NAME
disparity2d - Computes disparity between neighboring label
sites
SYNOPSIS
disparity2d [ -t1 input_file_1 ] [ -t2 input_file_2 ] [ -t3
input_file_3 ] [ -t4 input_file_4 ] [ -t5 input_file_5 ] [
-O file_out ] [ -coarse coarse ] [ -halfwin halfwin ] [
-fmat fmat_file ] [ -bmat bmat_file ] [ -nr nr ] [ [-ks|-ad]
] [ -? ]
DESCRIPTION
disparity2d This program computes the disparity between all
pairs of neighboring label sites. The two label sites in a
neighboring pair can be near neighbors or random neighbors.
The disparity for one of these pairs is computed by compar-
ing the attribute values in the texture analysis window of
one of the label sites with the attribute values in the tex-
ture analysis window of the other label site. This program
uses the Kolmogorov-Smirnov statistic (or some variant, such
as the Anderson-Darling statistic) to determine the similar-
ity of the cdf's of the attribute values in the two texture
analysis windows. If the textures (i.e. the cdf's) of the
two sites are different, then the disparity between these
two sites is 1. If the textures of the two sites are simi-
lar, then the disparity between the two sites is -1.
disparity2d gets all its parameters from command line argu-
ments. These arguments specify the input, output, etc.
COMMAND LINE ARGUMENTS
-t1 input_file_1
the input file containing the first attribute of all
the sites in the data lattice. The first attribute is
the data value at the data site. (optional)
-t2 input_file_2
the input file containing the second attribute of all
the sites in the data lattice. The second attribute is
the intensity range, i.e. the maxval minval in the
attribute analysis window of the data site. (optional)
-t3 input_file_3
the input file containing the third attribute of all
the sites in the data lattice. The third attribute is
the boundary residual, i.e. the data value at the data
site minus the average of the data values on the boun-
dary of the the data site's attribute analysis window.
(optional)
-t4 input_file_4
the input file containing the fourth attribute of all
the sites in the data lattice. The fourth attribute is
the horizontal directional residual, i.e. a gradient
approximation in the horizontal direction. (optional)
-t5 input_file_5
the input file containing the fifth attribute of all
the sites in the data lattice. The fifth attribute is
the vertical directional residual, i.e. a gradient
approximation in the vertical direction. (optional)
NOTE: at least one of the attributes must be used to
calculate the disparity values, i.e. at least one of
the options {t1, t2, ..., t5} must be specified on the
command line.
-O file_out
the output file which contains the disparity values.
For each label site there are num_of_rand_neigh + 4
disparity values, one for each neighbor of the label
site. (Default: stdout)
GENERAL OPTIONS
-fmat fifn
the input file containing the forward index used for
locating random neighbors. The forward index is gen-
erated by the program randomgraph
-bmat bifn
the input file containing the backward index used for
locating random neighbors. The backward index is gen-
erated by the program randomgraph
[-ks | -ad]
a pair of optional flags used to specify which statist-
ical test is used to compare textures. The
Kolmogorov-Smirnov test is used when the -ks flag is
specified. The Anderson-Darling test is used when the
-ad flag is specified. Specification of a test is
optional. If neither one of these flags is set the
Kolmogorov-Smirnov test is used. NOTE: at most one of
these flags can be set on a command line. NOTE: The
Anderson-Darling test is still being refined. So, it
is not implemented in the current version.
-coarse coarseness
inverse of the resolution at which the segmentation is
performed. That is, the ratio of the number of
samples/traces in the data lattice to the number of
samples/traces in the label lattice. The data lattice
is the input to the program texture, and the label lat-
tice is the output from any segmentation optimizer,
such as bethe. For highest resolution set the coarse-
ness to 1. (Default: 9)
-nr num_of_rand_neigh
the number of random neighbors for each label site.
The neighborhood for a label site also includes 4 near
neighbors. Thus, each label site will have
num_of_rand_neigh + 4 neighbors. The number of random
neighbors must be a power of two. (Default: 16)
-halfwin half_win_size
half window width used to determine the width of the
texture analysis window over which the cdf's are com-
puted in the program disparity2d. The width of the
texture analysis window is given by 2 * half_win_size +
1. (Default: 7)
-h, -?, -help
Help.
See Also
texture2d, bethe2d, cogeom2d, combine2d, randomgraph2d,
texstat2d, brightsizing
REFERENCES
Crawford, Kelly, and Marfurt, Kurt, 1997. 2D Texture
Analysis: A User's Guide, Amoco Geoscience Research Bul-
letin F97-G-14.
Hoelting, Cory and Kelly, Ken, 1996. Texture Analysis of
Spectral Decomposition Data Using a Segmentation Algorithm,
Amoco Geoscience Research Bulletin F96-G-21.
Matheney, Mike and Kelly, Ken, 1995. Texture-Based Segmen-
tation of 3-D Seismic Data, Amoco Geoscience Research Bul-
letin F95-G-29.
CONTRACT AGREEMENT
This product is brought to you by Research Agreement #548
(The Seismic Coherency Cube). Thank you for your support!
AUTHOR
Cory Hoelting and Ken Kelly (E&PTG, Tulsa, OK, USA).
COPYRIGHT
copyright 2001, Amoco Production Company
All Rights Reserved
an affiliate of BP America Inc.
Man(1) output converted with
man2html