How to rotate darks?

[DaveAus]

This may be a pretty dumb question – but here goes anyway.

When using Dawn I want to use darks that have been rotated by 180 degrees so that their orientation will correspond with the lights being used. So far as I understand it Dawn only accepts fits files so the rotated darks will need to be in fits format. I can't use Fits Liberator to convert the darks for two reasons (1) it only allows for “flipping” not rotating and (2) it only seems to save tiff files. I've tried using Photoshop but although I can rotate each dark and save the result as a fits file Dawn objects when I try to load the rotated files with the message “Error loading FITS file – Image not recognised as an Atik image”.

So how do I do it? Any help will be appreciated.

[NickK]

I'm assuming that the FITS header parameters have been stripped off by FITSLiberator. Is there a way you can add back the FITS Header parameters?

[DaveAus]

Thanks for your reply NickK.  But Fits Liberator doesn't work for the two reasons stated. Perhaps you meant to say “stripped off by Photoshop”?  If so, I have no idea how to go about adding back the Fits Header parameters.

[Chris]

Dawn only processes images from Atik cameras.

Chris

[MikeM]

There is no reason to "rotate" darks. The chip does not know what orientation the camera is in.

The only time you need to concern yourself with orientation is when shooting flats where the camera has to be in the same orientation in the imaging train as your "lights".

[NickK]

There is no reason to "rotate" darks. The chip does not know what orientation the camera is in.

The only time you need to concern yourself with orientation is when shooting flats where the camera has to be in the same orientation in the imaging train as your "lights".

That's a very good point - darks are related to the CCD sensor, flats related to the optic train..

[DaveAus]

Thanks for your replies - they are appreciated.

Attached is a crop from a RGB image taken a few days ago. When using Dawn I used darks taken in an earlier imaging session. You will see small RGB arcs resulting from where hot pixels were not cancelled out by the master dark frame. I had assumed that the arcs resulted from the darks not having the same orientation as the lights and reasoned that that must be due to rotation by 180 degrees between the two imaging sessions. But I see now that that reasoning is incorrect. I'll have to investigate further to see if I can determine the real reason for the arcs.

David

[NickK]

Travelling dots of noise are down to how alignment works.

If you take ten photos of an object and each time the object is moving diagonally up then when aligned so that the object is on top of itself for each frame, the noise dot is a traced line.

Here is an example - you note the small faint galaxy and a row of horizontal hot pixel of noise moving. I've not used any flats or darks on this, from memory - the galaxy is 28 arc seconds across and I was using my own form of super resolution/drizzle.

[MikeM]

I used to get those trailing dots on my old OSC camera. It is probably due to poor polar alignment. It possibly can be minimized by dithering.