M77 - Cetus A - Hitting a brick wall with Wipe

[ionia23]

Hello all,

Besides the Color module, Wipe has been one of the biggest challenges to figure out/understand for me. This image has driven me crazy trying to get it to a decent, even background.

This is a tricky object anyway. M77 has a fairly bright central area, and then this very faint oval "ring" extended out quite a bit. I managed to capture the center structure and the outer ring, but I'm having a bear of a time preserving that and wiping out the uneven field. Even masking the center structure and it's surrounding ring out (not green) doesn't seem to make a difference, and the more you crank up the aggressiveness, the more "crunchy" or "popcorn-like" images seem to become.

The software is amazing, and surprisingly intuitive even coming from a Photoshop background. I've had great success with many other images using this application for processing. Plenty of useful documentation and so forth, but I suppose sometimes it doesn't really hurt to ask.

So here I am, bend in knee, hat in hand, asking if anyone would be willing to take a shot at this. i'm sure I captured more than I think I did, but I must be doing something wrong. If you get some decent results, please share with me your observations that I may add them to my notes. I know I'm missing several "somethings". Who knows. What I learn here I might be able to apply to some of the others I've done.

This is the stacked FIT image, converted to RGB and all that before cropping. Original image is 752x582 (roughly).

Orion Starshoot G4 OSC with 0.8x focal reducer/field flattener
Williams-Thompson 105mm APO refractor
Orion Starshoot Autoguider in ST-4 mode (direct to mount)
Orion Atlas equatorial mount
No filters, urban Tucson skies (roughly 5.5 magnitude) no clouds, wind, in my front yard.


9x60 second exposures (Yes, yes, more is much more preferable to less)
subtracted dark and flat frames.
Stacking was done in Fitswork4, which I am absolutely loving vs. Deep Sky Stacker. Hell of a learning curve, though.

All debayering was done in Orion Camera Studio as it doesn't seem any other application out there can handle the CYMK matrix of this camera. No biggie.

This file should open straight up in Startools.

https://drive.google.com/file/d/1u4aGNQ ... sp=sharing

I believe I've shared this link out correctly so anyone can download it and the file is not large at all.

Thank y'all in advance for your assistance, and remember to keep looking up.

[almcl]

Interesting!

Had a go at your data and agree you have captured quite an amazing amount of detail in only 9 minutes.

Not sure if your chosen pre-processing method is actually helping though. The posted fits file seems to have been stretched which is best avoided for good results. Also the red channel appears quite weak, although this might be down to gradient removal or other issues.

Anyway, here's what I got on a quick run through:

m77.jpg (23.45 KiB) Viewed 3397 times

[Mike in Rancho]

I checked it out also last night with similar results. Interesting data, and yeah I almost can't believe it is all of 9 minutes?

Still, it is troublesome data. The stars and galactic core are very blown out. Very high gain, or as stated, maybe pre-stretched somehow? And while ST seemed to open it as well as it could, quick checks in both Gimp and DSS to try to see the histogram both mostly failed.

You mentioned something about subtracted dark and flat frames. Can you explain more? Standard calibration procedure of course is to subtract darks from lights, subtract dark flats or bias from flats, and then divide the corrected lights by the corrected flats.

Still, I did what I could and there's some detail to be had. The first Wipe was I believe 96% aggressiveness, with a high DAF of about 6. For a second try, I used the inverted mask to clear out a wide oval around the galaxy to see if it would preserve any faint detail while allowing removal of the gradient. That said, in this case I am uncertain of the validity of this move. Seemed too much like painting, and I get the feeling that I didn't preserve actual detail, but instead preserved noise and it just looks like possible detail.

Other than that I compressed the shadows in both AutoDev and Contrast, and variously used settings in HDR and/or SVD to try to tame things down. But it can only do so much here. Color was mostly a guessing game, as sampling is not going to work well here. I looked at Stellarium and some online samples to try to match, but, doubt I was able to succeed.

There also appear to be remnant hot pixels left in the stacking result that are pretending to be some brightly colored stars (they don't exist in Stellarium).

Anyway, these were what I came up with.

ionia23s - m77 - Cetus A ST8 1A.jpg (66.28 KiB) Viewed 3364 times

ionia23s - m77 - Cetus A ST8 2A.jpg (63.86 KiB) Viewed 3364 times

I didn't try it, but wonder if treating this as planetary might have helped. Might be just too oversaturated though.

[ionia23]

Hey,

I meant to extend a special thank you for pointing out that my data is pre-stretched. it turns out you are correct. I have to do all by debayering from Orion Camera Studio since I have yet to find another application that can properly handle its rather funky bayer matrix.

The results I've gotten are pretty close to what others have gotten with the same data. This did wonders for my self-confidence, and I am very grateful for that.

I had a setting enabled "Auto adjust images" or something like that. I took that out and redid the initial stack. Did better.

Im going to do a full re-shoot of M77. Shorter exposures but a whole lot more of them. What's really nifty is by readdressing pre-stretched data, I was able to pull this off the other night.



M106

This is 50 x 30 seconds, no flats, no bias, no darks. Stacked in Fitswork4 with about 10 alignment points. All the rest was just startools. And I really, really like this.