#255 closed defect (fixed)
psCell - removal of fromChip & toTP transforms
| Reported by: | Owned by: | eugene | |
|---|---|---|---|
| Priority: | high | Milestone: | |
| Component: | PSLib SDRS | Version: | unspecified |
| Severity: | normal | Keywords: | |
| Cc: |
Description
In the latest SDRS, the fromChip transform was removed from the specification,
but it is used in the psCellInChip function. Also, the toTP transform was
removed, but it is used in psCoordCellToSkyQuick as explicitly required by the
SDRS (see page 74 of SDRS r.10).
I assume the fromChip transform is basically going to be the inverse of the
toChip transform, but there currently is no means to do an inverse of a
transform, and furthermore, could be computationally expensive if the transform
is used frequently and the roots of the polynomial are needed to be calculated
every time the transform is performed.
-rdd
Change History (3)
comment:1 by , 22 years ago
| Owner: | changed from to |
|---|
comment:2 by , 22 years ago
| Status: | new → assigned |
|---|

the entry toTP became toSky, but this is probably wrong since it is
dimensionally incorrect (toSky is a PsPlaneTransform to go from psPlane to
psPlane, not to psSphere). To get to the sky, you really need to have at least
some of the info in the psExposure: ra, dec, or at least a psProjection. This
highlights some of the inconsistencies which we need to clean up a bit. This
will become a bit clearer as we specify the module bits which convert header
astrometry information (aka WCS info) into the corresponding psProjection,
psDistortion terms and match this with the psFPA I/O functions.
The fromChip part is a bit more subtle. The point of dropping this term is
that, in generat, the fromChip distortion term has very large linear terms and
very small non-linear terms. This means the inverse of the polynomial can be
approximated negating the coefficients of the higher order terms. This should
be clarified in the ADD.