2 replies to this topic

[falco72]

Hi everyone.
I have a 1-second exposure image, with 10x7.5deg FoV and a resolution of 20 arcsecperpix.
The image contains around 200 stars, which is more than enough to perform plate solving.
20% of my camera FoV is physically obstructed(the left part of the image) and my goal is to know the sky coordinates for that obstructed part too, not only for the part that contains starfield.
I can solve the starfield part with astrometry.net, and using the .wcs and astropy's allpix2world() or wcs-xy2rd, I can obtain the sky coordinates for that obstructed part. The solving precision (average distance between all stars in the image and all matched stars from the catalog) is around 6 arcsec. How precise I can be for that transformation to the obstructed part? How to estimate it? What about distortion at the end?
Has anyone had a similar problem?

[ButterFly]

I can solve the starfield part with astrometry.net, and using the .wcs and astropy's allpix2world() or wcs-xy2rd, I can obtain the sky coordinates for that obstructed part.

So what's the issue then?

 

 

How precise I can be for that transformation to the obstructed part? How to estimate it? What about distortion at the end?
 

The precision is zero becuase there is no sky there to analyze.  You can assume the distortion characteristics of your system is the same and solve at some unobdtructed part of the sky.  Then use those distortion characteristics of the unobstructed image in your obstructed portion.  Refraction may or may not matter, depending on what you are trying to do.

 

What are you trying to do?
 

[Alex McConahay]

If you have that many stars, could you crop to the center of the original image, and solve for that? YOu are, after all, looking for the center coordinates of the image. 

 

Alex