LDN1235 Shark Nebula

[ollypenrice]

Two panel mosaic (right and left halves in this orientation) with 64x3 mins on the left and 53x3 mins on the right. Some high haze for the right hand side meant the right had more noise. This is a crop.

RASA 8, ASI2600 OSC, Avalon Linear Fast Reverse, joint venture with Paul Kummer. Instrument based in our robotic shed and driven by Paul from the UK.

After ABE and SCNR green in PI, this is my Photoshop processing. Images stitched in Registar.

Olly

Edit: scroll down the thread for a more punchy processing.

Edited September 9, 2021 by ollypenrice
clarification

[ollypenrice]

LDN1235 to paddleboard, LDN1235 to paddleboard: You need a bigger boat.

[Davey-T]

Nice Shark Olly, I only managed the head half from a dark site before lockdown struck.

Dave

[gorann]

Stars are excellent Olly, I think you have that problem solved👍, but I suspect there is more colour and structure in your data to be brought out, especially in the dust. At least I found a bit more in my attempt on the shark with a similar integration time: https://www.astrobin.com/6s82qb/C/ That came at the cost of more star halos so maybe try the Starnet approach on the dust? (which I have not done yet)

Edited September 8, 2021 by gorann

[ollypenrice]

8 hours ago, gorann said:

Stars are excellent Olly, I think you have that problem solved👍, but I suspect there is more colour and structure in your data to be brought out, especially in the dust. At least I found a bit more in my attempt on the shark with a similar integration time: https://www.astrobin.com/6s82qb/C/ That came at the cost of more star halos so maybe try the Starnet approach on the dust? (which I have not done yet)

Yes, mine's rather a conservative rendition. I'll see what I can find. I was rather focused on the stars, I suspect.

Olly

[Magnum]

You are getting the hang of the new setup fast Olly, what gain an offset are you using? 

Lee

[ollypenrice]

Here's a more pushy version, still trying to keep the stars down. It's interesting just how different it is processing RASA data as compared with refractor.

Olly

 

[wimvb]

Nice catch, Olly. It looks like you also caught a few galaxies in your new photon hoover.

[peter shah]

A lovely rendition Olly, I am seeing dark haloing around your stars. Looks like it could be an artefact when recombining your stars with the background?

[ollypenrice]

43 minutes ago, peter shah said:

A lovely rendition Olly, I am seeing dark haloing around your stars. Looks like it could be an artefact when recombining your stars with the background?

I didn't remove stars this time, Peter. I don't think my PC will do it on these images. Starnet keels over. The effect is real, though, and probably arises from various tricks for enhancing local contrast. I've done some fixing already but clearly not enough!

Olly

[peter shah]

14 minutes ago, ollypenrice said:

I didn't remove stars this time, Peter. I don't think my PC will do it on these images. Starnet keels over. The effect is real, though, and probably arises from various tricks for enhancing local contrast. I've done some fixing already but clearly not enough!

Olly

Thats interesting could it then be coming from the gradient removal method or sharpening perhaps? EI know effect from a highpass filter can also do something similar to stars.

Edited September 9, 2021 by peter shah

[ollypenrice]

13 minutes ago, peter shah said:

Thats interesting could it then be coming from the gradient removal method or sharpening perhaps? EI know effect from a highpass filter can also do something similar to stars.

I think it comes from local contrast enhancement. I used three methods to get more structure out of the dust.

1) Noel's Actions applied only in Blend Mode Darken.

2) Application of a greyscale equalized and blurred image copy as a layer mask.

3) Local Histogram Equalization in PI.

I think that 1 and 3 always run the risk of darkening the areas around stars. However, most renditions of this target do have large stars and I wanted to see if I could have small ones.

Olly

[peter shah]

8 minutes ago, ollypenrice said:

most renditions of this target do have large stars and I wanted to see if I could have small ones.

I agree, you certainly have small stars and they bring such a wonderful amount of colour to the whole field..... I always enjoy your images Olly

[Tommohawk]

Smashing image Olly.

I've never been accused of being subtle, but FWIW I prefer the original version. 

[kirkster501]

This is a really hard subject to image. I have failed miserably in Bortle 5 skies. Great image Olly, chapeau mate.

[tomato]

"Big fish, swallow you whole"

Sorry, its my all time favourite movie. I'm also under a Bortle 5 sky but I will have to give this a go with the RASA.

[BrendanC]

I tried this the other night with appalling results. So yes, I agree with everyone else - very tough object, very well done.

[wimvb]

3 hours ago, ollypenrice said:

3) Local Histogram Equalization in PI

I have abandoned LHE in favour of MMT (Multiscale Median Transform). LHE increased noise in my images, and didn't give me the same control that MMT does. With MMT I can increase sharpening or local contrast (which is the same thing) exactly where I want it. It can even control colour saturation locally (by selecting Chrominance as the target).

I used MMT on my version of the IKI M81/82 dataset to get better definition in the arms of M81. I didn't need to use any sort of noise reduction (having 100 hrs of data helps).

I must admit that I've never used MMT on such a rich star field as this image. It is very possible that MMT won't give any better results, because it will fatten stars, so these will need to be protected.

Edited September 9, 2021 by wimvb

[ollypenrice]

8 hours ago, wimvb said:

I have abandoned LHE in favour of MMT (Multiscale Median Transform). LHE increased noise in my images, and didn't give me the same control that MMT does. With MMT I can increase sharpening or local contrast (which is the same thing) exactly where I want it. It can even control colour saturation locally (by selecting Chrominance as the target).

I used MMT on my version of the IKI M81/82 dataset to get better definition in the arms of M81. I didn't need to use any sort of noise reduction (having 100 hrs of data helps).

I must admit that I've never used MMT on such a rich star field as this image. It is very possible that MMT won't give any better results, because it will fatten stars, so these will need to be protected.

You've mentioned MMT before but I'd forgotten. I'll look it up. I only ever use LHE-modified images as a layer in Photoshop, so very selectively. (This kind of shameless promiscuity makes me fear a visit from the Spanish Inquisition who, as is well known, lie behind Pïxinsight...)

😁lly

[ollypenrice]

Using a pure log stretch of the stack for reference I've tried to correct the contrast artifacts around bright stars and groups of stars. Thanks to  @peter shah for picking this up.

Olly

[wimvb]

1 hour ago, ollypenrice said:

You've mentioned MMT before but I'd forgotten. I'll look it up.

And I realise that I sort of promised to do a write up on my work flow. I’ll try to squeeze that in somewhere this weekend.

[peter shah]

45 minutes ago, ollypenrice said:

Using a pure log stretch of the stack for reference I've tried to correct the contrast artifacts around bright stars and groups of stars. Thanks to  @peter shah for picking this up.

Olly

 That's made a big difference