Alignment issues in AA5

[Ivor]

HI,

I'm working on transferring more of my workflow over to AA5 and I've come up with a problem when aligning a set of RGB images for M33, GB align fine however there is a small rotation in the green channel and I'm not sure how to correct it. Any suggestions how I can get around this?

The files can be found here
Red https://www.dropbox.com/s/r59myhb1euv2bwu/m33_R_900_merged.fit
Green https://www.dropbox.com/s/r59myhb1euv2bwu/m33_G_900_merged.fit
Blue https://www.dropbox.com/s/r59myhb1euv2bwu/m33_B_900_merged.fit

My only alternative at the monent is PS but I'm not sure how to work out the axis point for green channel to rotate it.

Thanks

[MarkS]

Somewhere you'll find there's an option to automatically align images.  That ought to cope with translation, rotation and possibly other distortions.

[Ivor]

I found that but it didn't work. It's as though it didn't find the correct pivet point in the green frame so some stars are good but as you move away from the pivet point errors in the aligment appear.

[MarkS]

Sounds like it's not applying rotation correctly or maybe there's a preference setting or checkbox somewhere that governs whether or not rotations are calculated and applied.  It's only a guess because I don't know AA.

[Ivor]

My thought as well so I've put the question up on the AstroArt forum as well. It did bring up an interesting question (well I thought it was interesting) how can you manual workout the pivet point? I hada similar problem with my IC1805 image (the reducer slipped again causing a slight rotation) I'd hoped by framing the galaxy right in the centre of the image I'd reduce the challange if it happen again but that doesn't appear to be the case.

In PS I can put the frames in layers and rotate and get fairly close but not any better than the auto alignment in AA5. I was wondering how the software did it. I thought there might be a way to trianglulate the pivot point my hypnosis:

Choose two reference stars on two frames align them
draw a line between the two stars and at the mid point draw a further line perpendicular to it.
Choose a third star and repeat using star 1
Repeat again with star 2 and 3.
The 3 perpendicular lines should all meet in the centre of the triangle but is it the pivot point?

[MarkS]

Are you saying that AA does actually do a rotation (same thing in PS) but you are still left with a residual problem?

[Rocket Pooch]

Hi,

AA does rotation and should align, there's an option for star pattern matching somewhere in there.

Chris

[Ivor]

When you align the subs it gives you the option for rotation and alignment but I haven't seen that in the channel alignment functionality, as mentioned before I'm not sure if this is my inexperience of AA5 or a gap in the functionality.

[MarkS]

I don't understand your concept of "pivot point".

You can undo the rotation by rotating about the top left corner, the bottom right corner, the centre, or any other point.  Following this, a simple translation is all that is required to line everything up.  The utility of a pivot point is therefore transient and arbitrary.

[Fay]

I just use Auto Alignment, Star Pattern Translation only, Ivor

[Ivor]

I don't understand your concept of "pivot point".

What I meant by a pivot point was once the the subs had been combined the rotation point was off centre, Im guessing roughly where my x is, by default PS rotates around the centre of the image so I would be able to get all the channels aligned.

I just use Auto Alignment, Star Pattern Translation only, Ivor

Thank you, you made me realise a way around the problem. The problem appears to stem from the factor the auto alignment in the  trichromy panel doesn't appear to use the Star pattern translation + rotation option, if I align the channels before going into that function it works.