Adreneline

Very nice. You managed to achieve some real depth to the image, and I love the colours. Adrian

I have both Astronomy Tools Actions and Annie's Astro Actions - I find both very useful but I tend to find I use AAA more. AAA is worth a look. Adrian

Very nice Brendan- really like the colours. Adrian

Well maybe not all, but lots just isn't worth keeping. It took me a while to nail focus and guiding and whilst so doing I accumulated a lot of dodgy data. I also suffered a lot with "blue bloat" - not good! I might be a bit selective but I think I need a fresh start - that why we're moving! A new beginning in our retirement, just five minutes walk from the coast.

I might need to look at that one myself! The star reduced version is much better in my view - the stars in the original were a little distracting but now the star of the show (excuse the pun), the nebula, is more apparent. AP is a constant learning and adapting game when it comes to image processing, and what works for one image does not necessarily work for another. Thanks for sharing and good luck. Adrian P.S. Tomorrow we move to darker skies! I'm thinking of dumping all my previous data and starting a fresh. I can't see the point of combining Bortle 5/6 with Bortle 3/4 and it'll save buying a new remote hard drive. It will also be the first time I've been able to image the western sky. Exciting! Might be a few weeks before I'm up and runnning though.

That's a nicely framed image and despite your concerns about noise there is a lot of detail in the nebula. I use a CEM25-EC unguided for all of my imaging, so no dithering because I use the ASIair to control the whole process. I use an ASI1600MM mono and a Canon 70D or M6 MkII and todate I have been lucky I guess and seen no evidence of 'walking noise'. Like you I use 180s exposures with ASI16000 on all filters with darks, dark-flats and flats but all pre-processed in APP and the post-processed in PI. Exposure times with the dslr are much shorter as I live in a Bortle 5/6 region so I don't go beyond 120s. I do go to great lengths to level the mount and perform a PA using a PoleMaster each time I setup. I've tried guiding with the CEM25 a few times but always been disappointed with the result - maybe it's the ASIair derivative of PHD that's giving me the problem. Hope you get it sorted. Adrian

Very nice. The ZWI ASI533MC certainly does the business. Love the colours. Adrian

A random thought occurs to me. Could you run your dslr/osc linear master through ChannelExtraction and remove the stars from the individual R,G and B channels, much as I do with my Ha, OIII and SII masters. I haveno idea if this will work but it might be worth some experimentation.

I think this is the big difference with dslr/osc. I remove the stars from the Ha, OIII and SII originals after a small initial stretch so I've not had a problem with the 'holes' left behind. Tricky. "If it was easy it wouldn't be worth doing" - not sure who said that but sometimes difficult gets boring as well! Good luck! Adrian

Hello Matteo, I am not sure I know what is right but I am happy to tell you what I do, recognising that dealing with dslr/osc is a little different to RGB or narrowband. I agree with (1) but I would not do (1b). I tend to prefer MLT but that is personal choice (influenced by the Light Vortex tutorials). As soon as I have run MLT I then duplicate the image, one (original) for the starless and one (duplicate/clone) for the image that will keep my stars (the Blend). I then do incremental stretches using HT, three maybe four times, and then apply Starnet to remove the stars. Once the stars are removed I continue to do multiple small incremental stretches, each time moving the black point a small way across the bottom foot of the curve. I can do as many as 10 or 12 separate stretches and always finish with a reset of the black point and no change in the mid-point. At this point I have a nicely stretches starless image. I then use ColourSaturation to boost the colours to suit my taste. I don't use Curves Transformation because I would rather take the image at this point into Affinity and play around with the colour settings (Hue, Saturation, Colour Mix etc.) Going back to the duplicate/clone I now have a choice, thanks to you!. I can take the image into Affinity and stretch using a Curve as shown by Olly and then blend with the starless image using a Lighten layer, or I can now stretch in PixInsight and apply the PixelMath formula you found once I have brought my starless image back into PixInsight. At this stage I'm pretty much done with colour and all I do is a final bit of noise reduction and try not to be too heavy handed. I don't use the dark structure script and I don't mess around with the stars (because I will almost certainly make it worse!). I have used ArcsinhStretch and it can be really good for pulling out star colour in an rgb/osc image so you could certainly apply that on the duplicate instead of using HT. With all these stretch processes I find it is very easy to over do it. I nearly always get to the point where I think "just a minute - I've gone too far" at which point I press the Undo button a couple of times. I tend to prefer a more pastel/soft palette but I understand many prefer a bolder palette with striking colours. My personal view is that space is beautiful but it is not garish or "in your face". It's a bit like facial make-up - lots is not necessarily a good thing. My mother came from Yorkshire and she would have said "there's now't natural about that!". My mother had something to say about most things Anyway, thank you for providing me with a distraction from legal matters. Once again I find myself waiting for an email or a phone call that says we have exchanged contracts - it has been like this every working day since 19th October! Good luck and I hope I have been of some help. Adrian

Sorry for the delay in getting back to you. Once I have my master image which I've cropped, applied DBE/ABE and then MLT I duplicate that image (right click on the image and select Duplicate); at this stage the original and the duplicate are linear. I then start to stretch the original and after three (maybe four) stretches I apply Starnet and remove the stars; once they are removed I continue to stretch the image and then do whatever additional processing I think it requires. I have been experimenting with the table formulae you sent me. Lighten works very similar to what I'm used to in Affinity. I have modified the formula slightly to read max(starless,0.75*Blend) but the 0.75 needs experimentation depending just how much you end up stretching the duplicate. My duplicate (becomes Blend in the formula) is a ChannelCombination of my R,G and B components (which may be Ha,OIII,SII in whatever order I fancy). I have found a second application of ABE removes any background cast and I have also found an application of SCNR removes any halos (magenta halos are removed by inverting the image and then applying SCNR). The end result is not too disimilar to my version with Affinity. Apologies again for the delay. We are living a daily nightmare here thanks to the legal profession who are making a complete meal of exchanging contracts on our house sale - they've been trying for over two weeks. Maybe today the gods will look kindly on me. Adrian

Thank you. With the run of clear nights last week I managed to get some OIII and SII as well and managed to produce this: I'm a bit anti deconvolution really. As I understand it the technique was developed to correct out of focus stars on the initial Hubble images. Seems better to me to get the optics in focus in the first place! I stand to be corrected. Good luck - and one other piece of advice - enjoy the hobby and don't get too OCD about it. If my wife likes the images I produce I'm good to go and I can include them in our Apple TV screensaver and enjoy them whilst listening to our favourite music mix. Success! Adrian

What a great table! I'd not seen that one - and I've bought his book! It's funny how you only see what you're looking for when you refer to a book and can quite happily scan past things that could be really useful. Thank you for link - I shall be giving it a go for sure as I am no longer using PS as a matter of principle. Have to say Affinity does nearly everything I need except it is more difficult working at the channel level - layers are good - channels are a challenge. In answer to your question, I take a duplicate after I've cropped, background extracted and reduced the noise using MLT. I have never used convolution/deconvolution because it always seems to introduce more problems that it solves, but that is probably due to operator incompetance. Good luck and I look forward to seeing more of your images. Adrian

The approach is detailed in this thread: Adrian

I know star erosion can work but Starnet can be a game changer and although it can take a while (don't be tempted to set Stride to 128 - 64 is good enough) it is well worth doing and I have found has improved all my images. Following advice from @ollypenriceI have found the best approach is to use layers in PS (I use Affinity but GIMP works just as well) and use a Lighten layer to overlay a stretched (in PS etc.) duplicate image (post background extraction and noise reduction). I've even gone back and re-visited some of my previous images and been very pleased with the end result. My previous attempts always looked like the stars had been 'stuck on' and two-dimensional whereas using Olly's approach the end result looks much more natural and retains more depth. It's a constant learning game and with all these restrictions I've got plenty of time to investigate processing techniques. Good luck. Adrian

Wow! That is an awesome amount of integration time - well done you for sticking at it and producing such a good result. The stars are quite dominant. I have found it can be helpful to make a duplicate of the image before going non-linear. Do a few partial stretches with HT on the original and then whip the stars out using Starnet. Continue to stretch and process the starless image to your satisfaction. Once you're happy with the starless version you can put the stars back from a lesser stretch duplicate using PixelMaths or maybe using layers in something like GIMP, PS, etc. Great image though. The Heart is one of my favourites. Thank you for sharing. Adrian

The exposure is what it is but at least camera raw allows you to correct for the over exposure to some extent. That's why you shoot in raw mode and not jpg! The main thing is the hardware is not at fault. Good luck!

Looks like over exposure when you look at a raw file in Camera Raw - or similar.

I also used MLT in PI on all four channels whilst they were linear. Once combined I took the finished image into Affinity (you could use PS) and manually applied Dfine2. I have an old copy of Dfine2 from when it was free to use - DxO now charge for it in their NIK Collection. Adrian

I think that is much better - my attempt is not significantly different. I would be inclined to reduce the noise a little more but that is just a personal preference. I've taken the liberty of reducing the noise on your image. Adrian

All sorted - not sure what was going on but I've got all four files. I'll have a play and see and see what I can do. Thank you

Everything downloaded. Strange though that Blue is there twice but I seem to have everything.

Nicely framed image Wyvern, and one I've never attempted. To me the image seems to have quite a lot of noise. I had a go at reducing the noise using the jpg posted but then I lost a lot of the detail. Any chance you might post your RGB and Ha fit masters so I (and maybe others) can have a play. Now we're in Tier 2 (and no doubt heading for Tier 3) I've got no plans to go anywhere so I've time to play Adrian

Absolutely fine by me! What a great and encouraging start. I'm afraid I am in no position to offer advice because having taken an initial 'try it out' image of M31 (above) and had a more serious attempt at the Bubble ..... ...... my RC has been packed away in it's box in anticipation of a house move. Focus is tricky. I upgraded mine with a Baader Diamond SteelTrack and printed a bracket to allow me to attach a SW dc focus motor but other than taking one image of Mars which I'm too embarrassed to post that's all I've managed. Good luck with your RC and I look forward to seeing more of your images. Adrian

I agree - wide field imaging can be very rewarding as well as revealing. I really like using my Canon 200mm as well as the Samyang 135mm. I keep having wayward thoughts about buying another 1600 so I can have both going at the same time. One advantage of the CEM25-EC has been that I don't have to guide, at least up to 180s at these sort of focal lengths, so no more warning beeps and suspended sessions thanks to PHD/SGP. I just push on taking images and dump the bad ones the next morning! Thanks for sharing. Adrian