Is Wipe required for color processing later?

[Chunkles]

Hi,

I've been working on my OSC data that was taken through an L-Enhance filter (dual band). I noticed that my data didn't have any obvious gradients, and the Wipe module seemed to make the image look worse and noisier. So I skipped it but when I got to the Color module, I couldn't bring out any color at all (all grey). So I went back and ran Wipe and then I was able to see the colors just fine in the Color module.
So is Wipe a requirement during processing? And if not, why weren't any colors visible when I didn't use it?

Thanks!
Scott

[admin]

Hi,

I've been working on my OSC data that was taken through an L-Enhance filter (dual band). I noticed that my data didn't have any obvious gradients, and the Wipe module seemed to make the image look worse and noisier. So I skipped it but when I got to the Color module, I couldn't bring out any color at all (all grey). So I went back and ran Wipe and then I was able to see the colors just fine in the Color module.
So is Wipe a requirement during processing? And if not, why weren't any colors visible when I didn't use it?

Thanks!
Scott

Hi Scott,

A few things to unpack here;

I noticed that my data didn't have any obvious gradients

Wipe not only models and eliminates gradients, it also eliminates any bias signal (e.g. a single signal that is added to all pixels).

Wipe module seemed to make the image look worse and noisier

The Wipe module does not make your image noisier - it does not materially modify noise signatures. Can you provide an example?
Perhaps you are confused by the diagnostics stretch in Wipe? If that particular stretch appears to make your image noisier, then that is a good thing. It means Wipe has cleaned up your data so much that the diagnostics stretch now starts to latch on to shot noise, as there is nothing else to 'bring out' any more. If shot noise is the last 'issue' in your stack, then you are doing well!

So I skipped it but when I got to the Color module, I couldn't bring out any color at all

As said above, Wipe eliminates any bias signals, which is a prerequisite for color balancing. In short, in most instances, you cannot and should not skip Wipe.

If working on OSC data, you are told to inspect your chrominance (color) data as well in Wipe. You should observe Wipe making a marked difference there for your color data, even if in cases where it does not do much for your luminance data.

Hope this helps!

[Chunkles]

Thanks Ivo!

The more I played with it, I understood what you were saying. I just used the Wipe module very lightly. This is what I was able to get from my 7 hours of integration time in a heavily light polluted area using the L-Enhance filter on my ASI533:

If you or anyone else wants to play with the stacked data, I'd love to see how it can be improved. File here:
https://drive.google.com/file/d/1HgJJAN ... sp=sharing
Specifically, the gigantic stars. I know the bright one just below and to the left of the Horsehead is kind of a blue nebulous area, but I had trouble getting all the green out. And some of the other overexposed stars show little rings near their centers. I'm sure this was due to something I did in processing.

Thanks, as usual, for all your help!
-Scott

[admin]

Hi Scott,

How did you stack this?

Something odd is going on if this is really 7h worth of data;

Selection_434.jpg (222.5 KiB) Viewed 4387 times

StarTools intentional aliasing scaling is betraying a latent pattern in your data which spells trouble.
Indeed, close up we can see some severe correlated noise, as well as some odd Bayer artefacts.

1. Are you dithering at all?
2. Have you implemented all recommended settings for your stacker?

There is vignetting present that is not caught by your flats. I needed to use the Vignetting preset in the Wipe module to flatten the field.

The color data seems to be very, very skewed to the red side of things (some sort of stacker issue or setting?). That said, pulling things back yielded a color rendition that can be hoped for when using the L-Enhance filter;

StarTools_436.jpg (311.16 KiB) Viewed 4387 times

Just following the standard workflow (after binning) with mostly defaults, you should get something like this.

Horsehead2_SB.jpg (194.79 KiB) Viewed 4387 times

Hope this helps,

[Chunkles]

Hi Ivo,

Yes, I was dithering. I'd have to check my APT settings. I think it was every 4 images maybe. The distance dithered might have been very small though, I was totally guessing on those settings.
As to the stacking, I re-stacked those images a bunch of times. I might have posted the wrong stack. I had multiple nights of exposures and differing exposure lengths also on the different nights. So trying to get everything working in DSS was giving me a fit. I'll look more closely and see what I posted, as well as what I may have done wrong with DSS. I did just get the trial of APP that I wanted to try stacking with, but haven't had a chance to play with it yet. Thanks for your help and more to come ...

-Scott

(I did use the compose module in 1.7 for the first time and really liked how the Bi-Color setting works)

[Burly]

Dither calculator here for working out Apt dither using phd2 just edit to suit look for a dither of 12-15 pixels
https://docs.google.com/file/d/18wes9gw ... pe=msexcel

[Chunkles]

Thanks, Burly. I think I saw your posts about this on the APT forum. Since I am not using PH2 or a guide camera at all, I'm not sure if that calculator would work for the APT guiding feature. But regardless, I think I had it set way too low anyway ... not nearly enough pixels!

[admin]

Thanks Scott. Always interested to hear solutions to little problems like these. Do let us know how you get on!

PS Also give ASTAP a try for stacking (setting for StarTools here).