Content-type: text/html Man page of X-GEN

X-GEN

Section: X-GEN Commands (1)
Updated: April 2005
Index Return to Main Contents
 

NAME

X-GEN - calibrate  

SYNOPSIS

calibrate [-bceghijnopvz] [-aval...] [xcen] [ycen]  

DESCRIPTION

"calibrate" creates the mapping from pixels to centimeters and vice versa. This is ordinarily done by reading in a detector image known as a "brass plate image", processing its contents, and writing out a binary file containing both a forward mapping and an inverse mapping. For detectors that are nearly linear in their geometric response (either inherently or because the images managed by X-GEN have already had their nonlinearities mapped away), "calibrate" provides a mechanism for defining the linear map through the -l and -j options.  

OPTIONS

-b
Pixels outside the active area of the detector (as determined in the "border" or "spots" functionalities) will be excluded from the list of brass-plate spots used in the mapping from pixels to centimeters.
-c
Read a list of brass-plate centroids from an ASCII file with environment variable CALCENTROIDS and use those instead of finding the centroids in a brass-plate image.
-e
Do _not_ check to see whether a brass-plate fiducial spot has been excluded from the active-pixel list before using it. Thus it is the complement of the -b flag.
-g
Force the program to use an actual brass-plate image to do its job rather than using a linear mapping.
-h
Expect the brass-plate centroids to be arranged in a hexagonal pattern rather than a rectangular pattern.
-i
Expect that the list of fiducial points found in the CALCENTROIDS file will be indexed and use those indexings in setting up the calibration.
-j
Dispense with examining a fiducial image altogether and simply define a linear mapping from pixels to centimeters using the information contained in the image header to define the spacing in cm between fictitious columns and the number of lines across the detector face. Thus it does a similar operation as compared to the -l option except that the user does not need to specify the number of columns or the raster size.
-n
Do not extrapolate its mapping outside the points that are actually observed until AFTER it has performed the first inversion of the mapping.
-o
Do not compute a pixel-to-cm mapping at all; instead, write out an ASCII list of the fiducial-point centroids to the file with environment-variable name CALCENTROIDS. This file can then be edited and re-inserted into another run of "calibrate" using the "read calibration centroids in" Boolean.
-p
Extrapolate out beyond the observed points before performing the first inversion of the mapping. This is therefore the converse of the -n option.
-v
Write out large numbers of diagnostics to the XLOG log file and to stdout during the "calibrate" operation.
-z
Do _not_ write large numbers of diagnostics to the log file. This is the converse of the -v flag.
-a<val>
This defines the multiple of sigma(local background)
that will be used as the cutoff for counts in each region. Thus if sigma(local background count) in a region is 8 and this switch is set to 5, then a fiducial spot will be considered as bright enough to use if it has several pixels with more than 40 counts in them. Default value: 4 * sqrt(mean count), unless the calibration image is from a FAST detector, for which it is 0.5 * sqrt(mean count).
-f<val>
This value defines an overall cutoff intensity below which
a potential fiducial-point spot won't be considered for inclusion, independent of the local sigma cutoffs. Thus a value of 8 for this switch will require that all fiducial spots have several pixels with at least 8 counts in them, even if the cutoff defined by sigma(local background) would allow weaker spots. Default value: 0.
-l<val>
If this value is nonzero, "calibrate" will dispense with
examining a fiducial image altogether and simply define a linear mapping from pixels to centimeters in which the value given here defines the number of COLUMNS of imaginary fiducial spots that will be created in the analysis. The number of rows will be number of columns times the ratio of the detector height to the detector width. Thus on a square detector with this parameter set to 41, we will create 41 rows and 41 columns of imaginary spots. Default value: irrelevant if "-l" isn't specified at all; if -l is specified without an associated numerical value, it is set to 19.
-m<val>
This value specifies the minimum number of pixels above
the local cutoff that must be present in order for a fiducial spot to be counted. Thus if this value is 6 and there are only 5 pixels above the local cutoff in a given spot, then the spot won't be counted. Default value: 5.
-s<val>
This specifies the fraction of the dark current image that
will be subtracted from the brass-plate image in analyzing the latter to find fiducial spots. Thus if this value is taken as 1., we will examine (fiducial image - 1.0 * dark-current image). Default value: 0.
-d<val>
This forces the program to treat the Y dimension of the
detector to be <val>. This is useful if a subset of the actual detector face is to be analyzed, so that the dimension given in the header is inapplicable. Default value: actual Y dimension of the detector.
-w<val>
This forces the program to treat the X dimension of the
detector to be <val>. This is useful if a subset of the actual detector face is to be analyzed, so that the dimension given in the header is inapplicable. Default value: actual X dimension of the detector.
-x<val>
This specifies the spacing in centimeters between COLUMNS
of fiducial points in the detector image. Thus if the columns of holes are 0.254 cm apart on the brass plate, we specify this value as 0.254. Default value: 0.508.
-y<val>
This specifies the spacing in centimeters between ROWS of
fiducial points in the detector image. Thus if the rows of holes are 0.254 cm apart on the brass plate, we specify this value as 0.254. In the case of an hexagonal grid, this should be specified as HALF the vertical spacing between points within the same column. Default value: 0.508.
xcen, ycen
specify the pixel position in (X,Y) of the direct beam when
the detector's two-theta value is zero. (X,Y)=(0,0) in the upper left-hand corner of the detector, with X increasing to the right and Y increasing down; the view is from the detector toward the crystal. Default values: half the detector width and height.
 

EXAMPLES

calibrate -j
Determine the active area of a detector using a
linear calibration (constant pixel-to-centimeter values across the face) at a beam-center position equal to the middle of the detector:
calibrate -x0.254 -y0.254 508 513
Do a full calibration based on a
brass-plate image with the beam center at [508,513], where the spacing between holes in the brass plate set at 0.254 cm:
 

REPORTING BUGS

Report bugs to Andy Howard at howard@iit.edu or 312-567-5881.  

COPYRIGHT

Copyright © 2002, Illinois Institute of Technology. See the file 'LICENSE' for information on usage and redistribution of this file, and for a DISCLAIMER OF ALL WARRANTIES


 

Index

NAME
SYNOPSIS
DESCRIPTION
OPTIONS
EXAMPLES
REPORTING BUGS
COPYRIGHT

This document was created by man2html, using the manual pages.
Time: 02:08:09 GMT, October 03, 2005