Output quantities

Drop
A so-called ``Drop'' image produced by ConvPhot in the measure. The name derives from the fact that each detected object which is fitted by ConvPhot is multiplied by zero (drop) in order to leave only bright galaxies in the measure image not detected in the detection one.

ConvPhot produces in output several quantities, including parameters for each object as well as several type of images.

The most important output quantities are the best-fit solutions of the minimization procedure, i.e. the Fi parameters. Since the Pi model for each object is normalized to unit flux, the resulting total magnitude in the measure image is simply -2.5 log(Fi) + ZPm, where ZPm is the zeropoint of the measure image itself. Based on our tests, we have concluded that this total magnitude is a reliable measure of the actual total flux of the objects, somewhat less prone to systematic effects than the Kron magnitudes computed by SExtractor. However, these total magnitudes can be hardly compared with the SExtractor magnitudes of the detection image, so that reliable colors cannot be directly obtained. At this purpose, we use the total flux Di measured by ConvPhot itself in the detection image, and used to normalize the object profile Pi.

The resulting flux ratio is therefore Flux(measure)/Flux(Detection) = Fi / Di X 10-0.4(ZPm - ZPd) where ZPd is the zeropoint of the detection image. This flux ratio, or the equivalent magnitude color, mmeasure-mdetection = -2.5 log(Fi / Di) +ZPm - ZPd is the fundamental output of ConvPhot. These and other useful quantites are saved in the ConvPhot output files, as detailed in the help section.

In addition, ConvPhot produces several output images. In particular, we produce a ``Residual'' image, containing the fit residuals, i.e. the quantity {I(m,n)-Bi-M(m,n)}. This image is useful to judge the quality of the fit and of the background estimation. Since some residual typically occurrs in the center of bright objects, this image is not very useful to perform an automated search of objects that are detected only in the measure image (``dropout''). At this purpose, we also produce a ``Drop'' image, Fig.XXXX, where all the pixels within the (dilated) segmentation are set equal to the background.

Valid XHTML 1.0! Valid CSS 1!