1.51
#...
·
·
1
like
|
|---|
| Would love any thoughts around my latest image. https://www.astrobin.com/eknzmz/ |
13.14
#...
·
·
1
like
|
|---|
| Christopher, I'm fairly new at this avocation. You have great data for sure from dark skies and great setup. I've never seen this target so close up before. I'm impressed with the detail and natural color especially the dust both the cloudy white and brownish. I don't see how to improve except maybe getting a bit more star color. This is a fantastic image. |
|
1.51
#...
·
·
1
like
|
|---|
John Favalessa: Man imo you hit it on the head. I really struggled with getting the star color better here. I have 3 other revisions of this locally that each have more star color but unfortunately it also messed up some of the other details :*(. Thanks for the feedback John! |
|
13.14
#...
·
|
|---|
| Star color is something I've and am struggling with. I use StarExterminator to remove the stars and then work on their color, but just so so results. I hope we get some good things to try from others in this thread. |
|
1.51
#...
·
·
1
like
|
|---|
John Favalessa: Yeah, star color and not blowing them out with stretching is always tricky. I will say the new Starnet module was released Starnet2. And it is absolutely amazing. It handles star removal better than any other tool I have used. |
|
13.14
#...
·
|
|---|
| I didn't notice that and will give it a try. Thanks! |
 #...
·
|
|---|
|
This is a great image Christopher, no doubt about that. The star colors are always a challenge, but I believe in this area of the sky they are mostly white, hence it does not really hurt that much 😊. If you could get a bit more sharpness into the nebula structure it would probably look a tick better (I ran a starless version of your image through Topaz and it did improve I would say), but well ... what am I saying, this is up at 99% quality at least. One question: where did you get the Starnet version 2? I tried the usual links but it seems that it has been taken off. I accidentally downloaded the batch file, and not the PI version, which sucks of course. Do you know maybe where one can find it now? Uwe |
4.44
#...
·
·
1
like
|
|---|
|
Beautiful image, Christopher, I echo the other comments on the stars. Perhaps reduce the size too? Uwe, here is some information on Starnet2. I followed instructions on the following and got it working on my Mac... https://www.galactic-hunter.com/post/starnet2 It coexists with version 1. I had to download and followed the directions, there is a cli version which is wrong so get the Pixinsight version only The green highlighted one is not right, https://sourceforge.net/projects/starnet/files/v2.0/StarNetv2PI_MacOS.zip/download is the Mac one and there is a Windows version for PI too Hope that helps! Gowri |
#...
·
|
|---|
| Thanks Gowri! I used your link and it looks like that it was taken off from this site. Strange indeed!! |
|
13.14
#...
·
·
2
likes
|
|---|
| I found another site with the v2 download. https://starnetastro.com |
|
4.44
#...
·
·
1
like
|
|---|
| Oh I saw they changed the links! Does this work? https://starnetastro.com/ |
#...
·
·
1
like
|
|---|
| HA! I tried this site this afternoon as well and it did not work. BUT ... now it did!!! Thanks John and Gowri!!! Already installed. Will do some comparisons with StarXTerminator for sure now 😊😊. |
|
13.14
#...
·
·
1
like
|
|---|
| yes, but I'm having trouble installing on my Mac. have to give security permissions...I will try again to install. |
#...
·
·
1
like
|
|---|
| Yes, same here, had to overwrite the whining of my computer 😂😂😂. |
|
13.14
#...
·
·
1
like
|
|---|
| got it installed...testing now. Runs slower than StarXterm but really good results. 2x upsampling runs very slow, slightly different result, not better in my image though. |
|
4.44
#...
·
|
|---|
| I really like it! Curious what you both think of it too! |
|
13.14
#...
·
·
2
likes
|
|---|
| Starnett 2 is definitely better than the original version, much less artifacts. Not sure if it's better than StarXterm yet. I think it may depend on the image, so I'll be trying both in the future. These are a game changer for processing for me, although sometimes when recombining the stars get residual halos or artifacts. |
6.65
#...
·
·
2
likes
|
|---|
|
Lovely image Christopher--nice detail and color to the nebulosity! The only thing to me is your stars are either overexposed or overstretched. My own version of the Running Man suffers from the same issue, someday I will go back and redo it myself, now that I know how to fix this. This area does have lots of bright stars, which makes it difficult. The stars should follow a PSF (point spread function), which means they are not just uniform intensity dots, they should fall off gradually in intensity from their centers. First always check your individual light frames to check that stars are not saturated--if one or two color channels are saturated your star color will never be correct, if all are saturated you won't be able get color. If this is the problem, then you need to reduce your gain and/or exposure time for the stars--you can't process to good star color. Note the brightest stars may always be saturated, especially with a larger aperture, even at low gain extremely short exposures (<1s) might be required for some really bright stars. For bright stars as long as the halo is not saturated you can get some color there, even if the core is white. If you lights are good, then you can reprocess the stars separately from the Nebulosity as is, otherwise redo some lights for the stars. I find adding the stars back in as a screen layer in Photoshop works well, I presume there is something similar in PixInsight. If you have the stars separated, you can carefully add a mask with some blur to reduce the halos on the stars before you add them back in. I use StarTools for processing, it has great tools for reducing halos, both intensity and or color. You can presumably do the same in PixInsight. There are tons of tutorials out there for PixInsight if you search for them. Personally StarXterminator, which I run as a plug-in in Photoshop gives better results than I ever got with Starnet++, and much faster by a long way. And convenient integrated into Photoshop. Bright stars will need a manual fix as well, I use content-aware fill in Photoshop. Not removing stars works better if you first reduce the halos. Star extraction has more difficulty with removing the halos as they can look too much like nebulosity. Hope this helps. Rick |
#...
·
·
2
likes
|
|---|
|
As info for everybody on this chain: The screen function from PS can be done with PI in Pixelmath:a*(1-(1-top)*(1-bot)) + (1-a)*bot; a = screening factor (usually from 0.7 to 1). top and bot are the image names. Rick: the new Starnet++ (V2) shows definitely improvements, give it a try. |
|
6.65
#...
·
·
2
likes
|
|---|
Thanks Uwe! Great to see the math behind the screen function. And thanks, I just downloaded the new windows version at https://starnetastro.com (though it took me a few browsers to find one whose security would allow me to download it). It is faster than before and nice to have the GUI. Quality is similar to a little worse for the image I was currently working on, for leaving halos and bright stars. For my image anyway, StarXterminator was about 2x faster and shows a little less residuals. But I will keep StarNetv2 in my tool kit, as, if I have learned one thing, no single tool is best for all images. |
#...
·
·
2
likes
|
|---|
| Yep, was the same for me Rick, needed some trials to get it finally downloaded 😊. And yes, StarXT is still a tick better, but I must admire the success of the Starnet developer! |
