Planetary processing for 1.8

[Mike in Rancho]

The other night I captured some video of Jupiter and Saturn after a little visual observing. Conditions probably not the greatest and they were getting a bit low in the sky, but I figured it would still be fun to do.

I know the basics of lucky imaging from fiddling around with it a year or so ago with not so great equipment, so can make my way through PIPP and AS3 to get a final tiff even if I am no expert. I probably need to work on camera settings for the video, as well as quality ranking and percentages of frames to use.

That said, ST seemed to do a nice job of finalizing the image, though I'm not fully certain of the workflow. A few old threads and a video I found didn't help too much (older versions as well).

It seemed to make sense to open as non-linear, since it's a bright stacked video. Also there didn't seem to be any help (only harm) from use of stretching or Wipe. So, right to enhancements. I tried using the normal "start big and whittle down to finer detail" method that the icons are pretty much set up for. Deconvolution was amazing. However, it almost seemed that, with the planetary presets, deconvolution might almost be more of a big hammer to perhaps do early, and then use other modules for the smaller detail? I have not experimented with that yet.

Still, and though it's not saying much and as fuzzy as they may still seem, they did turn out to be my best Jupiter and Saturn ever!

After each was saved from ST I used Gimp just to put them side by side for easier attaching, and then scaled them up 2x before export to jpg. Amazing how many GB are used up just to finish with a tiny 50kb image.

Jup-Sat Oct 29 2021.jpg (49.91 KiB) Viewed 2786 times

[admin]

Nice!

Indeed, depending on the source material, your stack may have been color balanced and stretched (pretty common if the source is a video).

Did you use the Plnt/Lunar preset in SVDecon? Don't be afraid to bump up the Synthetic Iterations into the hundreds. Signal is typically quite high for planetary, solar and lunar data so it will fall apart less quickly under many iterations.

Be sure to tweak the Synthetic PSF Radius as well. If it is set to high, you can sometimes observe the background (or the darkness of craters) starting to "eat into" a planet or crater detail.

[Mike in Rancho]

Aha, thanks.

I also realized that the post-processing scale-up sort of washed things out. No bueno. So I actually restacked -- keeping PIPP out of the quality culling and letting AS3 do it all, along with a drizzle.

It's webcam video avi, and probably not the greatest? This was originally an MS Lifecam Cinema, which I took apart, removed the lens and stock filter, and stuffed into a 1.5" barrel, which now has an SVB UV-IR cut on the end. Maybe 1280x720? I actually built this to be my guide scope camera. The driver has lots of control settings, and frankly I wasn't sure where to put them. I tried to expose decently fast though, I believe around 8ms frames.

The strong deconv values you suggested worked, but with care. If it got too silly looking in the detail, or developed a ring floating above Jupiter, I backed it down.

I believe the results were better, and also allowed for better ensuing color and saturation adjustment. I was still a bit too dark, and FilmDev after the fact was not working well for minor gamma changes. On a whim I tried SS, thinking of the brightness preset, but instead it turned out that Dimsmall worked very nicely for what I wanted and a bit of clarity added too. Just a touch, strength of 25%.

Planets are quite tricky! And of course this is poor data, even though I am pretty excited about it. I did develop some horizontal banding somewhere along the line, noticeable along the bottom. Perhaps the drizzle? I will have to trace it out. Still, I think it's getting better.

Jup Oct 29 2021 PIPP-AS3_Drizzle15 ST8 1C.jpg (23.71 KiB) Viewed 2778 times