Next: Technical details of Up: The Source Extraction Previous: Introduction

The SAD routine

SAD is used as a framework in a batch program. It attempts to find all sources in an image whose peak is brighter than a given level. It searches for all points above this level and merges such points in contiguous ``islands''. For each island, initial estimates of the strength and size are generated. Then the gaussian fitting algorithm used in JMFIT is called to determine the least square gaussian fit.

The error estimates should be regarded as tentative. They are formal standard error estimates, obtained by multiplying the square roots of the diagonal elements of an approximation to the inverse Hessian of the chi-squared function by the final r.m.s. residual. These estimates assume that the errors in the input data are uncorrelated (which they are not!). As a result the error estimates should be regarded as lower limits which should be multiplied by a suitable fudge factor.

When beamsize information is available in the header of the map, SAD deconvolves the clean beam from the fitted component size. In our case we calculate what the beamsize is, because we don't use the CLEAN algorithm during reduction (except for very extended structures). The beam size is put in the header at the start of the source finding batch.



M.Bremer@sron.ruu.nl
Thu Mar 21 12:29:49 MET 1996