Galaxies... "Mark of Zorro" (& Gradients!) removed

[Macavity]

"Amateurs" (sic) of STACKED video images will be familiar with the "Mark of Zorro"!

No? Well, it's the penalty we pay for unguided imaging. As Registax locks onto stars

etc., (any) HOT pixels are rendered into a "Z shape" when the images are stacked. 

Such is fairly trivially removed by dark frame subtraction! Well, not always so here...

Usually dark frames make a big difference, but sometimes a residual "Zed" is still visible.

Alternatively rendered as a Z-shaped hole in m'galaxies. lol. But now I notice, Registax

contains a Darkframe "scaling" option 0.5, 1.0, 1.5... 10.0! - And this (sortof) works!

However, VIDEO cameras seem to be quirky beasts... they run HOT, they vary rapidly in

temperature? For whatever reason, it seems useful to be able to "scale" my dark frames 

beyond the limited factors 0.5x (1.5x thru 10x!) available in Registax... 0.6x, 0.8x etc.

To get a more accurate subtraction - Using any common imaging software, I found that, 

if I take the dark frame and add it to a layer above a *black* canvas image, I can "scale"

the dark frame by adjusting the transparency. Obvious in retrospect, but so useful!

Scaling is found by trial and error. But no more "Mark of Zorro" (+ve or -ve) for me!

Aside: I believe some VIDEO Cameras defeat attempts to take "dark frames" by the

AGC facility. Perhaps there is some hope for those too?

Now to the next problem! lol. Certainly with my Watec and *longer* (stacked) images

there is still a residual GRADIENT on the images. This is not removed by "darks".

However, I have found a good solution in the IRIS software. Even the default (menu

driven!) polynomial fit *usually* gives an improvement on my galaxy images:

BEFORE:

AFTER:

Okay, Okay... it's debatable! But I certainly reckon the *worst* of the gradient is

gone. There is some loss of good data? But the overall effect more pleasing!

Other problems (the various dust "blobs"! etc.) seem more worth addressing...

With 100 x 10.24 sec (256x) = 20 Minutes, you begin to see M51 halo stars.

A few more of the above ilk:

C32 (The Whale) interacting with...

... NGC4657 (The Hockey Stick)

And finally M101 revisited!

About as far as it can be messed-with? Beyond this you have to look to 16 bits,

and classical imaging techniques? (Longer exposure, auto-guiding etc. etc.) 

BUT I can still pop off about TEN of these per night, before fatigue sets in!

[Grogfish]

Great images, and thanks for sharing your hard-learnt techniques. The Polynomial gradient certainly appears to give improved contrast and remove amp glow, though maybe at the loss of a little detail.