Preprossing error

Image processing, astrometry, photometry, etc.
Post Reply
User avatar
Rudi
Posts: 152
Joined: 08 Jan 2019, 04:47

Preprossing error

Post by Rudi » 30 Oct 2019, 07:42

Hi,

An error occurs with the 4'th image, when I am trying to stack. What can I try to let AA7 succeed? Any parameters that I can tweak?


Preprocessing:

Code: Select all

#PREPROCESSING
Image=C:\Users\rbr\Dropbox\Main\astro\2019\1029\ngc7331\ngc7331_180s_004_L.fit
Image=C:\Users\rbr\Dropbox\Main\astro\2019\1029\ngc7331\ngc7331_180s_001_L.fit
Image=C:\Users\rbr\Dropbox\Main\astro\2019\1029\ngc7331\ngc7331_180s_002_L.fit
Image=C:\Users\rbr\Dropbox\Main\astro\2019\1029\ngc7331\ngc7331_180s_003_L.fit
CombineMethod=5
CombineDark=1
CombineFlat=1
CombineF.Dark=1
CombineBias=1
ScaledDarksImages=0
ScaledDarksFlats=0
DefectMapImages=0
DefectMapFlats=0
AlignMethod=2
AlignCopyBorders=0
FollowMP=0
DemosaicRGB=0
DemosaicCMY=0
RemoveNegative=1
NormalizeBackground=1
Sigma=1.8
ConfirmAll=0
ShowMasters=0
CreateLog=1
Multithreading=1
ExtraFilters=0
Error:

Code: Select all

Preprocessing. 2019-10-30 , 08:35:14

Sigma+Sum: 4 Images

1) ngc7331_180s_004_L.fit
Range = 4352 : 65504   Background = 6320
Alignment: dx = 0.0 dy = 0.0

2) ngc7331_180s_001_L.fit
Range = 4784 : 65504   Background = 6736
Alignment: dx = 1.17 dy = 1.01 rot = 0.0°

3) ngc7331_180s_002_L.fit
Range = 4496 : 65504   Background = 6608
Alignment: dx = -0.09 dy = 1.48 rot = 0.0°

4) ngc7331_180s_003_L.fit
Range = 4496 : 65504   Background = 6464
Alignment: Error
*** Reject bad images ***

Result image: 3/4 images.
Sigma: Rejected pixels = 0.0 %
77.72 seconds
Preferences are as follows: The files are here:
https://www.dropbox.com/s/ejqs3shzyi0nu ... 1.zip?dl=0
/Rudi

fabdev
Posts: 461
Joined: 03 Dec 2018, 21:43

Re: Preprocessing error

Post by fabdev » 30 Oct 2019, 16:30

Hi, the problem is big saturated stars, for the moment the only way to solve is to use the align method "One star" (or "Two stars" if you want to correct rotation).

Procedure:

1) Before launching the preoprocessing, open the first image, select a star (ideal is a medium bright star, isolated, eg. with no brighter stars near it).
2) Close the image, click Preprocessing, select Align, "One star" and the coordinates of that star will be shown.

Astroart does not consider big and saturated stars as "stars" because they are not good for astrometry and photometry, but for aligning and plate solve they can be useful, so a correction will be made in the future. Such stars can happen on images bigger than 3000 pixels (if strong saturation is present as well).

User avatar
Rudi
Posts: 152
Joined: 08 Jan 2019, 04:47

Re: Preprossing error

Post by Rudi » 30 Oct 2019, 18:12

Thanks Fabio, I will try two stars. But later, it's clear right now, so I am busy :D

/Rudi

User avatar
Rudi
Posts: 152
Joined: 08 Jan 2019, 04:47

Re: Preprossing error

Post by Rudi » 31 Oct 2019, 05:24

Thanks for the advise, your "two stars" suggestion works fine with the above four lights.

I've checked, and the "two stars" workarround works just fine:
  1. Open first fit file in the list to be stacked
  2. Select two stars
  3. Close the fit file again
  4. Start Preprocessing and use the autoalignment->two stars method
I assume, that the most optimal, will be to select two stars as far apart as possible, but present on all images?
And two unsaturated stars as well?
So two unsaturated stars near opposite corners present in all images?
/Rudi

fabdev
Posts: 461
Joined: 03 Dec 2018, 21:43

Re: Preprocessing error

Post by fabdev » 31 Oct 2019, 14:37

Yes, two separated stars, for example near the corners, but not too close to the borders.
By the way "two stars" is useful only if the image have rotation too, maybe "one star" is sufficient for those images. Try a comparison of the result.

A bit of saturation in the center of the star is not a problem, if the rest of the shape of star is correct.

Post Reply