using lights from two different nights

[llinnell]

I am trying to use lights from two different nights, each night shooting the same galaxy but not perfectly aligned with each other, and one night darker than the other. I am using a common reference image for star alignment but when I use ImageIntegration the resulting image shows a clear line where the lights don't perfectly align. What tool do I need to use to get the "brightness" of each set of images calibrated with each other?

[macnmotion]

You don't. Use Dynamic Crop to crop only the portion of the FOV in your stacked image that is common to both nights' images.

[llinnell]

Then how does one make a mosaic that contains multiple images, each overlapping 20-30%, taken on different nights?

[RickS]

There are a couple of tools for matching mosaics.  One is the Frame Adaption capability of StarAlignment.  The other is the GradientMergeMosaic process.  Normally these aren't necessary for a single panel image.  Sounds like your alignment difference between nights is a lot bigger than we'd usually expect

[macnmotion]

A mosaic is different, as the parts that will seam together and become part of the final image are all full signal. If you had one panel of the mosaic that had the issue you describe in your post, and the misalignment amount for that single panel is greater than the overlap you've set for your mosaic, you'd be out of luck with the mosaic. I suppose using PI or Photoshop you might be able to try to illuminate the darker part of the image to match the central portion, but as we have said you'd be dealing with a portion of the FOV that does not have the complete signal from both nights. That doesn't only result in a darker area, but less signal in your object.

Andy

[llinnell]

Andy:

So the situation is as follows: 32 stacked lights from night one and 32 stacked lights from night two. Night one is the left side of the galaxy and night two is the right side - with about a 30% overlap.
So the left and right sides of the mosaic have a strength of 32 images but the center 30% where they overlap has 64 images of strength. That center part is "darker" than the two sides that only have 32 images.

This must be a solvable issue or how do people make mosaic's without showing the "seams" where the images overlap.

Lloyd

[RickS]

This must be a solvable issue or how do people make mosaic's without showing the "seams" where the images overlap.

As I mentioned above, GradientMergeMosaic is a process that can help with this.  You'll need to process as two separate panels and then combine them.  You may find this tricky if your main subject is in the overlap area - that's something I'd try to avoid when planning a deliberate mosaic.

Cheers,
Rick.

[macnmotion]

Hi Lloyd,

As Rick mentioned, the mosaic merge process is different from the normal registration and integration process. What you're envisioning is the results from aligning and integrating, not mosaic-merging. Read this tutorial, see if it gets you started:

http://www.lightvortexastronomy.com/tutorial-preparing-a-mosaic.html

or this video:

https://www.youtube.com/watch?v=H2r3ZYqCaP8

Andy