Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/computing/software/gipsy/tsk/patch.dc1 Дата изменения: Mon May 7 12:45:18 2001 Дата индексирования: Fri Jan 16 00:07:45 2009 Кодировка: Поисковые слова: п п п р п р п

Program: PATCH

Purpose: Find data values at positions of blanks interpolating the
data in a small volume around the blanks.

Category: CALCULATION, COMBINATION, MANIPULATION

File: PATCH

Author: F.J. Sicking

Keywords:

INSET= Set and subsets containing the data.

NOISET= Set containing the errors on the data. [ask NOISE=]

NOISE= Give the noise on the data. [1.0]

OUTSET= Set to write the output to. [input set]

BOX= Area to replace blanks. For all subsets the same box is used.

BEAM= The major and minor axes (FWHM in arcseconds) of an
elliptical beam that can be used for weighting of the data
used in the interpolation. [no beam]
Only asked if the dimension of the subsets is less or equal
than two. A round beam can be specified by supplying only
one number.

BEAMPA= The position angle of the beam. Only asked, if both, the
subsets are two dimensional, and two elements were supplied
at BEAM=.

BEAMCUT= Cutoff for the beam. Positions where the value of the beam
(its central peak value is set to one) are lower than BEAMCUT=
discarded.

SIZE= Size of the area to collect the data for each individual
interpolation. The number of elements must be equal to the
number of subset axes. The values must be odd in order for
the routine to be able to centre the area on the position
of the blank. They must also be larger than one, since no
fitting can be done in a direction where the size of the
collecting area is only one pixel. If an interpolation
is wanted using only data from a lower dimensional structure
than the subset -for example a one dimensional interpolation
in a two dimensional subset- then the subset selection has
to be adapted.

NDATA= Minimum number of data points present inside the area
used for collecting the data. [0.0] The default is the
minimum amount of data possible depending on the dimension
and the degree of the polynomial.

ALL= Only asked if one position is supplied at BOX=. [y]
Interpolate all blanks inside the area defined in SIZE= or
just the one position specified at BOX= ?

DEG= Degree of least squares polynomial used in the interpolation.
[0] Can be zero one or two.

ANOTHER= Proceed with next box ? [n]

Description:
There are two ways to use PATCH. If at BOX= an area larger
than one position is given then for each blank inside this area
patch fits a polynomial through the data around it, ignoring
other blanks, and substitutes the blank by the value of the
polynomial. The data used in the fit are collected from an
area, which sizes can be specified using the keyword SIZE=,
around the blank.
If only one position is given in BOX=, the keyword ALL= is
asked. If ALL=n PATCH proceeds the same way as described
above, but the area of operation is only one pixel. If
ALL=y, however, PATCH fits only one polynomial, centered on
the position supplied at BOX=, but substitutes all blanks
inside the area specified by SIZE= --also used to collect
the data-- by the values of the polynomial.
Updates: