StarTools 1.9 preview

General discussion about StarTools.
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: StarTools 1.9 preview

Post by almcl »

Having a little bit of trouble with 1.9's SVDecon.

Several times now the auto mask generator presents a screen completely devoid of green or yellow star cores within the 'starfish'.

Here's an example:
SVDecon1.jpg
SVDecon1.jpg (202.68 KiB) Viewed 13820 times
Zooming in, while there aren't many candidates, there are some, but they appear to have been ignored in favour of what to me look like noisy pixels. Perhaps the data has been stretched too much, but this close up shows what I am seeing:
SVDecon 2.jpg
SVDecon 2.jpg (151.76 KiB) Viewed 13820 times
Should this be happening?
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: StarTools 1.9 preview

Post by admin »

almcl wrote: Thu Dec 15, 2022 3:01 pm Having a little bit of trouble with 1.9's SVDecon.
Interesting! I would love to have a look at the dataset and see what might be going on here. It is precisely these sort of (potential) failure cases I am looking for.
Ivo Jager
StarTools creator and astronomy enthusiast
almcl
Posts: 263
Joined: Wed Jan 21, 2015 7:15 pm
Location: Shropshire. UK

Re: StarTools 1.9 preview

Post by almcl »

PM sent with link, Ivo.

Let me know if you need anything else?
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
Mike in Rancho
Posts: 1141
Joined: Sun Jun 20, 2021 10:05 pm
Location: Alta Loma, CA

Re: StarTools 1.9 preview

Post by Mike in Rancho »

I had an interesting one too this morning. Fiddling with some older DSLR data on M13. The field of outlines did have a lot of small and peculiar selections, but there were also a lot of very good star candidates outlined as well, all across the field.

However, the blue box size was tiny! In most cases way inside the white outline. I'll have to try again to give it better inspection, but at first glance it seemed that a lot of stellar profile was being left on the table.

Now, I was testing this data out without any binning (full APS-C) and maybe that was part of it - the blue boxing just didn't scale itself well to the viable star candidates that were in fact found?
alacant
Posts: 211
Joined: Mon Jan 25, 2016 7:03 am

Re: StarTools 1.9 preview

Post by alacant »

SV Decon no longer requires separation of samples when using built-in mask generator
Hi everyone

Not sure about what this means.

The spiky style star mask in the new SVDecon makes it difficult to select stars for sampling. Maybe we don't need to select stars in 'Sampling'... (?)
Any guidelines for this? Optimum settings? Need for sampling?

TIA and clear skies,
Steve
decay
Posts: 433
Joined: Sat Apr 10, 2021 12:28 pm
Location: Germany, NRW

Re: StarTools 1.9 preview

Post by decay »

almcl wrote: Thu Dec 15, 2022 3:01 pm but they appear to have been ignored in favour of what to me look like noisy pixels
Something similar here: There are some correctly choosen starfishies, but a lot of good candidates are ignored and yes - a lot of empty starfishies are there as well. :think:

I could provide the dataset and log, if that would be of any use.

Best regards, Dietmar.
2022-12-16 15_40_50-StarTools.jpg
2022-12-16 15_40_50-StarTools.jpg (239.74 KiB) Viewed 13749 times
hixx
Posts: 250
Joined: Mon Sep 02, 2019 3:36 pm

Re: StarTools 1.9 preview

Post by hixx »

Hi Mike,
Now, I was testing this data out without any binning (full APS-C) and maybe that was part of it - the blue boxing just didn't scale itself well to the viable star candidates that were in fact found?
What image size does Your cam have? If I am not mistaken, the blue boxes should autoscale, covering the full star blob. It works for my 24 MPx full-frame. I do no BIN neither and still the blue boxes are shaped according to Star size, covering the full star on decent candidates. Even close stars get separated.

However,I find it hard identifying good candidates in a crowded star field which seems all yellow-green to me
Cheers,
Jochen
Mike in Rancho
Posts: 1141
Joined: Sun Jun 20, 2021 10:05 pm
Location: Alta Loma, CA

Re: StarTools 1.9 preview

Post by Mike in Rancho »

Dietmar and Jochen,

Yeah, it's early development, though working pretty well even so I must say. And frankly there's going to be a nearly infinite range of dataset types that will be thrown at this, so...a tough thing to do.

If I recall from the prior 1.8 starfishy discussion - the stellar profile and quality coloring is pulled up from the linear stage, but the white outlining can be affected by other actions such as your autodev?

My 2600 and D5300 are APS-C both more or less in the ballpark ( ;) ) of 6000x4000. Yes the blue boxing is autoscaling, but the question is what is SVD using as its model for that autoscaling. If it's some kind of average size of all the white outlines, it could end up too small if there's a field full of tiny noisy spots that got white outlined, leaving the box undersized for the fewer, larger, but actual quality stars outlined and that will be actually used for sampling. :think:

I had gone back to some data from earlier in the year on M13 for some testing (was actually testing something else but stumbled into this). Data is so-so. I hadn't yet refined my Newt for collimation and everything else, and didn't yet have my EFW and RGB filters. So I combined an hour of L from the 2600 and an hour of OSC from the D5300. It still came up with a fine result in 1.8 I think, and might be in the gallery here back in May or so.

Running these tests, including just on the D5300 data alone before extraction (not linked), depending on lack of bin and stretch/noise I can come up with a SVD sampling mask that pretty much has no stars to choose at all - just tiny outlines of mostly invisible noise or perhaps faint stars?

Here though, composited as the registered LRGB, and not binned, I get a ton of unusable outlines but also a decent sampling of good candidates. But note the blue box size (I've had even smaller). Note the zoom % of course, so I'm way way deep. It's not really too bad, but does strike me as undersized.

SVD9 sample screen.jpg
SVD9 sample screen.jpg (94 KiB) Viewed 13736 times

But if I then undo that SVD, bin to 35%, and run SVD again - I can get blue boxing much more appropriate to the white candidate outlines. Of course the SNR is improved too which may assist the white outlining. :confusion-shrug:

I haven't yet tried full size/no-bin on some of my recent data (almost all SHO) but will this weekend if I can, to see what more I can find.

This link has these old M13 files though - L and the extracted RGB, all registered; and the log I used for this testing if anyone wants to see the settings in crop, wipe, etc. and SVD masking.

https://drive.google.com/drive/folders/ ... sp=sharing
alacant
Posts: 211
Joined: Mon Jan 25, 2016 7:03 am

Re: StarTools 1.9 preview

Post by alacant »

This, for example, greatly reduces the need for things like the Shrink module.
Hi
JTOL...
But please don't remove 'shrink'. It has saved many an image for me!
The combination of svdecon and shrink produces a far superior result to those awful star-removal algorithms which are currently fashionable and making images look so false.
Cheers
jhart
Posts: 47
Joined: Wed Aug 26, 2020 10:08 pm

Re: StarTools 1.9 preview

Post by jhart »

Hi,

For what it is worth, I am having the same problem as Steve in SVDecon. With Ha and OIII data captured with an L-Ultimate, no matter how I load the data in ST9, the "starfish" white splatters do not line up with the stars.

Jeff
Decon.jpg
Decon.jpg (392.66 KiB) Viewed 13700 times
Post Reply