Methods to combine images of two different focal lengths

[fbutkovich]

I have two stacked images of M33, each at different focal lengths: A (400mm) and B (600mm). The B stack has 16h of integration versus the A stack only having 6hours.

The stars are nicer on version A and the framing gives me a little more room, however, the signal-to-noise ratio is better on the B stack. Is there a decent way to combine these?

Currently, this is the work flow I have done but I'm not sure if it's the best technique:

1. Dynamic crop both images
2. Dynamic background extract both images
3. Spectrophotometry–based color calibration on both images
4. Blur xterminator on both images (set to correct only)
5. Dynamic alignment using B as target result (align B to the framing of A, this generates a black backdrop to match the size of A)
6. Linear fit A to B using B as the reference
7. PixelMath
   1. Used simple addition expression ($T + B) with image A being the target

This doesn't work because I can still see the outline/profile of B within A

For context, I should probably also clarify that until this point neither version has had any stretching performed to them.

 

[cloudbait]

I have two stacked images of M33, each at different focal lengths: A (400mm) and B (600mm). The B stack has 16h of integration versus the A stack only having 6hours.

The stars are nicer on version A and the framing gives me a little more room, however, the signal-to-noise ratio is better on the B stack. Is there a decent way to combine these?

Currently, this is the work flow I have done but I'm not sure if it's the best technique:

1. Dynamic crop both images
2. Dynamic background extract both images
3. Spectrophotometry–based color calibration on both images
4. Blur xterminator on both images (set to correct only)
5. Dynamic alignment using B as target result (align B to the framing of A, this generates a black backdrop to match the size of A)
6. Linear fit A to B using B as the reference
7. PixelMath
   1. Used simple addition expression ($T + B) with image A being the target

This doesn't work because I can still see the outline/profile of B within A

For context, I should probably also clarify that until this point neither version has had any stretching performed to them.

The most common process would be to drop all of the subs into WBPP with their associated calibration frames and manually choose a reference frame, which is what everything will be aligned and resized to before being integrated together. You'd also want to set you exposure tolerance as necessary to ensure that all of them end up in a single master.

 

[fbutkovich]

@cloudbait thank you for the advice. I have yet to try WBPP and still stack with DSS, it's still an intimidating feature to me, however I'm sure once I learn it it will become invaluable.

I'm still at a point where I cross over my workflow between DSS, PixInsight, and Photoshop.

By the way nice username, pretty fitting for our hobby.

 

[cloudbait]

@cloudbait thank you for the advice. I have yet to try WBPP and still stack with DSS, it's still an intimidating feature to me, however I'm sure once I learn it it will become invaluable.

I'm still at a point where I cross over my workflow between DSS, PixInsight, and Photoshop.

By the way nice username, pretty fitting for our hobby.

Once you get comfortable with WBPP you'll never look back. It runs circles around any other calibration/stacking software out there.