| cenpars | xapphot | cenpars |
cenpars -- Edit the centering algorithm parameters
cenpars
The cenpars parameters control the action of the centering algorithms. The default values should produce reasonable results in most cases. Some of the centering parameters scale with the image data dependent parameters impars.iscale and impars.iskysigma .
For each object to be measured xguiphot extracts a data subraster 2.0 * cradius / impars.iscale pixels wide around the initial position. If impars.iscale is defined in terms of the half-width half-maximum of the psf, the same value of cradius will work well for image data taken under a wide range of conditions.
New centers are computed using the centering algorithm specified by calgorithm , the extracted image data, and points that are some threshold above or below the local minimum or maximum data value if impars.iemission is "yes" or "no" respectively. Values for cthreshold of INDEF, a positive number, and 0.0, invoke the default thresholding algorithm, set the threshold to impars.iskysigma * cthreshold, or set the threshold to 0.0 respectively.
After thresholding the signal to noise ratio (SNR) of the subraster is estimated using cminsnratio , the background noise estimate defined by impars.iskysigma, and the gain value defined by the impars.ikgain or igain parameters. If the SNR < cminsnratio, the new center is still computed but an error flag is associated with that object.
The default algorithm is "centroid1d". This centering algorithm computes the x and y marginal distributions from the extracted subraster. The intensity weighted mean and mean error of the x and y marginal distributions are computed using only points in the marginal array above or below the threshold value which is either the mean, impars.iskysigma * cthreshold +/- the background, or the local background, depending on whether cthreshold is set to INDEF, a positive number, or 0.0 respectively. The centroiding algorithm iterates until the new x and y coordinates are both less than 1 pixel from the previous guess or until the number of iterations equals cmaxiter . Centroid1d is the only centering algorithm for which the choice of impars.ihwhmpsf is not critical.
The centering algorithm "gauss1d" computes the new centers by fitting a 1D Gaussian function to the marginal distributions in x and y using a fixed half-width half-maximum of impars.ihwhmpsf. Initial guesses for the fit parameters are made from the data values. The default thresholding algorithm for the "gauss1d" algorithm is no thresholding. The "gauss" fitting1d routines iterate until a best fit solution is achieved or the limit of cmaxiter is reached.
The final choice "ofilter1d" employs a variation of the optimal filtering technique in which the profile is simulated by a triangle function of width impars.ihwhmpsf . The default thresholding algorithm for the "ofilter1d" algorithm is no thresholding. The "gauss" fitting routines iterate until a best fit solution is achieved or the limit of cmaxiter is reached.
If the computed shift in either coordinate > cxyhift / impars.iscale, the new center is returned but an error flag is set.
If the ctrpmark is "yes, then a mark of shape charmark , color ccolormark , and csizemark is drawn at the position of the centered object.
1. List the centering algorithm parameters.
xa> lpar cenpars
2. Edit the centering algorithm parameters
xa> cenpars
3. Edit the centering algorithm parameters and run xguiphot.
xa> epar xguiphot
... edit a few xguiphot parameters
... move to the cenpars parameter and type :e
... edit the cenpars parameters and type :wq
... finish editing other pset and task parameters
... type :go to run xguiphot
4. Save the current centering algorithm parameters set in the text file cennite1.par.
xa> cenpars
... edit the parameters
... type ":w cennite1.par" from within epar
epar,lpar,xguiphot,impars,findpars,skypars,photpars