My first M27

[mark7331]

Well here's my interim image from an ongoing attempt at capturing M27. So far I've collected 3hours 10min of data. I'm going to try to collect at least four times that, clouds permitting, over the next few weeks.

Details:-

Scope: SPX200 @ F4.5

Camera: Atik14L+, cooled to -5c

Subs: Luminance (clear IR) 8 x 5m, RGB each 6 x 5m, Ha 6 x 10m

Combined with Astroart and photoshop.

All comments & advice very welcome.

Thanks for looking.

Mark

[MarkF]

Nice one Mark, I am working on this one too with my Atik 16ic and Equinox 80. So far 25 out of focus poor visibility Lums that I have scrapped, then last night managed five 20 min HA subs and 21 5 min Lum subs in focus this time hope to get some RGB tonight moon permitting.

[johnh]

Thats a good start but noticed reds need tucking in with the stars and you have a magenta green diagonal gradient. You should flatten each of the RGB before stacking to eliminate this. How are you integrating the Ha into the image?

John.

[mark7331]

Thanks Mark, good luck for tonight - I'll be out there too.

John, thanks for the comments. I use AstroArt to stack the individual subs and then read the RG&B into photoshop as layers. I'm not entirely sure how/where to flatten the RGB? If you have any advice I would be very grateful. I've done the same for the L & Ha subs.

Edit: I set each layer to a different color in Advanced Blending. I've set the Ha layer to red.

The subs are all perfectly guided but across filter changes I get some rotation/movement which I realign, once the subs are combined, in photoshop. This happens because I'm using a manual filter wheel and I guess I'm being lazy here because I select the next filter while it's guiding. It causes the autoguider to complain for a moment but then picks up. I've done this in the past because of "newbie nerves - will I be able to get the guider working again?" I feel like I know what I'm doing with it now so I'm going to stop the guider and reframe with each filter change. Perhaps also refocus.

Mark

[mark7331]

Reprocessed the original data, after reading many posts and howto's. Also decided to read the manual of AstroArt. Now i know how to rotate and align. This took about 6 attempts at reprocessing.

I'm definitely going to add more data now I have a taste of what processing means. Not there yet but I feel like it's progress.

Thanks for the comments all.

[MarkF]

That looks fab, lots of whispyness, did you try the unsharpen mask with Gaussian blur method on that link you sent me?

Couldn't resist having a go hope you don't mind, then I did a highpass filter and softlight blend, maybe a bit too sharp, but the detail from that 314 is so much better than my 16ic, more pixels to play with I guess.

[mark7331]

Cheers Mark. I went right back to basics and learned how to stack, align and rotate the subs properly. I've been trying to rotate the RG&B by hand but it doesn't work for me. So after reading the AstroArt manual I now know how to get the software to do it. Amazing the difference it made when finally combining the RG&B.

I then played with the curves for the combined RGB and then the Ha. Once I'd pulled out some detail I used the Ha as a red layer and tweaked the opacity so that it didn't overwhelm the RGB.

I did about 12 iterations of curves (on both RGB and Ha) making small adjustments each time - reducing the levels of different nebulosity.

I've just run an unsharp mask on the last result and it really picks out the stars! So much to learn! I keep forgetting what to do . I'm writing up what I've done today and retracing my steps so I can follow it again.

Thanks for the comments and help Mark.

[MarkF]

hehe, and only 6 goes at stacking, I think I can treble that and still not get what I want, looking good though, I edited my last post with a trial sharpening. Hope you don't mind.

[mark7331]

Wow cheers Mark - that's really has picked up the detail! Fantastic!

[ollypenrice]

Thats a good start but noticed reds need tucking in with the stars and you have a magenta green diagonal gradient. You should flatten each of the RGB before stacking to eliminate this. How are you integrating the Ha into the image?

John.

John, could you expand on this point a little? I'm not sure what you mean, do you mean stack all the Rs, all the Gs and all the Bs before combining? I do this routinely but have sometimes noticed this diagonal colour gradient in images from my OSC camera and have never got to the bottom of it. Any clues most welcome.

This image is going well. Be very, very careful not to black clip it because the faint outer halo is easy to amputate by mistake!

When using unsharp mask I would be careful to select only the bright parts of the image for this treatment and also eliminate the stars. One of the reworks has all the hallmarks of USM on areas of weak signal and this throws up star haloes and noise. You can find out how to create a star layer in Martin B's excellent tutorial and use colour select to avoid sharpening anything faint and not inherently sharp anyway, like the background sky.

Olly

[mark7331]

Thanks Olly, I've noticed in some of my previous processing runs how easy it is to clip. Thanks for the USM tips. Will give it a go.

Cheers

Mark

[adamsp123]

Nice image Mark, processing is another big area to get to grips with.

[mark7331]

Thanks Pete. Processing the data has taken about 4 times longer than it did to capture it! And I thought capture was the hard part

[James]

... I'm learning the hard way that processing seems to be the tougher part... at least I have no problem setting up and staying up all night capturing data but scratch my head with the processing bit.

The image is looking good though and can only get better with more data

James

[mark7331]

Today has been a real eye opener for me. I've learnt a lot thanks to the guys here. I have a feeling there's an awful lot more to come.

Thanks for the comment James, I'm looking forward to adding more data and getting more input to the whole process.

Cheers

Mark

[johnh]

Well done Mark its really nice to see you coming along so well with M27, stars seem well aligned now and the colours are better now with the blues and reds contrasting beautifully. I would advise more colour data for each channel and more Ha will give you some outer halo if you desire.

You have also reduced the gradient but usually I flatten the gradient present in each stacked mono channel in Maxim (under filters) before combining the RGB in Maxim and this gets rid of the gradient.

OSC ccd's I have no experience with but Action's has a useful col gradient removal process or could try splitting colour into mono channels and flatten in Maxim then recombine.

John.

[mark7331]

Thanks very much for the comments John. I'm hoping to double the data I've collected tonight. It'll be really interesting to see how the extra data changes the image.

I'll have a look at the gradient filter (I have AstroArt) and give it a try before combining.

Thanks again.

Mark

[mark7331]

Reproces update: After reading Olly's adventures with PixInsight I decided to download it and have a go and now that my brain has stopped hurting, after crashing into the learning curve, I've managed to put together the original data using it.

I had a really difficult time aligning the RGB originally. The red was 180 flipped and it came down to manually nudging but in PI it was a breeze. I didn't even rotate the red to the same orientation as the others!

It's still not there yet - more data and process skill required but I'm going to buy a copy of PixInsight!

[MarkF]

Looking great Mark, I too have been following the pixinsight threads and will try the demo when I have enough data to play around with in the 45 day trial period.

It certainly made a difference to the colours

[mark7331]

Cheers Mark. It's a little difficult to get your head around to start with but they have some video tutorials on another site that have been pretty good:-

http://www.harrysastroshed.com/pixinsighthomenewbie.html

Have a go as soon as you have the data. I'd like to see what you can do with your M57 and if you can get more detail out of the Ha M27 images.

I was going to replace my manual filter wheel with a motorised one but I PI has taken away the issue of rotation/alignment. And it's half the price of the wheel!

Mark

[Joe gray]

Awsome

[mark7331]

Thanks Joe

[gary1968]

NIce image, I take my hat of to you guys that use RGB imaging, soooo much work and time involved, thanks for sharing.

Gary

[msinclairinork]

Woh!! nice one mate! you must be happy with that!!!

Well done!

Michael