1.43
#...
·
 |
|---|
|
Hi! I have for a long time been struggling to control the blue star halos that I tend to get in my images. I have tried removing both my Optolong L-enhance filter and the MPCC Coma corrector from the imaging train, but the halos tend to remain. Hence, I have now tried to change my processing routines to try to reduce the halos. My normal postprocessing software so far has been Startools. I now also have evaluation licenses for Pixinsight and StarXterminator. I have a number of examples on different processing methods on the Rosette nebula in this post in my staging area: https://astrob.in/42jx3u/B/ Please read my comment in there and give me comments on the different processing revisions, and let me know which one you like best. And if you have any suggestions on how to reduce blue star halos, please share them. Best Regards and Clear Skies during the New Year Ulf |
11.01
#...
·
|
|---|
|
I'm trying to understand what you really mean by blue star halos? I can see some bloated stars in your images, but not specifically halos. Have you tried using Starnet to extract the stars before you over-stretch the image; that then lets you focus on stretching the nebula. You can then add the stars back in after the nebula has been stretched. {update] Ooops, just saw your post about using starxterminator .. might be worth trying starnet++ in pixinsight to see if you have better results. |
|
1.43
#...
·
|
|---|
|
Andy, thanks for your answer. In revision E I have what I would call blue halos. But in most of the other revisions I have tried to reduce them in different ways, and so they are not so apparent. In rev B and C I did what you suggest and removed the stars before stretching, as is mentioned in the description. A more obvoius example of what I mean by blue halos can be seen here: https://astrob.in/pq39qg/E/ But I guess what is meant by a "halo" could be a bit different for you and me. Regards Ulf |
|
1.43
#...
·
|
|---|
| As for Starnet++ and StarXterminator, I have tried both, and my experience is that StarXterminator does a better job with stars that have diffraction spikes. |
|
11.01
#...
·
|
|---|
Andy, thanks for your answer. In revision E I have what I would call blue halos. But in most of the other revisions I have tried to reduce them in different ways, and so they are not so apparent. In rev B and C I did what you suggest and removed the stars before stretching, as is mentioned in the description. A more obvoius example of what I mean by blue halos can be seen here: The example you shared looks more like a heavy processing artefact rather than your raw data. I used Startools a lot a few years ago, however I'm now finding that Pixinsight gives me more control and insight into exactly what is being processed. So much so, that I eventually paid for the full license tonight as my trial license just ran out (ouch that was painful). |
|
11.01
#...
·
|
|---|
| One last comment: I think we both need to learn more about the star-handling thing. For me, I'm going to try stretching the image to the point where the stars are a little subdued compared to where I eventually want them before using either starnet or starxterminator to extract them knowing that I'll probably boost the combined image a little bit in photoshop at the end. |
|
7.42
#...
·
|
|---|
| As Andy said it seems to me that the so called halos are more the result of heavy stretching and manipulation (especially heavy median filltering to remove noise) rather than actual halos which are by definition a ghost image of the inlet pupil of the scope/lens, and most common with refractors. Try to post an unprocessed image to see what we're actually talking about. |
|
1.43
#...
·
|
|---|
| Andrea, I totally agree that heavy stretching is needed to bring out the blue rings around the stars. But that does not mean that it is an artefact that is not in the data to begin with. The heavy stretching may also be needed to bring out the desired nebula data. It is not clear to me what you mean by an unprocessed image. Would you mean a raw uncalibrated single frame? Or an integrated but unstretched image? I can not really post linear full depth images to Astrobin, but if you would be willing to look at it that way, I could post a download link to a raw fits or tiff file. |
|
7.42
#...
·
|
|---|
Andrea, I totally agree that heavy stretching is needed to bring out the blue rings around the stars. But that does not mean that it is an artefact that is not in the data to begin with. The heavy stretching may also be needed to bring out the desired nebula data. It is not clear to me what you mean by an unprocessed image. Would you mean a raw uncalibrated single frame? Or an integrated but unstretched image? I can not really post linear full depth images to Astrobin, but if you would be willing to look at it that way, I could post a download link to a raw fits or tiff file. Probably the last option is the best, in terms of quality of data and resolution. Or use STF in PI with a stacked-up but otherwise unprocessed image. |
|
1.43
#...
·
|
|---|
|
This a full depth tiff that was only stretched with STF in PI. https://drive.google.com/file/d/1Oa3ddExYJsW8okeCGQnrqxhrKQ9mBJr4/view?usp=sharing /Ulf |
|
7.42
#...
·
|
|---|
This a full depth tiff that was only stretched with STF in PI. Thanks for the file Ulf. As expected there is no halo at all. It is just the diffraction pattern due to the mirror clips (mostly) and obviously the secondary's spider. but mostly the former. The only cure is maybe the application of erosion/dilation morphing in PI to minimize it. Or learn to live with it. |
