INP=A Input IBIS tabular file containing the
input tiepoints. If INP is specified
then the tiepoints will be taken from
the IBIS interface file rather than the
TIEPOINT parameter or the parameter
data set.
COLS=(C1,C2,C3,C4) Columns in the IBIS tabular file that
contain the tiepoints. C1 has new line,
C2 has new sample, C3 has old line, and
C4 has old sample.
OUT=B Output parameter data set containing
gridded tiepoints suitable for the
GEOM programs. If the GEOMV keyword
is used then this is in an IBIS file.
If the plot option is chosen, then
this is a plot file.
by routine XVPOUT. This data set con-
tains keywords and data for TIEPOINT
NAH and NAV and can be used instead
of specifying these keywords in the
TAE COMMAND LINE.
TIEPOINT=(NL1,NS1, these specify the input tiepoint
OL1,OS1, . . ., pairs for GEOM applications.
NLk,NSk,OLk,OSk) Maximum k is 100 (due to TAE).
TIEPOINT=(NL1,NS1, this form of parameter specifies
DZ1, . . .,NLk, the input tiepoint pairs for GEOMZ
NSk, DZk) applications or image surface
generation. Maximum k is 133 (due
to TAE).
NAH=n the integer n specifies the number
of grid cells horizontally in the
output grid (default is 30 except
in the case of LGEOM which is 10).
NAV=m the integer m specifies the number
of grid cells vertically in the
output grid (default is 30 except
in the case of LGEOM which is 10).
MINL=w the integers w, x, y, z define the
MINS=x lower and upper bounds of the
MAXL=y output grid in terms of line and
MAXS=z sample. The default is to make the
grid exactly contain the convex
hull of the input tiepoints.
MINL=w the integers w, x, y, z define the
MINS=x lower and upper bounds of the
MAXL=y output grid in terms of line and
MAXS=z sample. The default is to make the
grid exactly contain the convex
hull of the input tiepoints.
MINL=w the integers w, x, y, z define the
MINS=x lower and upper bounds of the
MAXL=y output grid in terms of line and
MAXS=z sample. The default is to make the
grid exactly contain the convex
hull of the input tiepoints.
MINL=w the integers w, x, y, z define the
MINS=x lower and upper bounds of the
MAXL=y output grid in terms of line and
MAXS=z sample. The default is to make the
grid exactly contain the convex
hull of the input tiepoints.
REJECT=r the floating point value r
specifies a radius within which
separate points will be considered
as duplicate points (default is
.01).
GEOMV this keyword specifies that the
output dataset is to be formatted
for GEOMV use. The output disk
dataset will be an IBIS file. Note
that GEOMV no longer requires the
NAH or NAV keywords.
GEOMA this keyword specifies that the
output dataset is to be formatted
for GEOMA or POLYGEOM use. The
output disk dataset will contain
the proper GEOMA or POLYGEOM
keywords and format so that no
addition parameters need to be
specified unless desired.
GEOMZ this keyword specifies that the
output is to be formatted for GEOMZ
use. The output disk dataset will
contain the proper GEOMZ keywords
and format so that no additional
parameters need to be specified
unless desired.
LGEOM this keyword specifies that the
output is to be formatted for LGEOM
use. The output disk dataset will
contain the proper LGEOM keywords
and format so that no additional
parameters need to be specified
unless desired. The user is
cautioned to observe LGEOM
application size limitations.
MGEOM this keyword specifies that the
output is to be formatted for MGEOM
use. The output disk dataset will
contain proper MGEOM keywords and
format so that no additional
keywords need to be specified
unless desired.
ABEND=ABEND specifies that the routine should
or 'ABEND abend if duplicate points are
found.
ABENDG=ABENDG specifies that the routine should
or 'ABENDG abend if the grid finding routine
fails to find a grid.
POLY=LINEAR LINEAR fits a linear surface through
the x-distortion and the y-distortion
using least squares. This requires at
least three input data points. More
tiepoints will be handled with a least
squares fit and the residuals can be
viewed. All input output formats are
valid with these options except for PLOT.
POLY=KEYSTONE KEYSTONE fits a bilinear surface through
the x-distortion and the y-distortion
using least squares. This requires at
least four input data points. More
tiepoints will be handled with a least
squares fit and the residuals can be
viewed. All input output formats are
valid with these options except for PLOT.
POLY=QUAD TIECONV can be used to prepare various
polynomial fits. For a general quadratic
fit of the form x' = ax^2+bxy+cy^2+dx+ey+f
y' = gx^2+hxy+iy^2+jx+ky+l, use the
keyword QUAD, which requires six tiepoints.
More tiepoints will be handled with a least
squares fit and the residuals can be
viewed. All input output formats are
valid with these options except for PLOT.
POLY=CUBIC TIECONV can be used to prepare various
polynomial fits. For a general cubic
fit of the form
x' = ax^3+bx^2y+cxy^2+dy^3+ex^2+fxy+gy^2
+hx+iy+j
y' = kx^3+lx^2y+mxy^2+ny^3+ox^2+pxy+qy^2
+rx+sy+t use the
keyword CUBIC, which requires ten tiepoints.
More tiepoints will be handled with a least
squares fit and the residuals can be
viewed. All input output formats are
valid with these options except for PLOT.
PLOT=PLOT bypasses tiepoint generation phase.
or 'PLOT creates graphics file of triangula-
tion for plotting.
'PRINT Keyword to print informative data.
If the input is a grid (and no poly keywords are used) the program is sped up by a huge factor. The grid is detected automatically and has a tolerance for points being out of regularity on the grid. Each point is predicted by its two previous neighbors and the distance from the predicted to the actual is ratioed to the neighbor grid distance. GRIDTOL is the maximum ratio allowed. Failure to find a grid due to this parameter causes the program to apply thiessen triangulation to the point set which can be slower by a factor of hundreds. You can be generous with this parameter, say 10 or 20, but keep it smaller than the nah or nav by half. The row or column that is found must match the length of the other rows or columns and (nah+1)*(nav+1) must equal the total number of points.