intrep -- Find guide star coordinates for an image.
intrep gsctab image
This task will reposition the initial guesses of guide star positions on an image. Users can select marked stars and specify better positions for them. Pixel positions will be centered based on the peak value in a sub-raster of the image at the cursor location. The plate solution will also be fitted to the coordinates. The output will consist of the equatorial coordinates of cursor locations, based on the computed fit.
This task is interactive and uses the image display server (imtool or SAOimage) and the image display cursor. See the man pages for the appropriate display server for more information on its operation. The image must be displayed using the display task in the images.tv package. When intrep starts, the image cursor (i.e., a blinking circle in imtool and a plus symbol in SAOimage) will appear in the display window. The user interacts with intrep by moving the image cursor with the mouse and entering values with the keyboard.
The plate solution option (the "f" command key) attempts to fit a set of points (X,Y,RA) and (X,Y,DEC) with a model F(X,Y). The fit routine is a version of the Single Value Decomposition (SVD) method used in the routine svdfit (and is described the book "Numerical Recipes", by Press, et al.); this version is modified for double precision. Positions (X,Y) are offsets from the lower left corner, which is defined as (1,1).
Pixel values are first converted to millimeters, with respect to the given reference point (CRPIX1,CRPIX2) and the pixel size (in microns). The equatorial coordinates values (RA,DEC) are converted to standard coordinates (XI,ETA) in arcseconds, given the equatorial coordinates for the reference point. The coordinates used in intrep can be written to an output file in a format that can be used by more complete fitting tasks, such as pltsol. The fit coefficients computed by intrep can also be written in a format suitable for the xyeq task, which computes equatorial coordinates for a list of pixel (X,Y) coordinates based on the fit.
After obtaining the plate solution, coordinates can be computed for any point by evaluating the fit. The "c" command finds the coordinates of the star from the input list nearest the cursor. The "p" command key finds the coordinates at the peak value in a small subraster of the image near the cursor. Colon commands (i.e., those preceded by the ":" character) can accept arguments.
The following command keys are recognized by this task:
c Show the star closest to the cursor. d Delete a star (i.e., flag as not valid). f Plate solution. l List the stars. p Find the peak of the subimage bounded by the search box. q Quit. s Select a star to be repositioned. u Undelete a star (i.e., flag as valid). :b size Set or show the search box size (in pixels). :d row Delete the listed star by number. :D Delete all stars. :f file Write the fit coefficients to a file. :n num Set or show the number of fit coefficients. :s file Write the valid coordinates to a file. :u row Restore the listed star by number. :U Undelete all stars.
- gsctab [file name]
- Guide Star Catalog table name. This table is assumed to consist of columns containing right ascension, declination, X and Y pixel coordinates, and a validity flag. One row is used for each star selected from the guide star catalog.
- image [file name]
- Image name; this should be the original unmarked image---not the output from overlay. The "s" and "p" commands (explained above) use a subraster of pixels from the image to compute the peak value.
- (boxsize = 7) [integer, min = 3]
- Box size in which to search for the center of the star.
- (valid = 0) [integer, min = -1, max = 1]
- Initialize validity of all stars? The following values are
-1 Set all flags to not valid. 0 Don't change flags. 1 Set all flags to valid.
- (xcrpix) [real]
- X reference pixel.
- (ycrpix) [real]
- Y reference pixel.
- (racen) [real]
- Right ascension at the reference pixel (in degrees).
- (deccen) [real]
- Declination at the reference pixel (in degrees).
- (xpxsize) [real]
- X pixel size (in microns).
- (ypxsize) [real]
- Y pixel size (in microns).
- (pltscl) [real]
- Plate scale (in arcseconds per millimeter).
- (solout = no) [boolean]
- Write equatorial and pixel coordinates in a text table suitable as input to the pltsol task?
- (coord) [file name]
- Name of the image display cursor file.
Type "help gasp opt=sys" for more information about the gasp package and the Guide Star Catalog.