Given an astrophoto with a resolution between 0.5 and 5 arcseconds per pixel, which ways exist to identify the direction of view, field of view, and objects in the picture?

I believe most amateur astrophotographers are in this resolution range, and I believe most of us would like to confirm the captured images. I have read about the http://astrometry.net project and I'm wondering if there are any similar services available.

  • $\begingroup$ What's wrong with astrometry.net, other than they're still in alpha? $\endgroup$ – Joe Jul 26 '11 at 17:31
  • $\begingroup$ Nothing at all, in fact I hope I'll be able to try the solution on one of my servers. I just want to broaden my knowledge of the current initiatives on the subject. $\endgroup$ – Omar Salinas Jul 26 '11 at 17:39

If you don't already have an estimate of where you are pointing, the only other option I know of is WCSFixer. There also used to be the Pittsbugh WCS correction service, but it seems to be defunct now.

These tools only work with FITS files, so your first step would be converting whatever format you have into a FITS file. The FITS website has a FITS viewer page that also has a list of tools that can convert formats, although converting from FITS to something else is more commonly supported.

A note on terminology: WCS stands for "World Coordinate System", the standard used for metadata in FITS files that lets software transform between pixel coordinates and sky coordinates. If you have good WCS values, tools such as DS9 will let you find the coordinates of objects in a FITS image interactively.

If you already have an idea where you are pointing, there are more options. The simplest conceptually is to find the field were you think you are in SkyView or Google Sky and see if you can match up objects. If you can, any of the several professional data reduction toolchains have the tools that can be used to generate WCS metadata. For example, the IRAF data reduction environment has a few different options. The learning curve is steep, but not insurmountable for someone computer savvy with some time. DAOphot and FOCAS (now incorporated into IRAF?) are other options.

If you already have an approximate WCS solution, there are several other tools that will help refine the solution. The astromatic.net toolchain is one example. (I believe astrometry.net uses astromatic.net software "under the sheets," while WCSFixer uses IRAF.)

The algorithms that can determine pointing without an initial guess generally use triangles of stars in the field, because the angles in a triangle are the unchanged by rotation and scale. However, these algorithms can work poorly in highly distorted images. See Campusano et al. 1994 and Tabur 2007 for more details.

Another tool of interest is DS9's catalog overlay feature, but its more useful for confirming that your WCS is correct than it is at doing something about it if you find that it isn't.

  • $\begingroup$ Most of the things I mentioned are probably better tested at higher resolution that you requested, but I don't expect that it will matter very much: they should be able to handle the lower resolution images just fine. Of more concern is less complete starting metadata, particularly your initial guess about where you are pointing. I doubt if this is a show-stopper, but it will make using these tools on amateur data harder than data from a properly working professional instrument. $\endgroup$ – EHN Jul 27 '11 at 1:15
  • $\begingroup$ If your image is undersampled (that is, the full width at half maximum (FWHM) of a star is less than 2.3 pixels or so), you may need to blur your image for some of the object detection routines used in these tools to work properly. Most can handle undersampled images at some level, though. $\endgroup$ – EHN Jul 27 '11 at 11:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.