This application creates an NDF
from a supplied catalogue. The output NDF can be either a simple 1-dimensional list of
column values without any spatial information, or it can be a 2 or 3 dimensional array in
which column values retain their spatial positions (see parameter SHAPE). The columns
containing data value and (optionally) variance are specified using parameters COLDAT and
COLVAR.
If parameter SHAPE is set TRUE, a 2 or 3 dimensional NDF is created in which the spatial position of
each data value is read from the catalogue columns specified using parameters COLX, COLY and
COLZ. The NDF is formed by binning the catalogue values into a grid of equally sized rectangular
cells, the dimensions of each cell being given by parameter BOX. Each pixel in the output NDF
corresponds to one of these cells. The data values for the cell are formed by combining together the
COLDAT values of all input positions which fall within the cell, using the method specified by the
parameter METHOD.
If parameter SHAPE is set FALSE, a 1D NDF is created in which the spatial position of each data value
is ignored. The data values are just copied into the 1D NDF in the same order that they
appear in the input catalogue. That is, the first value in the catalogue becomes pixel 1, the
seconds catalogue value becomes pixel 2, etc. This avoids any binning of the data values,
and is useful if applications which do not use spatial information (such as the KAPPA
applications STATS, HISTOGRAM, etc) are to be used.
BOX( 3 ) = _REAL (Read)
The x, y
and z bin sizes. These values refer to the co-ordinate Frame given by parameters COLX
and COLY. Only accessed if parameter SHAPE is TRUE. If not supplied, the third value
defaults to 1.0 and the second value defaults to the first value.
COLDAT = LITERAL
(Read)
The name of the catalogue column holding the values to be stored in the DATA
component of the output NDF. A list of available column names is displayed if a non-existent
column name is given. An arbitrary algebraic combination of columns may be used by
supplying a expression instead of a single column name. The expressions should be in the form
expected by the CURSA package - see
SUN/190 for details of the syntax.
COLVAR =
LITERAL (Read)
The name of the catalogue column holding the values to be stored in the
VARIANCE component of the output NDF. A list of available column names is displayed if a
non-existent column name is given. If a null (!) value is supplied, no VARIANCE component
is created. An arbitrary algebraic combination of columns may be used by supplying a
CURSA expression instead of a single column name. For instance, supplying the string
"
DP2"
causes the square of the values in column DP to be used. See SUN/190 for details of the syntax of
these expressions. [!]
COLX = LITERAL (Read)
The name of the catalogue column which gives the
coordinate of each data value along the first axis. A list of available column names is displayed if a
non-existent column name is given. An arbitrary algebraic combination of columns may be used by
supplying a CURSA expression instead of a single column name. See SUN/190 for details of the
syntax of these expressions. Only accessed if parameter SHAPE is TRUE. [X]
COLY = LITERAL
(Read)
The name of the catalogue column which gives the coordinate of each data value
along the second axis. See COLX for further details. [Y]
COLZ = LITERAL (Read)
The
name of the catalogue column which gives the coordinate of each data value along a third
axis. If a null (!) value is supplied the output NDF will be 2-dimensional. The dynamic
default is "
Z"
if the catalogue contains a column named "
Z"
, and is null (!) otherwise. See
COLX for further details. []
IN = LITERAL (Read)
The name of the input catalogue.
This may be in any format supported by the CAT library (see SUN/181). A file type of
.FIT is assumed if no file type is supplied.
METHOD = LITERAL (Read)
The method
to be used when binning data values. This may be set to any unique abbreviation of the
following:
-
MEAN – Mean of the input data values
-
MEDIAN – Median of the input data values
-
SIGMA – A sigma clipped mean
Only accessed if parameter SHAPE is TRUE. [MEAN]
MINVAL = _INTEGER (Read)
The minimum
number of good input values which must be present in a cell to create a good output value. Only
accessed if parameter SHAPE is TRUE. [1]
OUT = NDF (Read)
The name of the output NDF.
SHAPE = _LOGICAL (Read)
If a TRUE value is supplied for parameter SHAPE, then the output NDF
is 2 or 3-dimensional and inherits the spatial positions given in the columns specified by COLX, COLY
and COLZ. If a FALSE value is supplied, the output NDF is 1-dimensional and the spatial position of
each data value is ignored. In this case, the number of pixels in the output NDF will equal the
number of rows in the input catalogue. The data values are stored in the NDF in the same
order as in the input catalogue. [TRUE]
SIGMAS = _REAL (Read)
Number of standard
deviations to reject data at. Only used if METHOD is set to "
SIGMA"
. Only accessed if
parameter SHAPE is TRUE. [4.0]