IC1396 in 7nm Ha

[ianaiken]

Hi all. My first image this season is IC1396 in Ha. I used a Baader 7nm Ha and captured 20 subs @ 1200s over three nights. Imaging camera was an Atik 314L+ on a SW ED80 DS-PRO sitting on top an NEQ6 guided with Finderguider/PHD. I used Sequence Generator Pro for acquisition which helped significantly to centre the image each session with the plate solving functionality.

Stacking and post processing (albeit not a lot) was done in PixInsight. Processing was Crop, Rotate, Histogram Stretch, a bit curves for contrast and a tiny bit ACDNR. Exported to TIF and saved to JPG via PhotoShop. Still trying to get a 100% true representation in JPG format to match how the FIT looks.

I may obtain more subs to further reduce the noise on this image and smooth it out a bit more. I had aimed to get 30 subs but it's difficult when you have to build and tear down the kit each session. I believe when my OBS is built I will have double the imaging time per session! I also may add some RGB from my OSC and create a HaRGB in the future.

Image can also be seen on http://www.astrobin.com/full/20527/

[Black Knight]

Looking really good. I'd be interested to see how the combine with OSC data goes.

[Gina]

Nice start

[ianaiken]

Thanks both! I'm looking forward to getting some RGB now with the OSC. I wouldn't mind getting some more narrowband but I don't have any filters other than Ha at the moment.

[DMouse]

You have some great detail there, especially for an ED80. But you have quite a bit of noise. Try looking up the tutorial on the PixInsight forum, for ATWT noise reduction, there are a couple of good ones out there. But you must apply this to a linear image before it is stretched. It does a better job than ACDNR.

Clear Skies

Paul

[ianaiken]

Paul - cheers. I'm always looking for new skills with PI thanks for the lead. I'll go give it a go now.

[swag72]

For my taste this is a little heavy on the processing - By getting such a light area in front of the trunk you are almost losing some of the contrast as it works its way to the bottom part of the screen. Personally I'd be looking at processing to bring out the trunk more without such a light background, if that makes sense. There's some lovely light areas around the clouds but they are allmost lost in the brightness.

Just a personal preference though if you like it as it is and others do to.

[ianaiken]

Sara - thanks for the feedback. In fact I thought the processing was minimal but maybe I've overdone it with the curves on the contrast. I've attached the .fit if anyone wants a go I'm all for seeing what can be done and to learn. Download FIT Saving to JPG is definately causing

Here's another process using ATWT.

[ianaiken]

Sara,

What do you think of this revision? I have used HDR which has resulted in a bit more detail in the brighter areas and it has also darkened them too as a result. I also did another pass with ATWT.

Edit: Of course when I get more subs I will end up doing less processing and I think the noise will be reduced further.

[swag72]

I prefer that last edit Ian, the Trunk is much more defined against the background.

[fwm891]

Tried a process taking the FIT file into Fits Liberator and saving as a Tiff. Then simple curves in PS and saved as png

[Gina]

Yes, I agree

[ianaiken]

Francis - I like the simple processing you've done and I never thought about saving as png I will give that a try.

Thanks Sara and Gina for the feedback on the last revision I posted. It is helpful.

[ollypenrice]

I think you're taking the data well past its noise limit. I do go past the un-noise-reduced limit myself and then apply NR but I never do so by much. The only proper way to reduce noise is ... yes, more data! T'was ever thus.

What I did here was stetch just into the noise, select the faint stuff, noise reduce it, then add a contrast curve to the brighter parts of the nebula, notably the ionization front. This could then stand just a tad of sharpening around the rim. On this target I don't see much point in stretching the background nebulosity across the board. I prefer to bring out the brighter accents in the Ha. Otherwise the result strikes me as flattened.

The data is perfect. It's lovely. You just need twice as much of it.

This was all in Photoshop. I do use PI but I couldn't see any need to battle with it on this image!

Olly

[ianaiken]

Olly thank you for the valuable input. Without your input over the last year I wouldn't have been where I am today in such a short period of time!

Stretching into the noise is not a concept I have given any thought. I am going to do some stretches and see just what you mean, I can see from your processing noise is minimum and from the processing I did on the image noise was quite high so I think I was stretching too much particularly on the mid levels.

Are you using masks to select fainter or brighter areas of the image and inverting it depending on if its fainter or brighter?

Hoping to get more data on this tonight pending cloud, supposed to be clear all night so it would be nice to get another 3 hours or so on it.

[ollypenrice]

Hi Ian,

My stretch was, perhaps, on the conservative side but that tends to be my way. I didn't use a mask though you could do so to keep the stars down. Once the lower brightnesses, de-noised, had given what they could I pinned their lower part of the curve and lifted the upper-middle part with a straight line to the white point so as to have the least effect on the stars. You've got the tracking and focus bang on, though, so the stars remained tight. The data really is bang on the money. Good stuff.

Olly

[Gina]

Ah yes, more data... that needs more clear night skies... hmmm... I want some