The first two pixel axes (x and y) in the output cube correspond to the pixel axes in the input image. The third pixel axis (z) in the output cube is proportional to data value in the input image. The value of a pixel in the output cube measures the difference between the data value implied by its z-axis position, and the data value of the corresponding pixel in the input image. Two schemes are available (see Parameter MODE): the output pixel values can be either simply the difference between these two data values, or the difference divided by the standard deviation at the corresponding pixel in the input image (as determined either from the VARIANCE component in the input NDF or by Parameter SIGMA).
"Data" — the value of each output pixel is equal to the difference between the data value implied by
its position along the data value axis, and the value of the corresponding pixel in the input
"Sigma" — this is the same as
"Data" except that the output pixel values are divided by the standard
deviation implied either by the VARIANCE component of the input image, or by the SIGMA
lower,upper — You can supply explicit lower and upper limiting values. For example,
would set the lower limit on the output data axis to 10 and its upper limit to 200. No method name
prefixes the two values. If only one value is supplied, the
"Range" method is adopted.
The limits must be within the dynamic range for the data type of the input NDF array
"Percentiles" — The default values for the output data axis range are set to the specified
percentiles of the input data. For instance, if the value
"Per,10,99" is supplied, then the
lowest 10% and highest 1% of the data values are beyond the bounds of the output data
value axis. If only one value, p1, is supplied, the second value, p2, defaults to (100 - p1).
If no values are supplied, the values default to
"5,95". Values must be in the range 0 to
"Range" — The minimum and maximum array values are used. No other sub-strings are needed by
this option. Null (
!) is a synonym for the
"Sigmas" — The limits on the output data value axis are set to the specified numbers of standard
deviations below and above the mean of the input data. For instance, if the supplied value is
"sig,1.5,3.0", then the data value axis extends from the mean of the input data minus 1.5
standard deviations to the mean plus 3 standard deviations. If only one value is supplied,
the second value defaults to the supplied value. If no values are supplied, both default to
The limits adopted for the data value axis are reported unless parameter RANGE is specified on the command line. In this case values are only calculated where necessary for the chosen method.
The method name can be abbreviated to a single character, and is case insensitive. The initial default
"Range". The suggested defaults are the current values, or
! if these do not exist.
"Sigma". If a null (
!) value is supplied, the standard deviations implied by the VARIANCE component in the input image are used (an error will be reported if the input image does not have a VARIANCE component). If a SIGMA value is supplied, the same value is used to scale all output pixels.
m31.sdfcontains a VARIANCE component, this will create a three-dimensional NDF called m31-cube in which the third pixel axis corresponds to data value in NDF m31, and each output pixel value is the number of standard deviations of the pixel away from the corresponding input data value. If you then use Gaia to view the cube, an iso-surface at value zero will be a carpet plot of the data values in m31, an iso-surface at value
-1.0will be a carpet plot showing data values one standard deviation below the m31 data values, and an iso-surface at value
+1.0will be a carpet plot showing data values one sigma above the m31 data values. This can help to visualise the errors in an image.