Star Alignment Failures

meavydev

New member
I have just got the PixInsight trial, so I have been trying to compare images to APP.
However I can't get PixInsight Star Alignment to work with images that were fine in APP.

I started off with the WeightedBatchPreProcessor script, which failed at the StarAlignment step, so now I have been trying running just StarAlignment with the debayered calibrated images.

The best I have get is about 5 pair matches:
Code:
Loading target file:
D:/Astro/ZWO_Galaxies/PixInsight/calibrated/light/debayered/M_51_Light_016_c_d.xisf
Loading image: w=4944 h=3284 n=3 RGB Float32
3 image properties
92 FITS keyword(s) extracted.
ICC profile extracted: 'sRGB IEC61966-2.1', 3144 bytes.
Noise reduction:
Structure map:
Detecting stars: done
20 stars found.
Computing intersection ...
Intersection:
R1          : {  421,  189, 4871, 3145}    31 stars
R2          : {  423,  182, 4873, 3138}    17 stars
alpha_21    : 0 deg
Matching stars:
6 putative star pair matches.
Performing RANSAC: done
* Previous attempt failed - this is try #2
useScaleDifferences=false
Matching stars:
10 putative star pair matches.
Performing RANSAC: done
* Previous attempt failed - this is try #3
useScaleDifferences=true
* Reference image: Limiting to 30 brightest stars.
Matching stars:
*** 4 star pair matches found - need at least six matched stars.
* Previous attempt failed - this is try #4
useScaleDifferences=false
* Reference image: Limiting to 30 brightest stars.
Matching stars:
8 putative star pair matches.
Performing RANSAC: done
* Previous attempt failed - this is try #5
useScaleDifferences=true
* Reference image: Limiting to 15 brightest stars.
* Target image: Limiting to 15 brightest stars.
Matching stars:
*** 3 star pair matches found - need at least six matched stars.
* Previous attempt failed - this is try #6
useScaleDifferences=false
* Reference image: Limiting to 15 brightest stars.
* Target image: Limiting to 15 brightest stars.
Matching stars:
*** 3 star pair matches found - need at least six matched stars.
* Previous attempt failed - this is try #7
useScaleDifferences=true
* Reference image: Limiting to 8 brightest stars.
* Target image: Limiting to 8 brightest stars.
Matching stars:
*** 5 star pair matches found - need at least six matched stars.
* Previous attempt failed - this is try #8
useScaleDifferences=false
* Reference image: Limiting to 8 brightest stars.
* Target image: Limiting to 8 brightest stars.
Matching stars:
*** 5 star pair matches found - need at least six matched stars.

*** Error:
*** Error: Unable to find an initial set of putative star pair matches.

I have uploaded this image and a reference one, plus a JPG from using the APP integration which I have stretched in PI Histogram tool (which is much better than APP... I hoped to just use PI for all the steps)
See my GDrive link: https://drive.google.com/drive/folders/1UwcadxBg6JglcRssQu-kthyZWNEpKu6_?usp=sharing
 
are you in Mosaic mode in StarAlignment? or do you have a preview defined on the reference image (which must be a View?)

this line:

R1 : { 421, 189, 4871, 3145} 31 stars R2 : { 423, 182, 4873, 3138} 17 stars

indicates that you are probably in mosaic mode.

fields such as m51 can be hard as it is out of the plane of the milky way and thus there are not very many stars. the best thing to do is run StarAlignment on one of the two images in "detected stars" mode. that way you can figure out if SA is picking up noise (not likely here given the small # of stars) or if it is just not finding most stars (in which case you can decrease the sensitivity slider, or increase the peak response slider) and run again.

rob
 
Default StarAlignment parameters have been optimized for accurate image registration working with data of reasonable quality. High noise, poor focus and bad tracking are preventing normal star detection on these images.

To register the two images you have uploaded with StarAlignment, set detection scales = 6 and noise reduction = 4 (Star Detection section). Leave the rest of parameters with default values. At any rate, try to acquire better data with higher SNR and better star profiles.
 
Default StarAlignment parameters have been optimized for accurate image registration working with data of reasonable quality. High noise, poor focus and bad tracking are preventing normal star detection on these images.

To register the two images you have uploaded with StarAlignment, set detection scales = 6 and noise reduction = 4 (Star Detection section). Leave the rest of parameters with default values. At any rate, try to acquire better data with higher SNR and better star profiles.
OK thanks.
As far as focus and tracking go, that would be nice, but no bahtinov mask yet and having problems with guider, so I wanted to do a comparison between PI and APP, so perhaps PI is only for professionals?
 
Define professional :)

PixInsight has a large user base in more than 95 countries, and is being used successfully with an immense variety of instruments and conditions, mostly by amateur astronomers. It is also working in academic research and development environments. What is important here is not whether you are a professional or an amateur. You should realize that astrophotography is an extremely difficult and demanding discipline. There are several basic, critically important procedures that you must implement correctly before even thinking about taking images. Among these, you must be able to acquire data with reasonable SNR, good focus and good tracking. Otherwise you'll be acquiring marginal data. Whoever tries to convince you that these steps can be obviated is just lying.
 
Define professional :)

PixInsight has a large user base in more than 95 countries, and is being used successfully with an immense variety of instruments and conditions, mostly by amateur astronomers. It is also working in academic research and development environments. What is important here is not whether you are a professional or an amateur. You should realize that astrophotography is an extremely difficult and demanding discipline. There are several basic, critically important procedures that you must implement correctly before even thinking about taking images. Among these, you must be able to acquire data with reasonable SNR, good focus and good tracking. Otherwise you'll be acquiring marginal data. Whoever tries to convince you that these steps can be obviated is just lying.
Juan: Are you able to give us some ballpark estimate of the size of the user base? Or, as I suspect, that is proprietary info?
 
Back
Top