Registration of different scales

Camera and field of view are the same, but different telephoto lenses. I work with an external SSD hard drive, type Sandisk Extreme 900.

@minusman,

So the 2 lenses have the same field of view ?

Another possible place for the bug, could be normalization ( allthough I doubt it), You used advanced normalization right?

Mabula

Or do you mean with different fields of view, different focal lengths? Because then the fields of vision are different. 🤔 

Posted by: minusman

Or do you mean with different fields of view, different focal lengths? Because then the fields of vision are different. 🤔 

Yes, indeed, the field of views then don't match which shouldn't be a problem. But if the area of bad integration matches the field of view of part of the images, than that's an indication for me.

@minusman,

Can't seem to reproduce the problem for now I am afraid ... 😐 

I integrated 200 frames of a Nikon D5100 with exposures of only 2,5 seconds shot from a simple tripod, no tracking, no guiding.

LNMWC 1 iteration kappa 4 , diffraction protection set at 20 (so 10% of number of lights)

Bayer Drizzle, droplets 2.0 pixel, scale 1.0

LNC 1ste degree 3 iterations, MBB 25%

The edges are the results of the field of view slowly moving between the frame since no tracking, guiding.

So the integration with the LN Winsor clip filter brought no change. Try LN MAD Sigma Clip now. Have additionally deactivated diffraction protection.

Data Set 1.  FOV 108.00x72.00 arc-minutes with Baader UHC-S Filter    Refraktor

Data Set 2.  FOV 122.00x78.00 arc-minutes                                                    Maksutov

Data Set 3.  FOV 122.00x78.00 arc-minutes                                                    Sigma Telephoto lens

Data Set 4.  FOV 78.00x52.00 arc-minutes                                                      Maksutov

Data Set 5.  FOV 80.00x53.00 arc-minutes                                                      Maksutov

Camera Canon EOS 6D unmodified Pixelsize 6,54 microns

Hello Mabula, I will change my test procedure. I first examine the individual data sets. 🧐 

Ok, thanks, let me know if you encounter anything suspicious... 😉 

Hello Mabula, could successfully integrate the data set individually. Then I have the data set started to integrate with each other, which has also worked, see picture.

With the increase of the LNC Degree from 1 to 8 and then the psychedelic integration came about.

Posted by: minusman

Hello Mabula, could successfully integrate the data set individually. Then I have the data set started to integrate with each other, which has also worked, see picture.

With the increase of the LNC Degree from 1 to 8 and then the psychedelic integration came about.

 

@minusman,

Thank you very much 😉 for testing this and locating the problem, which then is high degree LNC clearly.

I will run tests to see if I can duplicate this now 😉 and fix / upgrade LNC which is almost due..

Regarding LNC, in my own experience, the lowest degree that gives a good result is probably the best. Higher degrees tend to introduce same waves in the data, which I need to fix in a future version.

Cheers,

Mabula

😀

Hello Mabula, here is the finished picture. Just do not know if I'm right with the Star Color Calibration. So I'm completely satisfied. I only had difficulty with uploading the picture. There always came error message.

@minusman,

That looks great 😉 thanks for sharing !

Colors are very nice as well I think, maybe a bit on the blue side, but only slightly. It's very close I think.

Nice saturation in the stars.

Mabula