Issue with centroiding

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Issue with centroiding

ghawkins
I'm not sure if this is an issue, or the fact I don't understand what is happening.

I'm trying to analyze the standard starfield M67. I've download the AAVSO photometric table and separated out the RA/DEC for the 178 comparison stars.

I import this RA/DEC list for an M67 image and I see apertures appear in PINK over the stars.  In most cases they are slightly off center.  With centroiding set, when I click place apertures, the majority of the apertures reposition correctly over the appropriate star, turning RED in the process.  However, for a small number of stars the centroiding does not work so well, even occasionally selecting a brighter nearby star, or the center point between two closely located stars.

What can I do to stop/control this?

Best regards,

Gary

Reply | Threaded
Open this post in threaded view
|

Re: Issue with centroiding

karenacollins
Administrator
Hi Gary,

When you have faint stars near bright stars, the centroiding operation is likely going to move the aperture to the brighter star as you have found. The only way to work around it is to turn off centroiding for the faint star (or all of the stars except the first one in the radec list as described below).

You can specify to centroid or not centroid each aperture in the .radec input file format using a 1 or 0, respectively, in the 4th column as shown below.

# Aperture file format for use in AstroImageJ (.radec format)
# RA (should have proper motion applied)
# Dec (should have proper motion applied)
# Ref Star: 0=target star, 1=ref star
# Centroid: 0=do not centroid, 1=centroid
# Magnitude (optional)
# RA,            Dec,      Ref Star, Centroid, Mag
12:14:48.899,  -63:05:12.346, 0, 1, 10.167

If all apertures are offset somewhat from stars, then the plate solution for the image may not be precise. Or, if only some apertures are not centered on the star, the problem may be that those stars have higher proper motion and the proper motion has not been accounted for in the RA and Dec coordinates.

A workaround for a plate solution that is somewhat off is to set the first aperture to centroid and the others to NOT centroid in the radec file. After loading the radec file into AIJ and starting Multi-Aperture, ENABLE "Use previous N apertures" and DISABLE "Use RA/Dec to locate apertures". Then click Place Apertures and click near the star corresponding to the first entry in the RA/Dec file. This will cause the first aperture to centroid on the star where you clicked, and the other apertures will be placed relative to their distance from the first star (T1) according to difference in RA/Dec coordinates, rather than the absolute RA and Dec.

If you have some apertures that are still not centered on a star, zoom in close on the aperture, left-click and hold the mouse button down inside the aperture and drag the aperture by hand to center it on its respective star. Release the mouse button when it is centered. Once they are all centered, press enter or right click to start photometry.

In case you want to analyze the field again in the future, be sure and save a new radec file from AIJ after you run photometry. Then you can reload the adjusted aperture locations later.

Karen
Reply | Threaded
Open this post in threaded view
|

Re: Issue with centroiding

ghawkins
Thanks Karen, I'll try these suggestions.