Tadpoles Ha-Oiii

[tooth_dr]

This is an image taken on the morning of the 3rd of October, between 00:30 and 05:47.  Using two scopes I was able to gather just over 5 hours data each channel Ha and Oiii.

Stacking done in APP and processing done in PS. 

98-99% moon during data capture.

 

 

 

[AKB]

Really dramatic!  That's a lot of data in a short time.  Like it!

Tony

[Skipper Billy]

That's a glorious image and very elegantly processed. Well done.

[tooth_dr]

12 minutes ago, AKB said:

Really dramatic!  That's a lot of data in a short time.  Like it!

Tony

Thanks Tony.  I've been emailing Takahashi back and forth for some help, plus have had good advice on here, and whilst the Epsilon isnt 100% collimated with my Nikon, it's pretty decent at this sensor size, so I'll not touch it again until I'm feeling bored.   It certainly collects serious data when it gets going.  Below is my starless Ha stack from the Epsilon F2.8 and the Oiii stack in the ED80 F6.37 - some difference.  The alignment of the scopes is also off, but I have plan to correct this when I get a few hours spare.

 

Edited October 6, 2020 by tooth_dr

[tooth_dr]

11 minutes ago, Skipper Billy said:

That's a glorious image and very elegantly processed. Well done.

Thanks David, that means a lot to me 👍

[MarkAR]

Superb image.

[symmetal]

Excellent.  I always thought it wasn't worth doing Oiii with significant moonlight, but it certainly adds a significant amount to the final image. 😀

Alan

[Xiga]

Cracking image Adam. That's a lot of nebulosity, and i really love the detail in the dark dust! 

Something looks off with the Oiii data though, it looks too soft to my eye. I went back and had a look at my own Oiii stack (done in APP) of this area from a couple of years ago, which had a similar amount of exposure (4 hrs) but mine was shot with the Nikon D5300. I quickly ran it through Topaz Denoise AI and then starnet, then resized it in PS to roughly match yours. No sharpening or contrast enhancement was done. As you can see, it looks quite a bit sharper, which i wouldn't have thought was possible. 

How was your focus on the Oiii?

 

[tooth_dr]

4 hours ago, Xiga said:

Cracking image Adam. That's a lot of nebulosity, and i really love the detail in the dark dust! 

Something looks off with the Oiii data though, it looks too soft to my eye. I went back and had a look at my own Oiii stack (done in APP) of this area from a couple of years ago, which had a similar amount of exposure (4 hrs) but mine was shot with the Nikon D5300. I quickly ran it through Topaz Denoise AI and then starnet, then resized it in PS to roughly match yours. No sharpening or contrast enhancement was done. As you can see, it looks quite a bit sharper, which i wouldn't have thought was possible. 

How was your focus on the Oiii?

 

Thanks Ciarán!  Looks like i attached the Oiii that i applied the tone mapping tutorial to, so it has some Gaussian blur etc applied.

 

[geeklee]

Beautiful image and just as detailed browsing @ 1:1

What an instrument that Epsilon is for capture (plus your post processing skills of course! 😁)

When I saw the brilliant Ha stack in another thread (I think), I was looking forward to more.

[Xiga]

5 hours ago, tooth_dr said:

Thanks Ciarán!  Looks like i attached the Oiii that i applied the tone mapping tutorial to, so it has some Gaussian blur etc applied.

 

Ah yes, makes sense now.

It's been a while since I last looked at J.P Metsavainio's Tonemapping guide, but it's been around a while now so there are parts of it that can be improved upon imho. Have a go at this rough workflow...

Crop all your data at the beginning, right after you've stretched and saved them. Layer them all up in PS, then crop, then save each one off individually so they are all still aligned and cropped equally.

Throw the Oiii data into Topaz Denoise AI. Turn down the sharpness to zero and experiment with the amount of NR, e.g somewhere between 5 and 15 is usually good. Save. Then use PS to convert it back to Grayscale. If the stars now have some artefacts around them, which is common, then use a layer mask (of the image itself, with contrast boosted) which should fix them.

Now run starnet on it, followed by some spot healing brush if needs be.

Doing it this way, you should end up with a much cleaner tonemap, without as much reliance on then having to blur the data.

HTH.

[tooth_dr]

59 minutes ago, Xiga said:

If the stars now have some artefacts around them, which is common, then use a layer mask (of the image itself, with contrast boosted) which should fix them.

Firstly Ciarán thanks for this!  You have so much knowledge!  Is this literally just create a layer mask then copy and paste the image into the later mask and boost the contrast on the later mask.  Do you use curves to do this?

 

1 hour ago, Xiga said:

followed by some spot healing brush if needs be.

I never seem to do this ‘right’. Any links or pointers?

 

Thanks again Ciarán. I now plan another bights imaging on this then a full reprocess using the above. 

[Xiga]

2 hours ago, tooth_dr said:

Firstly Ciarán thanks for this!  You have so much knowledge!  Is this literally just create a layer mask then copy and paste the image into the later mask and boost the contrast on the later mask.  Do you use curves to do this?

 

I never seem to do this ‘right’. Any links or pointers?

 

Thanks again Ciarán. I now plan another bights imaging on this then a full reprocess using the above. 

What i like to do is, intentionally use a bit too much NR in Topaz Denoise ai. Then bring it into PS, layer it on top of the Pre-DNAI version and apply a layer mask to dial the effect back a good bit. The layer mask is just an inverted copy of the original image itself, with a Levels adjustment to clip the white point and also bring the black point up to just past the heel of the curve, enough to make all the stars nice and black. This way, the stars are protected, and the brightest parts of the image won't have as much NR being applied. Topaz does tighten up the stars quite a bit. You might like how it looks, or you might not, so it'll be up to you to decide if you want to fully protect them or not. There's always the Opacity slider too 😀 Personally, i like a certain amount of the tighter stars it produces. It can give the appearance of having used Deconvolution. Certainly the really teeny tiniest stars will need some protecting though, as they will almost surely end up looking like artefacts if neglected. 

Now that i think about it, i may have got the order wrong above. I can't recall if i do Denoise ai then starnet, or the other way around, so try it both ways and just go with whichever one gives you the cleaner result. 

As for the Spot Healing brush, there's not much i can add really. I just adjust the size of the brush for each click (using the square brackets keys). Don't go any bigger than you need to. That's it really. 

[Xiga]

Ran a short 100 min Ha stack through Denoise AI as an example. These are just jpg screenshots, but they should be good enough. 

Original stack:

 

With some Denoise AI. Notice the harsh edges to some of the stars, especially the very smallest ones:

 

Layer mask applied. Inverted version of the original image:

 

Levels adjustment to the layer mask:

 

Final version. Notice the star edges are now much better. Less NR has also been applied to the high signal areas. If you want more NR, then just use a higher setting in DNAI.

[tooth_dr]

1 hour ago, Xiga said:

Ran a short 100 min Ha stack through Denoise AI as an example. These are just jpg screenshots, but they should be good enough. 

Original stack:

 

With some Denoise AI. Notice the harsh edges to some of the stars, especially the very smallest ones:

 

Layer mask applied. Inverted version of the original image:

 

Levels adjustment to the layer mask:

 

Final version. Notice the star edges are now much better. Less NR has also been applied to the high signal areas. If you want more NR, then just use a higher setting in DNAI.

Legend again Ciaran, thanks for this.

[Adam1234]

Great image

[tooth_dr]

1 minute ago, Adam1234 said:

Great image

Thanks Adam.

[newbie alert]

Have always enjoyed your images Adam...this one's no different.. superb

[tooth_dr]

2 minutes ago, newbie alert said:

Have always enjoyed your images Adam...this one's no different.. superb

Very kind, thanks for that, really appreciate your comment.

Adam.

[astro mick]

Another really good result Adam.

I,ve never been too pleased with my starless images,the more i stretch it seems to degrade rapidly.

Hey but thats me.

M.