[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: correcting slope
- To: email@example.com
- Subject: Re: correcting slope
- From: Michael Gutzwiller <firstname.lastname@example.org>
- Date: Mon, 25 Aug 1997 11:55:59 -0400
- Old-Return-Path: <email@example.com>
- Organization: D&M Electronics
- References: <9708251521.AA19895@runner.nofs.navy.mil>
- Resent-Date: Mon, 25 Aug 1997 12:00:17 -0400 (EDT)
- Resent-From: firstname.lastname@example.org
- Resent-Message-ID: <"E9cFbB.A.8FC.PuaA0"@kani.wwa.com>
- Resent-Sender: email@example.com
> Mike G. was kind enough to post the results from a typical image. I have
> a couple of questions.
> The +- column is from Star's estimate of the magnitude error?
> The catalog mag field is the transformed VT into V value?
Yes, using the last translation set you posted.
> (A moderately ignorant question) the Mag column from the Star program
> is based on a constant sized psf for the entire frame? I know you've given
> this algorithm several times; my memory is just poor.
Yes. An average PSF is calculated for the entire image and used to
derive an optimized aperture. While the instrumental magnitudes used
are those from this optimized aperture, the magnitudes from the circular
aperture of approximately the same size tracks pretty well with the PSF
> (Now, a truely ignorant question) the Y direction is in RA, and the
> X direction is in Declination? I'm a little suprised that the fwhmx
> varies by a factor of two over the declination range. That is a big
> change in the image shape.
Yes. I agree the range is diconcerting but it also does NOT correlate
well with declination.
> You are certainly correct that, for proper zero point setting, you need
> to remove outliers from the fit. There will always be objects that are
> blended in TASS images that match with a single Tycho star, or variables,
> or objects with unusual colors, or stars contaminated with cosmic rays,
> etc. You just want the number of rejects to be 'small' in comparison with
> the number of accepted objects. This rejection scheme gets tricky when
> dealing with changing sky conditions.
Right now I am experimenting with a method described in the Numerical
Recipes book. Essentially it is a median solution to a linear fit. The
routine, for those who have the book, is called medfit. The results
seem encouraging so far.