Solution not found and checking bugs
Stacking program, astrometric (plate) solver, fits file viewer
Brought to you by:
han59
Menu
▾
▴
On the Intel Mac version (that I know of) the stacking has an issue with solving and selection when you have more that 35 or so images to stack.
If I have 35 images selected out of 101 , with 2 automatically unselected because of quality, (35 out of the first 37 included) it works fine.
As soon as I include another image, and select stack, the 2 previously deselected images (because of quality) become selected and it then has 38/101. But it then only randomly solves the images. In my last test only the last one solved, with all the others showing no solution. Previous attempts had the first 6-8 with no solution. It seems random.
I tried an older version from early May and found it also failed.
I have now tried the Linux version on a Raspberry Pi 4 and I am getting similar problems.
The individual images solve but are only "randomly" solving when all files are being stacked.
Paul
Last edit: Paul Homer 2023-07-30
Some further information.
The first part of the log below is where the stacking solves the first image but fails on the next 5.
When I click images 2 and 3 individually and solve each separately a solution is found within a fraction of a second. I have tried altering the downsampling and hash code tolerance and a few other settings but no difference. The only thing that seems to work is to reduce the number of images being solved to less than 38.
16:55:19 Calibrating image prior to solving.
16:55:19 Skipping dark calibration, already applied. See header keyword CALSTAT
16:55:19 Skipping flat calibration, already applied. See header keyword CALSTAT
16:55:19 Using star database D50
16:55:19 Creating grayscale x 2 binning image for solving or star alignment.
16:55:19 188 stars, 139 quads selected in the image. 142 database stars, 105 database quads required for the square search field of 0.5°. Search window at 200%
16:55:19 4 of 4 quads selected matching within 0.007 tolerance. Solution["] x:=0.006949x+ -0.744672y+ 920.846411, y:=-0.744190x+ -0.007109y+ 1246.521024
16:55:19 Solution found: 00: 46 29.0 -73° 06 26 Solved in 0.3 sec. Δ was 0.6". Mount Δα=-0.6', Δδ=1.1'. Used stars down to magnitude: 14.3
16:55:57 Calibrating image prior to solving.
16:55:58 Using star database D50
16:55:58 Creating grayscale x 2 binning image for solving or star alignment.
16:55:58 203 stars, 149 quads selected in the image. 153 database stars, 112 database quads required for the square search field of 0.5°. Search window at 200%
16:55:58 3 of 4 quads selected matching within 0.007 tolerance. Solution["] x:=0.006575x+ -0.744711y+ 921.074586, y:=-0.744335x+ -0.006522y+ 1245.618561
16:55:58 Solution found: 00: 46 29.0 -73° 06 27 Solved in 0.7 sec. Δ was 2.4'. Mount Δα=-1.1', Δδ=2.1'. Used stars down to magnitude: 14.3
16:56:07 Stack method Sigma clip average
16:56:07 Oversize 0 pixels
16:56:07 Analysing lights.
16:56:07 Analysing group NGC 261_H_300 for outliers. Average image quality (nrstars/sqr(hfdbinning))=63, σ=10.4
16:56:07 Grayscale stack (classify by light filter unchecked)
16:56:07 Stacking (Sigma clip average), HOLD ESC key to abort.
16:56:07 Analysing darks.
16:56:07 Master darks(s) ready.
16:56:07 Analysing flats.
16:56:07 Analysing flats
16:56:07 Master flat(s) ready.
16:56:07 Reference image selected based on quality (star_detections/sqr(hfd)) is: /Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_001.fits
16:56:07 Loading master dark file /Users/phomer/Astronomy/Imaging/QSI 583/darks/dark_b1g1T-15/dark_b1g1T-15_5m/2023-07-16/master_dark_61x300s_at_-15C_2023-07-16.fit
16:56:08 Loading master flat file /Users/phomer/Astronomy/Imaging/QSI 583/flats/C9.25 f6.3/2023-07-15/H/master_flat_corrected_with_flat_darks_H_51xF_51xFD_2023-07-15.fit
16:56:08 Adding light file: 1-174 "/Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_049.fits dark compensated to light average. Using 61 dark(s), 51 flat(s), 51 flat-dark(s)
16:56:08 Creating grayscale x 2 binning image for solving or star alignment.
16:56:10 Adding light file: 2-174 "/Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_002.fits dark compensated to light average. Using 61 dark(s), 51 flat(s), 51 flat-dark(s)
16:56:10 Creating grayscale x 2 binning image for solving or star alignment.
16:56:10 3 of 3 quads selected matching within 0.007 tolerance. Solution x:=0.999239x+ -0.011608y+ -11.870346, y:=0.007303x+ 0.998326y+ -3.757670
16:56:11 Adding light file: 3-174 "/Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_003.fits dark compensated to light average. Using 61 dark(s), 51 flat(s), 51 flat-dark(s)
16:56:11 Creating grayscale x 2 binning image for solving or star alignment.
16:56:11 Not enough quad matches <3 or inconsistent solution, skipping this image.
16:56:12 Adding light file: 3-174 "/Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_004.fits dark compensated to light average. Using 61 dark(s), 51 flat(s), 51 flat-dark(s)
16:56:12 Creating grayscale x 2 binning image for solving or star alignment.
16:56:12 Not enough quad matches <3 or inconsistent solution, skipping this image.
16:56:13 Adding light file: 3-174 "/Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_005.fits dark compensated to light average. Using 61 dark(s), 51 flat(s), 51 flat-dark(s)
16:56:13 Creating grayscale x 2 binning image for solving or star alignment.
16:56:13 Not enough quad matches <3 or inconsistent solution, skipping this image.
16:56:14 Adding light file: 3-174 "/Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_006.fits dark compensated to light average. Using 61 dark(s), 51 flat(s), 51 flat-dark(s)
16:56:14 Creating grayscale x 2 binning image for solving or star alignment.
16:56:14 Not enough quad matches <3 or inconsistent solution, skipping this image.
16:56:15 Adding light file: 3-174 "/Users/phomer/Astronomy/Images/Galaxies/SMC/NGC 261/C9.25 f6.3 QSI583/2023-07-29/H/NGC 261_b1g1T-15_5m_H_007.fits dark compensated to light average. Using 61 dark(s), 51 flat(s), 51 flat-dark(s)
16:56:15 ESC pressed. Execution stopped.
16:56:26 Calibrating image prior to solving.
16:56:26 Using star database D50
16:56:26 Creating grayscale x 2 binning image for solving or star alignment.
16:56:27 203 stars, 145 quads selected in the image. 153 database stars, 109 database quads required for the square search field of 0.5°. Search window at 200%
16:56:27 4 of 4 quads selected matching within 0.007 tolerance. Solution["] x:=0.006068x+ -0.744734y+ 921.946881, y:=-0.744299x+ -0.005831y+ 1244.693402
16:56:27 Solution found: 00: 46 29.4 -73° 06 31 Solved in 0.8 sec. Δ was 3.6'. Mount Δα=-1.7', Δδ=3.2'. Used stars down to magnitude: 14.3
16:58:38 Calibrating image prior to solving.
16:58:39 Using star database D50
16:58:39 Creating grayscale x 2 binning image for solving or star alignment.
16:58:39 229 stars, 172 quads selected in the image. 172 database stars, 129 database quads required for the square search field of 0.5°. Search window at 200%
16:58:39 8 of 8 quads selected matching within 0.007 tolerance. Solution["] x:=0.005428x+ -0.744668y+ 922.927429, y:=-0.744430x+ -0.005564*y+ 1244.582070
16:58:39 Solution found: 00: 46 29.5 -73° 06 32 Solved in 0.9 sec. Δ was 4.7'. Mount Δα=-2.1', Δδ=4.2'. Used stars down to magnitude: 14.4
Paul,
Can you share four or five images so I can have a look. E.g. upload them as one zip file to
https://ufile.io/
and share the link.
Han
Check also visualy the reference light. This is marked by a small crown. In rare cases it could select a frame where guiding failed and the stars are streaks.. The quality reported of this image should be low.
I checked the reference frame and it doesn't look like it is one of the better frames as it lacks some contrast but it appears to be fairly sharp.
I have noticed that when I did a test with 174 frames that the Result column has about half the frames showing 'no solution', however, there is a Solution column that shows all ticks.
Having some of your files will make it easier for me to find the cause..
Stacking 101 5 minute exposures with latest Raspberry Pi update.
The analysis stage is unflagging 5 exposures as poor quality.
After alignment and stacking 28 images have been included in the stacked image.
It ended up with 96 - 28 as having no solution, but they solve individually.
Paul
Just one more thing, the images I shared, will probably get removed soon.
Paul
A further update.
I have only encountered this bug with longer focal length images, 1400mm, but I never noticed it with my 600mm images, but due to the faint objects I am imaging I need lots of frames to remove the noise. I went back to older versions of ASTAP but it always fails.
With my latest testing I have 66 x 10 minute exposures and it works correctly, but add one more image to the stack and most images fail to solve and it only stacks 5 x 10 minutes, although the exact number is inconsistent and depends on which images I include in the stack.
Check the reference image indicated by a crown. It is automatic selected as the image with the highest quality.
Check also settings. See tab alignment. Set Downsample at 0, Maximum number of of stars to use at 500. Hash tolerance at 0.007 or 0.005. Ignore stars less the {HFD} at 0.8.
Share the series and I will have a look.
Han
I am not sure how easily this can be shared it is almost 1GB.
Interestingly 057 is the one with the crown when 66 are stacked but when you add one further image it has no solution and 075 has the crown. So most have a solution but add one more to the selection and most no longer have a solution.
I have tried most settings but currently have downsample 0, Max 500, tolerance 0.005, ignore 0.8.
I assume what is happening is some sort of overflow error.
Then image 075 is most likley to blame. Inspect that image or just don't use it.
If that doesn't help, share one gbyte is not a problem for me. But pack it in one zip file and upload to link below and share the link.
https://ufile.io/
Han