
functions of CURSOR have been moved to this task.
- Finds the first empty FRAME
picture in the graphics database. This is
most useful with PICGRID to fill in a grid of FRAME pictures
without having to know or compute their labels.
- Finds the first unobscured
and unobscuring FRAME picture in the graphics database. This tries
to locate a FRAME picture such that additional plotting within it
will not affect earlier plots.
- Transforms co-ordinates between
the current and BASE pictures of the graphics database.
- Finds the first unobscured FRAME
picture in the graphics database.
- Adds a constant to a sub-section of an
NDF to correct rift-valley defects.
- Sets the values for an axis array
component of an NDF. There are several options which permit the
deletion of the axis system, or an individual variance or width
array; the replacement of one or more individual values; the
assignment of the whole array using a Fortran-like mathematical
expression, or from a text file, or set to pixel co-ordinates.
- Manipulates the contents of a
specified NDF extension. There are six options including erasure,
renaming, and assignment of extension components.
- Sets one or more of the
axis-normalisation flags in an NDF.
- Makes an IRAS astrometry extension
for a two-dimensional NDF, thereby allowing sky co-ordinate
information to be stored with an arbitrary image. The co-ordinate
system can be defined by suppling explicit projection parameters, or
by providing a list of sky and corresponding image co-ordinates, or
a combination of both. This task enables non-IRAS data to use the
tasks in the IRAS90 package which perform
various astrometric tasks such as plotting axes and overlay grids in
- Inverts a transformation
stored in a TRANSFORM structure within an existing HDS file.
- Joins two transformation stored
in the TRANSFORM format.
- Makes a transformation structure
from supplied forward and inverse mappings. The structure can
have classification qualifiers and a comment. The mappings are
defined using Fortran-like expressions, and may include
sub-expressions and constants. There are also preset
two-dimensional transformations: polar about a supplied origin, and
the six-term linear which encompasses shift of origin, rotation,
magnification, and shear.
- Applies a transformation
stored in TRANSFORM structure to an NDF by resampling. There is
full control of the shape, origin, and co-ordinate limits of the
transformed NDF. The resampling use either linear interpolation or
the nearest-neighbour method.
- Lists the contents of a
- Plots a two-dimensional vector map
of an NDF. The vectors may be arrows or lines, and there is
control of the frequency, justification, and scaling of the
vectors. There is an optional key, and many parameters for
customising the appearance of the plot.
- Applies a Wiener filter to a one- or two-dimensional
array. It offers a much faster, but cruder, deconvolution method.
This task uses a `white' noise model or a power spectrum supplied
in another NDF. The noise power is adjustable.
Next: New synonyms
Up: Release Notes V0.9
Previous: Global changes
KAPPA --- Kernel Application Package
Starlink User Note 95
Malcolm J. Currie & David S. Berry
2013 February 14
Copyright © 2013 Science and Technology Facilities Council