Horsehead Nebula Mono Hydrogen Alpha

[Ikonnikov]

This is the Ha (15x 30 mins with the Chroma 3nm) to go with the LRGB data I used here:

I haven't managed to satisfactorily incorporate it yet, I think because it's a lot stronger than the RGB and it reddens the image too much (and overwhelms a lot of the brown dust which I want to preserve). Ideally I'd increase the amount of LRGB data substantially to help with this but it's too late for that now this year; guess I'll have to work on my processing skills a bit more. Will also try to rework the stars in (L)RGB to tame them somewhat.

Paul

 

 

[Tim]

That's nice and deep. Combining strong Ha like this with LRGB and keeping it "real" is very tricky.

I don't know if you have Registar, but that program makes it easy to quickly assess the impacts of various mixes. For instance, you can combine a percentage of the Ha with each colour channel, gauging what percentage by estimating the amount of the Ha which is picked up in each colour filter. For instance, I would estimate that my gear picks up around 5-10% of the Ha through the green filter, 15-20% through blue, and about 35% through red. My luminance filter probably gets about 25%. In Registar you can quickly test these out and make scaling adjustments.

One of the challenges is making sure the tight details picked up in the Lum subs make it through to the final image without getting swamped by the Ha. For instance, the wispy little dark tendril in the fuzz of bright nebulosity just below the horsehead, it just doesn't show up in the Ha, but is quite prominent in the Luminance subs.

At any rate, your LRGB picture is first class, a real belter, should be amazing once you get the Ha in there. Good luck!

Tim

[newbie alert]

Amazing horsehead..i really do like a ha image in greyscale format..the horse sitting on a cushion of dust is just awesome..almost 3d..fantastic..heres my effort just to show you just how good yours is

[tooth_dr]

It’s hard to get the right words! Some much detail EVERYWHERE! Incredible. 

[pipnina]

An incredible image!

The detail, the small stars, the wispy filaments & defined edges of the dust and gas is incredible. You should be well proud of this, I would be!

[Ikonnikov]

Thanks for the positive comments Tim, newbie alert, tooth_dr and pipnina.

1 hour ago, Tim said:

That's nice and deep. Combining strong Ha like this with LRGB and keeping it "real" is very tricky.

I don't know if you have Registar, but that program makes it easy to quickly assess the impacts of various mixes. For instance, you can combine a percentage of the Ha with each colour channel, gauging what percentage by estimating the amount of the Ha which is picked up in each colour filter. For instance, I would estimate that my gear picks up around 5-10% of the Ha through the green filter, 15-20% through blue, and about 35% through red. My luminance filter probably gets about 25%. In Registar you can quickly test these out and make scaling adjustments.

One of the challenges is making sure the tight details picked up in the Lum subs make it through to the final image without getting swamped by the Ha. For instance, the wispy little dark tendril in the fuzz of bright nebulosity just below the horsehead, it just doesn't show up in the Ha, but is quite prominent in the Luminance subs.

At any rate, your LRGB picture is first class, a real belter, should be amazing once you get the Ha in there. Good luck!

Tim

I've considered getting Registar for combining data from different focal lengths (although both my scopes now have diffraction spikes which may prove tricky to align) but didn't know anything about its other functions. Not thought about adding Ha signal to G and B in RGB images; is this to simulate Hbeta? I guess I could do similar in pixinsight with pixelmath but it would be more clunky.

In previous attempts at imaging the Horse I'd assumed the L and Ha were essentially the same (albeit cleaner signal in Ha) except for the blue reflection nebulae, but putting together this image noticed some of the other differences as you say. When the snr of L and Ha are similar I've found doing a maximum intensity combination works ok to make a new L channel and preserve features from both but with this data set some of the residual noise from the L sits on top of the smoother Ha nebulosity in a lot of places and looks ugly; it probably needs adding more carefully (and iteratively).

Paul

[Ikonnikov]

1 hour ago, newbie alert said:

Amazing horsehead..i really do like a ha image in greyscale format..the horse sitting on a cushion of dust is just awesome..almost 3d..fantastic..heres my effort just to show you just how good yours is

Looks like you just need more data to give you some more signal to work with, how many subs did you use and what setup?

[Tim]

I use Registar to register LRG&B data to my Ha, and also to quickly test out different colour combos for narrowband. It has a "Merge to RGB" function where you can quickly assign any of the data sets to any colour, or combine them with whatever weighting you choose, and then merge to RGB.

For instance, I sometimes add OIII to both blue and green channel, in various weights, before combining to give Ha-Oiii+G-Oiii+B etc.

Or I noticed a while back that Sii data most often is very similar to a short Ha exposure or even a medium Red one. If I dont get time to do Sii, I simply use Red datasets with a small percentage of Ha and the result is remarkably similar to using Sii, without the gruelling gathering of generally weak noisy data!

Please @Tim tag me when you post the final image, would love to see it.

Cheers

 

[Ikonnikov]

4 minutes ago, Tim said:

I use Registar to register LRG&B data to my Ha, and also to quickly test out different colour combos for narrowband. It has a "Merge to RGB" function where you can quickly assign any of the data sets to any colour, or combine them with whatever weighting you choose, and then merge to RGB.

For instance, I sometimes add OIII to both blue and green channel, in various weights, before combining to give Ha-Oiii+G-Oiii+B etc.

Or I noticed a while back that Sii data most often is very similar to a short Ha exposure or even a medium Red one. If I dont get time to do Sii, I simply use Red datasets with a small percentage of Ha and the result is remarkably similar to using Sii, without the gruelling gathering of generally weak noisy data!

 

Good to know, especially considering the price of a 50mm Chroma SII filter these days (the only NB one I'm lacking).

7 minutes ago, Tim said:

Please @Tim tag me when you post the final image, would love to see it.

 

Sure (although it might be a while before I get something I'm happy with), would be grateful if you could tell me how though as never done that before! 

[Tim]

17 minutes ago, Ikonnikov said:

Good to know, especially considering the price of a 50mm Chroma SII filter these days (the only NB one I'm lacking).

Sure (although it might be a while before I get something I'm happy with), would be grateful if you could tell me how though as never done that before! 

Yep, I could never bring myself to shell out for the 2" Sii either!

If you just type the @ symbol it will bring up a list of names and automatically tag the one you choose  Thanks

[Ikonnikov]

1 minute ago, Tim said:

Yep, I could never bring myself to shell out for the 2" Sii either!

If you just type the @ symbol it will bring up a list of names and automatically tag the one you choose  Thanks

Ok cheers!

[Tim]

As it goes, I've just spent the last 5 hours calibrating, stacking, local normalising, and drizzling my own attempts at a Ha version of the same target. Determined to do things by the book in Pixinsight for this one, including all the weighting equations. Would love to have the speed and gather of an Epsilon on it, but don't think my skies or wallet will justify one  

[Ikonnikov]

Which scope are you using (from previous posts I thought you were toting an Esprit 150 which is about two E130D's worth!)? Will you be posting it soon?

[Stephen Green]

A very nice image and detail

[newbie alert]

10 hours ago, Ikonnikov said:

Looks like you just need more data to give you some more signal to work with, how many subs did you use and what setup?

Yes you're probably right..think it' around the 10x 10 min subs area..maybe less.. As I don't have good south views at home I have to go out,and if it's clear on a work night it' not possible..roll on autumn I say..

I use a William optics ZS80D,6nm HA astronomik clip filter in a canon 100d

[Allinthehead]

That's looking great Paul. I was wondering how the Ha would improve your already great Rgb image.

[Ikonnikov]

1 hour ago, newbie alert said:

Yes you're probably right..think it' around the 10x 10 min subs area..maybe less.. As I don't have good south views at home I have to go out,and if it's clear on a work night it' not possible..roll on autumn I say..

I use a William optics ZS80D,6nm HA astronomik clip filter in a canon 100d

Yes it's not in a great place in uk skies to image. DSLR and Ha filter combination is also a slow business. Have you tried a uhc-s type filter (the IDAS LP V4 perhaps being the pick of the bunch)? Although they've got a wider bandpass around Ha than your astronomic filter they also let you collect some blue reflection nebula and oiii emission nebula signal at the same time (if you want to make a colour image).

[newbie alert]

2 hours ago, Ikonnikov said:

Yes it's not in a great place in uk skies to image. DSLR and Ha filter combination is also a slow business. Have you tried a uhc-s type filter (the IDAS LP V4 perhaps being the pick of the bunch)? Although they've got a wider bandpass around Ha than your astronomic filter they also let you collect some blue reflection nebula and oiii emission nebula signal at the same time (if you want to make a colour image).

Yes that filter has been suggested to me before..just starting to dip my toe in the realms of ccd imaging..so maybe once it comes around again at the end of the year I can have another try..

[Barry-Wilson]

Keep experimenting Paul as the Ha will certainly bring further depth and detail.  However you're right that it can morph brown dust into red.

I use PI's MultiChannel Synthesis HaRVB-AIP script, experimenting with the amount of Ha to be incorporated and then often blending the two resultant versions to get my desired result.  Blue star halos and salmon-pink Ha can then be muted in the blend.

[Ikonnikov]

Thanks for the advice Barry, I haven't used the HaRVB script before and on further reading it seems that it's been removed from the version of PI I'm currently running (1.8.05); any idea on how to get it back? Also I presume you use this in preference to the NBRGB script (which is still present)?

[Barry-Wilson]

Hi Paul

I was quite disappointed when it was removed from the latest release.  It is still available to download and install as a script - do a Google search.

Let me know if you can't find it and I'll upload to to Dropbox.

HTH

[Ikonnikov]

Thanks Barry will give it a go.