How It Works

The Stars Show the Blur

A star is a point of light, so the shape it arrives in is the blur itself, with nothing else mixed in. Atmospheric seeing, defocus, spherical aberration, coma, astigmatism, trefoil, chromatic aberration, and elongation from guiding errors, wind or the mount each change that shape in their own way.

That makes every star a measurement. Together, the stars in your frame show Wavefront Restore which aberrations are present, how strong they are, and how they change from the center of the frame to the corners. You don’t supply a PSF or pick a reference star.

One Correction for the Whole Image

Your optics blurred everything in the frame, not only the stars: a galaxy’s dust lanes, a nebula’s filaments, the edge of a globular cluster’s core. At any point in the frame, a star and the nebula behind it were blurred in exactly the same way, so the physically correct way to undo that blur is the same for both.

Wavefront Restore therefore applies what it reads from the stars to the whole image. The correction changes across the field, because the optics do. It never changes between stars and everything else, and there is no separate setting for each.

Making stars smaller for the look of the picture is a different job from deconvolution. If you want it, do it as a separate step afterward.

Linear Data Only

Use Wavefront Restore on linear data, before any stretch. It is built and tested only for linear data.

On linear data, the image is exactly the scene blurred by your optics, mount and sky, which is what Wavefront Restore reverses. A stretch changes bright and faint parts of the image by different amounts, so after one the blur is no longer the same everywhere.

Exceptions

A few areas are deliberately treated differently:

• Halos around very bright stars are left as they are.