PKFIT

       Subroutine of  GETPSF to perform a one-star least-squares fit

       Part of the DAOPHOT PSF photometry sequence

       PKFIT, f, scale, x, y, sky, radius, ronois, phpadu, gauss, psf,
                               errmag, chi, sharp, Niter, /DEBUG

       F      - NX by NY array containing actual picture data.
       X, Y   - the initial estimates of the centroid of the star relative
               to the corner (0,0) of the subarray.  Upon return, the
               final computed values of X and Y will be passed back to the
               calling routine.
       SKY  -   the local sky brightness value, as obtained from APER
       RADIUS-  the fitting radius-- only pixels within RADIUS of the
               instantaneous estimate of the star's centroid will be
               included in the fit, scalar
       RONOIS - readout noise per pixel, scalar
       PHPADU - photons per analog digital unit, scalar
       GAUSS -  vector containing the values of the five parameters defining
               the analytic Gaussian which approximates the core of the PSF.
       PSF   -  an NPSF by NPSF look-up table containing corrections from
               the Gaussian approximation of the PSF to the true PSF.

       SCALE  - the initial estimate of the brightness of the star,
               expressed as a fraction of the brightness of the PSF.
               Upon return, the final computed value of SCALE will be
               passed back to the calling routine.

       ERRMAG - the estimated standard error of the value of SCALE
               returned by this routine.
       CHI    - the estimated goodness-of-fit statistic:  the ratio
               of the observed pixel-to-pixel mean absolute deviation from
               the profile fit, to the value expected on the basis of the
               noise as determined from Poisson statistics and the
               readout noise.
       SHARP  - a goodness-of-fit statistic describing how much broader
               the actual profile of the object appears than the
               profile of the PSF.
       NITER -  the number of iterations the solution required to achieve
               convergence.  If NITER = 25, the solution did not converge.
               If for some reason a singular matrix occurs during the least-
               squares solution, this will be flagged by setting NITER = -1.

       No parameter checking is performed

       Adapted from the official DAO version of 1985 January 25
       Version 2.0 W. Landsman STX             November 1988
       Converted to IDL V5.0   W. Landsman   September 1997

[Home Page] [Software, Documentation] [IDL Documentation] [Quick Reference] [Feedback]