Frustrating Star Alignment

[RichInLesta]

Although new to PixInsight, I was able to successfully process an Andromeda image and am quite pleased with the results.  For my second astrophoto, I took a number of images of Triangulum that I'm trying to process.  Admittedly, they are pretty noisy having been taken with a DSLR (Canon 70D) through a Celestron 8" SCT.  I don't yet have guiding capability, so I limited my exposures to short (15 second) bursts at 12,800 ISO to avoid trailing.  I think it worked well for star shape, but am wondering if the higher noise is the cause of the issue I'm having, which is as follows:

I followed the pre-processing steps posted here (http://www.lightvortexastronomy.com/tutorial-pre-processing-calibrating-and-stacking-images-in-pixinsight.html), so I'll skip all the details up to the point where I'm having problems.  And the problem is that Star Alignment fails.  I keep getting an error message (actually, numerous) regarding the failure of detecting adequate stars.  Here is a sample:

Matching stars:
*** 0 star pair matches found - need at least six matched stars.
* Previous attempt failed - this is try #12
useScaleDifferences=false
* Reference image: Limiting to 8 brightest stars.
* Target image: Limiting to 8 brightest stars.
Matching stars:
*** 0 star pair matches found - need at least six matched stars.

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

I've searched and found numerous discussions regarding this error, and have tried several recommended solutions, but to no avail.  I believe the images I'm trying to align are adequate, and have attached an image showing a sample of one of the images, as well as a "Structure Map" derived from Star Alignment.  I've tried as few as two or three images, so that I can see if it was related to drift, but nothing helped.  To me, it certainly looks like I have enough stars, and they're pretty closely aligned from one image to the next, but cannot figure out why Star Alignment can't detect them.

I realize that I'm probably not providing nearly enough information, but I can provide anything needed to help me solve this dilemma.

Thanks, in advance, for any help you can provide.

[mschuster]

Can you post a dropbox link to a zip containing two images, the reference and one of the failing targets?

Sometimes it takes some parameter tweaking. Sometime StarAlignment > Star Detection > Noise Reduction 1 or 2 helps, but looking at your images this may not work.

Also why the high ISO? It seems to me high ISO just scales up both signal and noise, does not change SNR, and hence is of no use. Except maybe for saturating bright stars, which is a negative.

Thanks,
Mike

[RichInLesta]

The first link is to the reference image.  The image is one of ten that I'm trying to align, which I understand is "do-able"; that is, I can use one of the target images as a reference...if this isn't accurate, please let me know.  I first put the images through the Subframe Selector script and chose the best rated image.  The second link is to the lowest rated target image i'm trying to align.  And the others (8 at this time until I figure out what I'm doing wrong) are in between those two.

https://www.dropbox.com/s/uppk90etbbghl3p/debayer_LIGHT_15s_12800iso_%2B45f_20161227-18h45m06s096ms_c_cc_a.fit?dl=0
https://www.dropbox.com/s/9trj9i4xhee8707/debayer_LIGHT_15s_12800iso_%2B48f_20161227-18h51m09s974ms_c_cc_a.fit?dl=0 

As for the reason for the high ISO:  When shooting with my DSLR, I aim to get an image that peaks at ~25% of the histogram.  I know I can get that several ways; longer exposures at lower ISO or shorter exposures at higher ISO.  When I was taking test shots at lower ISO's (800, 1600, etc.) I was getting elongated stars because I'm not guiding (yet).  So at 12,800 and 15 seconds, I was achieving my 25% histogram target and getting decent star shapes.  So I went with those settings for this target.

Thanks.

[mschuster]

Hi Rich, the first link won't download. The second works.

Thanks,
Mike

[RichInLesta]

Don't know what happened there, but this one should work.  This is the reference image.

https://www.dropbox.com/s/uppk90etbbghl3p/debayer_LIGHT_15s_12800iso_%2B45f_20161227-18h45m06s096ms_c_cc_a.fit?dl=0

[mschuster]

Ok, thanks.

Star Alignment >Star Detection > Detection scales = 6 works on this pair.

The stars are a bit large, elongated, and saturated. 6 helps detect large stars.

I suggest abandoning your 25% histogram target to avoid saturating the stars, and make them easier to process. Try capturing at ISO 100, stay with your 15 sec exposures. IMO 25% histo goal is hurting not helping.

Thanks,
Mike

[RichInLesta]

Based on your testing, I set the Detection Scale to 8 (max) and played with the sensitivity (which I had read as an option) and was able to successfully align my images.  I'll finish up later to see what the results of these settings give me, and will tinker with those to optimize the results, but wanted to let you know that your feedback worked for me.  I greatly appreciate your time.

Also, although a long-time terrestrial photographer, I'm finding that imaging the sky is quite a different animal.  I'll give a lower ISO a shot at my next opportunity, and forego the 25% "target" for now.

Thanks again for the help.