Index: /trunk/doc/modules/ModulesSDRS.tex =================================================================== --- /trunk/doc/modules/ModulesSDRS.tex (revision 1109) +++ /trunk/doc/modules/ModulesSDRS.tex (revision 1110) @@ -1,3 +1,3 @@ -%%% $Id: ModulesSDRS.tex,v 1.3 2004-06-26 00:34:15 price Exp $ +%%% $Id: ModulesSDRS.tex,v 1.4 2004-06-26 03:42:48 price Exp $ \documentclass[panstarrs]{panstarrs} @@ -394,14 +394,12 @@ Given an input image and region to keep, \code{psPhase2Trim} shall -trim the input image, leaving only the specified region. The API -shall be the following: -\begin{verbatim} -/** Trims the input image to remove the edges corrupted by OT shifting, and the overscan. */ -psReadout *psPhase2Trim(psReadout *in, ///< Input image to be trimmed, and output - const psImageRegion *region ///< Region to keep - ); -\end{verbatim} - -Note that the trim shall be performed on the input image in-place. +trim the input image, \code{in}, leaving only the specified region, +\code{region}. The API shall be the following: +\begin{verbatim} +psReadout *psPhase2Trim(psReadout *in, const psImageRegion *region); +\end{verbatim} + +Note that the trim shall be performed on the input image in-place, so +that the input image is modified. The offset of the input image (\code{in->x0} and \code{in->y0}) shall @@ -420,13 +418,16 @@ \subsection{Non-linearity} -Given an input image, and the coefficients of a polynomial, -\code{psPhase2CorrectNonLinearity} shall correct the input image for -non-linearity by applying the polynomial to the flux of each pixel. +We shall specify two functions to perform the non-linearity +correction, in order to implement the possible methods of specifying +the correction. + +The first, \code{psPhase2NonLinearityPolynomial} shall correct the +input image for non-linearity by applying the given polynomial, +\code{coeff}, to the flux of each pixel in the input image, \code{in}. The API shall be the following: + \begin{verbatim} /** Applies the correction for detector non-linearity. */ -psReadout *psPhase2CorrectNonLinearity(psReadout *in, ///< Input image to be corrected, and output - const psPolynomial1D *coeff ///< Polynomial with which to correct - ); +psReadout *psPhase2CorrectNonLinearity(psReadout *in, const psPolynomial1D *coeff); \end{verbatim} @@ -434,13 +435,31 @@ caller, likely from the image metadata. +The second function, \code{psPhase2NonLinearityLookup} shall correct +the input image for non-linearity by using a lookup table. The API +shall be the following: + +\begin{verbatim} +psReadout *psPhase2NonLinearityLookup(psReadout *in, const psVector *inFlux, const psVector *outFlux); +\end{verbatim} + +For each pixel in the input image, the function shall find the flux +value in the \code{inFlux} vector, and replace the flux with the +corresponding value in the \code{outFlux} vector. If the sizes of the +\code{inFlux} and \code{outFlux} vectors differ, then the function +shall generate a warning, and the longer vector shall be truncated to +the length of the shorter. The \code{inFlux} vector may be assumed +to be pre-sorted. + +Both \code{psPhase2NonLinearityPolynomial} and +\code{psPhase2NonLinearityLookup} shall modify the input image +in-place. + \subsection{Flat-fielding} Given an input image and a flat-field image, \code{psPhase2FlatField} -shall divide the input image by the flat-field image. The API shall be the following: -\begin{verbatim} -/** Flat-fields the image. */ -psReadout *psPhase2FlatField(psReadout *in, ///< Input image to be flat-fielded, and output - const psImage *flat ///< Flat-field image - ); +shall divide the input image by the flat-field image. The API shall +be the following: +\begin{verbatim} +psReadout *psPhase2FlatField(psReadout *in, const psImage *flat); \end{verbatim} @@ -456,8 +475,10 @@ Pixels which are negative or zero in the \code{flat} shall be masked -in the input image. +in the input image with the value \code{PS_MASK_FLAT} (see +\S\ref{sec:maskValues}). The function shall not normalise the \code{flat}; this responsibility -is left to the caller. +is left to the caller. This function is basically equivalent to a +divide, but with care for the region that is divided. \subsection{Subtract sky} @@ -484,6 +505,8 @@ \end{verbatim} -Note that the input image shall be subtracted in-place. The function -shall return the subtracted image, and also update the polynomial, +The polynomial specification, \code{poly}, specifies the order of the +polynomial to be used in fitting the background. Note that the input +image, \code{in}, shall be subtracted in-place. The function shall +return the subtracted image, and also update the polynomial, \code{poly}, to hold the polynomial coefficients used in the subtraction, and update the vector \code{outNorms} to hold the @@ -491,5 +514,6 @@ subtraction. The \code{outNorms} vector shall be resized if necessary to match the number of template sky images in the list -\code{skyImages}. +\code{skyImages}. If the polynomial specification is \code{NULL}, then +no polynomial shall be fit to the input image. Note that the input image, \code{in}, and the template sky images, @@ -503,10 +527,14 @@ function shall generate an error. -If the size of a source in the \code{catalog} is \code{size}, then the -radius to mask around the source before fitting the background shall -be \code{radiusMultiply} $\times$ \code{size} + \code{radiusAdd}. +The \code{catalog} shall be a linked list of \code{psObject} which +corresponds to objects on the input image. If the size of a source in +the \code{catalog} is \code{size}, then the radius to mask around the +source before fitting the background shall be \code{radiusMultiply} +$\times$ \code{size} + \code{radiusAdd}. \tbd{Are the catalog stars in celestial or pixel coordinates? I lean to pixel coordinates.} + +\tbd{What size do we use? FWHM?} When fitting a polynomial, the function shall bin the input image by @@ -529,4 +557,24 @@ The following modules mask various defects on an image. +\subsection{Mask values} +\label{sec:maskValues} + +We define several mask values for use in the phase 2 processing: +\begin{verbatim} +/** Mask values */ +typedef enum { + PS_MASK_TRAP, // The pixel is a charge trap + PS_MASK_BADCOL, // The pixel is a bad column + PS_MASK_SAT, // The pixel is saturated + PS_MASK_FLAT, // The pixel is non-positive in the flat-field + PS_MASK_CR_MORPH // The pixel is determined to be a cosmic ray, based on its morphology +} psMaskValue; +\end{verbatim} + +Of these, masks for the charge traps need to be grown by the +extent of the OT convolution kernel. For other pixel types, +orthogonal transfer of the flux in this pixel will not (necessarily) +affect the flux in neighbouring pixels + \subsection{Bad pixels}