Re-use of coordinate system information with AST files

Coordinate systems can be exported from and imported to image files using the commands:

• ASTEXP
• ASTIMP

(the MAKESET routine can be used as a substitute for ASTIMP if CCDPACK Sets are being used -- see Section ).

Sometimes the coordinate transformations which must be applied to one set of observations are the same as those required for many other sets. One example of this is the observations from a mosaic CCD camera, in which multiple CCD chips are fixed adjacently in the same focal plane to avoid needing a single very large device; the relative position of each chip to the others will not change as a function of pointing direction. Another example is when the optical distortion of an instrument (at a given wavelength) is known; if this can be applied correctly to one observation it can be applied to many. A third example, which is common because the large focal planes implied by use of a mosaic camera often lead to significant optical distortions, is the combination of these two.

CCDPACK allows this kind of coordinate transformation information to be written to (using ASTEXP), and read back from (using ASTIMP), an external file called an AST file. This effectively stores a coordinate system, and enough information to graft it onto suitable files, for each one of a related group of images. Thus applying the AST file to a set of images adds a new coordinate system to each of them, which can be used for the registration process. Where there are several images in a set, as in the case of a mosaic camera, ASTIMP can determine which coordinate frame to use for which image either by the order in which they are presented, or by usng FITS headers in the file, according to how the AST file was constructed by ASTEXP.

Full details on how to use ASTIMP and ASTEXP can be found in an appendix, but the following gives an example.

% astexp '"reg_data[1234]"' astfile=inst.ast idtype=fitsid fitsid=chipname
% astimp 'new_data*' astfile=inst.ast

The first command takes a set of 4 images, one from each chip of an array, which are aligned in their Current coordinate system, (possibly by object matching) and constructs a file inst.ast from them; the parameters `IDTYPE=FITSID FITSID=CHIPNAME' mean that the AST file labels each coordinate system with the value of the `CHIPNAME' FITS header from the reg_data* images, since this dentifies which CCD is which in the mosaic camera. The second command applies the file inst.ast to a set of (any number of) images from the same instrument which have not yet been registered; for each one ASTIMP works out which coordinate system to add by matching the `CHIPNAME' FITS header in the image with one of the ones in the AST file.

If a previously prepared AST file for the instrument you are using exists, then the ASTEXP step can be avoided. In any case, once a suitable AST file is available, it can be applied to many sets of images from the same instrument, as long as the instrument's characteristics remain the same.

There is a more detailed example of the use of ASTIMP in section §.