Combining Two Different Nights

[Aidi]

I've got a set of images and associated calibration frames from a few weeks ago, I imaged the same target last night and did the flats this morning. Run them through Pixinsight and now have two images stacked etc.

I'm a bit confused as to how best to stack the two light frames together to combine the different sessions into one 4 hour capture.

I also tried in batch pre processing using the custom option to add the lights & flats using a different filter name for each session - so the process would associate the correct lights & flats together... but ended up with a master light for each session! My understanding was the process would stack the all the lights from both sessions into one master.

I did try image integration, but the process will not run with less than three images


Any assistance would be appreciated

Rgds
Aidan

[Niall Saunders]

Hi Aidan,

If you consider your first set of images - you must have performed a StarAlignment during that process, and you must have used a referece image against which you aligned all other images.

If so, then what you now need to do is to StarAlign all of your new images against the same (original) reference image.

At least all of your data is now correctly aligned.

You could now proceed with ImageIntegration of your new images - if you did this you would then have two 'MasterImages', and you could simply run both of these through a 'standard' (all parameters reset) ImageIntegration Process to generate a final MasterImage. The alternative - though pretty much identical - method (here) would be to add both of the images together using PixelMath.

Of course, if you still had all of your original calibrated Lights, then you could run that set, and your new set - together - through a single ImageIntegration Process. You might find that this gives you the best noise rejection (and hence the greatest increase in SNR).

You should consider experimenting (or save the task for a rainy day!).

[Aidi]

Thanks for the response.

I'm still quite new to all this, but I think I have the basics..so, I am happy that the images are all aligned.

 I did try your suggestion about image integration, but it errors as I only have two frames.  I suppose I could create two separate masters from one set of images, but that seems quite long winded.

I'll give pixel math ago.  I've got an idea of how that works!

I still have all the original data from both sessions.  Think is, I use both a mono and osc, and swap between sessions... So I need to make new flats, which is the case here.  I have different flats for both sessions!

[Niall Saunders]

I did try your suggestion about image integration, but it errors as I only have two frames

I am surprised that this is what happens - but, in all honesty, I have never tried II with just two frames. Still, I know that it will work in PixelMath, so if you are willing to give that a go, then you keep on learning!

Post your PixelMath expression back here, along with how you set up the rest of the PM Process window. (You want to 'create a new image', and drop the PM instance onto either of the two opened Masters).

[msmythers]

Yes Image Integration has always requires a minimum of three images. If you want to try Image Integration just make a clone of both images so you now have 4 images.


Mike

[Niall Saunders]

Ah well, Mike. Every day is a learning day !!

[Aidi]

Many thanks.   Cloned both images and ran it through Image Integration.  Seemed to have worked ok.

I'll still try the Pixel Math method..   No idea on the formula though..  Ill see what i can work out.

Rgds
Aidan

[msmythers]

Aidian

The simplest PixelMath expression would be (image1+image2)/2. You can select Create new image and call it image3 so you can check the differences in the images.


Mike

[vicent_peris]

Hi,

It's important to apply this process through ImageIntegration and not through PixelMath. The reason is that ImageIntegration will measure the noise in both images and will weight both according to that measurement.

Also, please deactivate any pixel rejection, as it does not have any sense to apply it in this process.

This process is the same as the one to create a synthetic luminance. In that case, we simply integrate the master lights from each filter together.

Best regards,
Vicent.