Sh2-86 in Ha

A place to share images that you have taken with SharpCap.
Forum rules
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.

Please share the equipment used and if possible camera settings to help others.
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Re: Sh2-86 in Ha

#11

Post by timh »

Hi Brian,

On the question you posted above on mixing RGB and Ha using tools the only method that I know about uses the tools in PixInsight. However the same principles must apply...? So in case it is of any use to you

The steps are ..

1) Align the combined RGBL image and the HA images either in preprocessing (aligned versus the same reference image) or postprocessing using an alignment tool (in PixInsight this is 'Dynamic Align')

2) Take the combined RGBL image - do all the steps that you normally would while it is still linear (background subtraction etc) - then stretch the image as normal so that it is in a non-linear form and looks right

3) Extract the luminance channel from the RGB. - call it RGB-L

4) Take the Ha image do all the steps that you normally would while it is still linear (background subtraction etc) - then stretch the image as normal to non-linear

5) Re-calibrate the extracted RGB-L so that it is stretched to the same scale/ dynamic range as the greyscale HA image (in PI this process is called 'Linear Fit' and would be applied to the RGB-L image using the HA image as the reference)

6) Now add the RGB-L image to the HA image using the 'Max' function. This works in PixInsight Pixelmath to produce a combined image - call it RGB-L-HA - where each pixel takes the maximum of the two images from which it is constituted.

7) Finally replace the luminance of the RGBL image with the new luminance channel , RB-L-HA , created in the above process. The PI tool for this is called LRGB combination.

Thus far I have found that the above works very well indeed on lots of objects --- the HA adding a great deal of extent and detail to RGB images of for example M1 and the flame nebula and tadpoles in Auriga.

best wishes
TimH
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Sh2-86 in Ha

#12

Post by oopfan »

Tim,

Thank you! I never would have figured that out on my own.

Well, last night was a "fail" but also a good learning experience. I captured RGB. However, as you can see, there are five stars that the Star Size Reduction failed to pick up. I tried all the sliders to no avail. I can understand it, from the algorithm's perspective. Those stars look like elliptical galaxies.

I was mostly concerned with over-exposing the stars in the cluster. I started with a single 150s test exposure in green. The cluster was just a bit too blown out, so I reduced the exposure to 120s. That satisfied the algorithm, but those five other stars are about 0.3 magnitude brighter. I think if I had lowered the exposure to 90s, I would have been OK.

Here is the 2-hour RGB stack:

Brian
Attachments
Sh2-86_R-19x240s-b1_G-23x120s-b1.jpg
Sh2-86_R-19x240s-b1_G-23x120s-b1.jpg (988.04 KiB) Viewed 1033 times
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Sh2-86 in Ha

#13

Post by oopfan »

I was able to squelch those five remaining bright stars. Sorry about that.

Explanation: As you know I don't capture Blue, only Red and Green. And for Luminance, I use Wratten #12. However, since I have a 5-position filter wheel, I don't have room for the Wratten #12 right now. So, last night I only captured Red and Green data. In APP's Color Combine, my standard workflow is to assign Wratten #12 data to Luminance, 50% of Red data to Red, 50% of Red data to Green, 50% of Green data to Green, and 50% of Green data to Blue. That works out quite well. In my previous reply, since I don't have Wratten #12 data, I made the mistake of not assigning anything to Luminance. That unfortunately had a negative impact on the Star Size Reduction algorithm. Now, I assigned Red data in equal amounts to Red, Green, and Luminance. And, Green data to Green, Blue, and Luminance. Fortunately that satisfies the Star Size Reduction Algorithm.

Brian
Attachments
Sh2-86_R-19x240s-b1_G-23x120s-b1.jpg
Sh2-86_R-19x240s-b1_G-23x120s-b1.jpg (951.83 KiB) Viewed 1030 times
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Sh2-86 in Ha

#14

Post by oopfan »

What do you think of the following image? I did everything within APP except for the final denoising and curves adjustment. I really had to boost saturation since I've only got 2 hours of color, compared to 5 hours of luminance (Ha).

It needs more color.
It needs more overall integration time.

First, I think I'll bulk up on color so that the colors are richer, and the image doesn't look so strained.
And then I'll increase integration time for all to smooth out the noise.

Anyone want to take a guess as to what the total integration time will eventually be?

Brian
Attachments
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1.jpg
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1.jpg (947.42 KiB) Viewed 1027 times
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Sh2-86 in Ha

#15

Post by oopfan »

Alright, so I discovered that if I inject a little of the Red and Green data into Luminance, then the red in the nebula really pops.

I readily admit that I need more data. This image is looking a bit contrived. That is a sure sign that it is over-processed.

Brian
Attachments
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1_RG-Lum.jpg
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1_RG-Lum.jpg (977.5 KiB) Viewed 1021 times
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Sh2-86 in Ha

#16

Post by oopfan »

Well, I think this is a dramatic improvement.

I changed strategies. In my previous attempts, I loaded Ha into Luminance, and relied on the RGB stacks to supply the nebula's color. In theory it should work, but it requires LOTS of integration time in RGB. Now I load Ha to Red, which if you think about it makes a lot of sense. However, the problem is star colors. Proper star colors are in the RGB stacks. If I were to simply combine the Ha with the RGB, then there would be an over-abundance of red which skews star colors. The solution is to remove the stars from the Ha stack. Thankfully, I can achieve that with APP 1.0.8.3-beta2.

Next step is to combine the nebula and the stars. I used Affinity Photo. (I should look to see if APP can do it, but Affinity seemed to be the intuitive choice.) Affinity has a tool available from the File menu called "New Stack..." (not to be confused with "New Astrophotography Stack...") Afterwards, I cropped it, denoised it, and adjusted shadows and highlights.) That was it!

Let me know if you think there are too many stars. I did use the Star Size Reducer on the RGB stack, but I only reduced them by about 40%. I found that I began to lose star color if I reduced more. All stars wanted to go white. I think that the proper solution is to recapture RGB using a shorter exposure to lessen star saturation/star bloat.

Here is the result:

Brian
Attachments
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1.jpg
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1.jpg (901.23 KiB) Viewed 1002 times
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: Sh2-86 in Ha

#17

Post by turfpit »

A good image there Brian. You must have spent many hours on the capture and experimenting with processing.

Dave
timh
Posts: 515
Joined: Mon Aug 26, 2019 5:50 pm

Re: Sh2-86 in Ha

#18

Post by timh »

Hi Brian,

I think that you have got there (although in this hobby improvement is neverending :-) ) and it looks good. Your best solution ended up the converse of the way that worked for me with the tadpoles (viewtopic.php?f=16&t=3835) where adding HA to luminance worked better than adding it to the red channel. It may be to do with the objects --or maybe the software?

I'd be very happy to have a go at postprocessing your data in PixInsight if you like -- just to see what it makes of it ?

best wishes
TimH
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Sh2-86 in Ha

#19

Post by oopfan »

Thanks gentlemen, I finally settled on a different variation:

1. Reduced size of stars by 60% instead of 40%.
2. Pushed the contrast harder to clearly show luminance variations within the nebula.

Item #2 was only possible after applying item #1. However, there is a cost. The bright stars are beginning to show artifacts. Like I mentioned, the proper thing is to recapture RGB with a shorter exposure to prevent star bloat.

Tim: Thanks for the offer! I'll look into uploading the frames to the Cloud. The calibration frames will add a lot to the size of the upload. Perhaps you could first try processing the FITS files that APP generated: the Ha stack and the RGB stack. What do you think?

Brian
Attachments
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1_ver2.jpg
Sh2-86_Ha-7nm-31x600s-b1_R-19x240s-b1_G-23x120s-b1_ver2.jpg (921.23 KiB) Viewed 991 times
Post Reply