Image Integration

[Nigel Ball]

Hi all

Following Jordi's excellent presentation I've been inspired to try ImageIntegration in PI.

Now i have some questions regarding the various outputs

I have 40 Luminance subs on NGC1333 - very faint stuff

Using LinearFitCurve I've been adjusting the Linear fit low and Linear fit high under Pixel Rejection(2) to achieve a higher average SNR as seen in the Console Window. Now the theoretical maximum for 40 images is 6.325 I've only managed to achieve 5.1663 by settign both LFHigh and LFLow to the maximum values of 10.

Now the questions
- There is a signifcant gradient in the image - will this affect the SNR calculation?
- what am i looking for in the slope window
- how do I interpret the various parameters in the console window namely
        Gaussian noise estimate
        Reference SNR
        
Finally my initial results are very encouraging as shown by the screenshot below. The PI integrated images are much smoother than the CCDStack combine (bottom right)

Thanks in advance

[Nigel Ball]

Another question

Is this banding normal in the slope image?

I get different orientations with each filter ....

[Harry page]

Hi

Don't know if I can be wise on this  but here goes

1) I do not know if the gradient will affect the svr calculation , sorry
2) I remember juan saying he will find a use for the slope , but just pretty at the moment 
3) Higher Referance SNR numbers are better

But generally I do not use this I refer to the clipped pixels knowing on average I only want perhapes 2 or 300 pixels rejected  (in  6 meg camera ) I spend my time looking at this , therefore leaving all the good stuff in of course if there is sat / plane trails you will get more pixels rejected,

I have seen banding before , this has been down to relections in my scope do not know if its the same for you

Regards Harry

[Nigel Ball]

Harry,

Thanks for taking the time to look at this and comment.

Yes I appreciate higher SNR numbers are better but how close to the theoretical value should I be getting? Do values far from the theoretical maximum indicate poor subs/ underexposed subs/ bad seeing etc?

As you can see I have lots of questions ....

Maybe the fact that nobody has responded to my post indicates extra info / guru input is required on this

Perhaps one of the PI Team could chip in at some point on this?

Nigel

[Harry page]

still Much to learn I have 


Harry

Harry,

Thanks for taking the time to look at this and comment.

Yes I appreciate higher SNR numbers are better but how close to the theoretical value should I be getting? Do values far from the theoretical maximum indicate poor subs/ underexposed subs/ bad seeing etc?

As you can see I have lots of questions ....

Maybe the fact that nobody has responded to my post indicates extra info / guru input is required on this

Perhaps one of the PI Team could chip in at some point on this?

Nigel