M27 - Dumbbell Nebula - my first attempt with APP

[StuartT]

Here is my first try at using Astropixel Processor just with all the default settings straight 'out of the box'

I'm still very new to astrophotography, but I'm reasonably happy with this effort.

45 x 60sec lights (Canon EOS750D, 400mm lens, ISO1600)
15 darks
19 bias
27 flats 

 

Edited June 16, 2021 by StuartT

[Clarkey]

Pretty good for 45 minutes data. Have a play with the APP sliders - you can nearly always tease out some more.

[david_taurus83]

Looking good to me. Its like how it looks through an eyepiece 

[AstroMuni]

43 minutes ago, StuartT said:

Here is my first try at using Astropixel Processor just with all the default settings straight 'out of the box'

That looks beautiful. Was it taken through your Meade telescope with the 750D?

Edited June 16, 2021 by AstroMuni

[StuartT]

2 hours ago, AstroMuni said:

That looks beautiful. Was it taken through your Meade telescope with the 750D?

Thanks. No, it was shot with a Canon 400mm f/5.6 telephoto lens.

But my next job is to try it with my 8" SCT for a larger image (though with shorter subs, as I am not guiding)

[StuartT]

3 hours ago, Clarkey said:

Pretty good for 45 minutes data. Have a play with the APP sliders - you can nearly always tease out some more.

Thanks. Please feel free to fiddle with the original and see if you can get more out of it (as I wouldn't know which ones to fiddle with!)

 

M27_Dumbbell_Nebula-RGB-session_1.fits

[Padraic M]

Well Stuart, that's a very fine image to start with, both as an 'early work' and for 45 minutes of exposure. I'm not an SLR expert but I had a go in APP and Gimp. My suggested version is below.

My main problem was that there is a strong red cast in your version, and actually quite a bit of vignetting in red. I'm not sure if that's a feature of SLR or OSC cameras (I've only done mono) but I started by loading your FITS file into APP as a Light frame (it objected but did it anyway), and separated the channels into red, green and blue. As expected, there was quite a difference in light pollution response in green and blue, so in the combined rgb image the white balance changed across the image. I used the Remove Light Pollution tool on all three channels and then recombined them as RGB. That unified the background quite a bit. I cropped out quite a bit at the edges. There was very little other colour modification required, but in the Calibrate Star Colours tool I just tweaked the slopes and constants very slightly, and saved the stretched image to 32-bit TIFF.

In Gimp, there was some Levels work to be done, and then I needed to desaturate the Red channel a little for the background (by masking out the nebula). I scaled the image down by 50% too (equivalent to binning 2x2 I think) because you didn't need the resolution, and there was some colour mottling in the background.

Finally then just exported as JPEG and here it is below. I didn't do any noise reduction, and you might like to keep some more colour saturation in the stars but I quite like it. Btw the data was really nice to work with.

 

[StuartT]

43 minutes ago, Padraic M said:

Well Stuart, that's a very fine image to start with, both as an 'early work' and for 45 minutes of exposure. I'm not an SLR expert but I had a go in APP and Gimp. My suggested version is below.

My main problem was that there is a strong red cast in your version, and actually quite a bit of vignetting in red. I'm not sure if that's a feature of SLR or OSC cameras (I've only done mono) but I started by loading your FITS file into APP as a Light frame (it objected but did it anyway), and separated the channels into red, green and blue. As expected, there was quite a difference in light pollution response in green and blue, so in the combined rgb image the white balance changed across the image. I used the Remove Light Pollution tool on all three channels and then recombined them as RGB. That unified the background quite a bit. I cropped out quite a bit at the edges. There was very little other colour modification required, but in the Calibrate Star Colours tool I just tweaked the slopes and constants very slightly, and saved the stretched image to 32-bit TIFF.

In Gimp, there was some Levels work to be done, and then I needed to desaturate the Red channel a little for the background (by masking out the nebula). I scaled the image down by 50% too (equivalent to binning 2x2 I think) because you didn't need the resolution, and there was some colour mottling in the background.

Finally then just exported as JPEG and here it is below. I didn't do any noise reduction, and you might like to keep some more colour saturation in the stars but I quite like it. Btw the data was really nice to work with.

 

wow! Thanks so much for going to the trouble. I wish I understood all the stuff you did, but it sure looks fine! Maybe one day I can learn to do that myself.

When you say the 'data' was nice to work with, I am not sure what you mean. Is a FITS file not just an image? (forgive my ignorance, but I'd never come across this file format before using APP)

[Padraic M]

The APP learning curve isn't that steep - when you've had a chance to experiment with a few more images you will get a better idea of what it can do. The tools on tab 9 are very interesting to play with.

2 hours ago, StuartT said:

Is a FITS file not just an image?

Yes it is - and that's what we mean by the data - but there are images and there are images! Dealing with sensor noise, bad vignetting, light pollution, thin cloud, poor focus, etc. etc. makes an image so much harder to work with. I notice that images captured in the city (Bortle 8 bright skies) take a lot of processing work, whereas images taken at a dark sky location (Bortle 2) just seem to 'pop out' on their own. The image format doesn't matter; fits and tiff are very similar in that they're both high quality loss-less image formats, but fits is an astrophotography-specific image format with lots of useful information stored in the image headers. You can see all of this in APP when you open a fits file.

Just to add a little to the last post - on the Calibrate tab you can split a colour image into three separate images, one for each channel. Here are the blue, green and red channels; you can see in the blue channel that the top left corner is dark, and the image gets brighter as you travel to bottom right. Similar with green, but with a slightly different distribution. The red channel is actually very even from side to side (and it has more detail in the nebula). This gradient from side to side or corner to corner is very standard with light pollution, as the sky is brighter towards the horizon, or close to a light source. It's also very standard that the different colour channels will pick up the light pollution to different amounts, which means that the combined RGB will have an uneven colour balance across the frame. The Remove Light Pollution tool in APP is excellent at evening out this type of gradient, and really easy to use. There may be a better way to do this with a OSC/DSLR image, but as I said, I just do mono so I do it at the channel level and then combine to RGB.

Blue:

Green:

Red:

 

Edited June 16, 2021 by Padraic M

[StuartT]

34 minutes ago, Padraic M said:

The Remove Light Pollution tool in APP is excellent at evening out this type of gradient, and really easy to use.

Thanks! I have been using an Astronomik CLS filter thus far, but would I need that if I am using this feature of APP? In other words, do I need both?

[Padraic M]

Not an expert on this, but I would imagine that you will use both. Starting with the best data (using the filter to remove as much of the LP as possible) means less work to do in APP. It is likely that there will still be some gradients or vignettes left in the image, even if you use a good LP filter and good flat frames.

[alacant]

On 16/06/2021 at 15:30, StuartT said:

still very new

Hi

Loadsa detail and nice stars. Love it:)

Suggestions? Lose the dark frames, dither and stack using a clipping algorithm.

Cheers

Edited June 17, 2021 by alacant

[almcl]

Had a play with your data and agree with what has been said above: it's very nice!

Here's a crop of the central area just to show the detail that is present, well done! You can probably achieve something similar in APP?

[StuartT]

wow! thanks everyone. Your comments are really encouraging!

I wonder if anyone has any thoughts about how long I can go with subs (I am not guiding) with this 400mm lens? I made 60sec my upper limit and have nice round stars, but I'm wondering if I could go a bit longer?

[almcl]

1 hour ago, StuartT said:

I wonder if anyone has any thoughts about how long I can go with subs (I am not guiding) with this 400mm lens? 

Maybe that's something to try while the light nights are with us?  I managed 3 minutes unguided with my 200 mm lens, so you might well be able to get past 60 sec.  Of course the seeing, polar alignment, target altitude &c will all have  a bearing, but certainly worth experimenting to see.

[The Lazy Astronomer]

3 hours ago, StuartT said:

wow! thanks everyone. Your comments are really encouraging!

I wonder if anyone has any thoughts about how long I can go with subs (I am not guiding) with this 400mm lens? I made 60sec my upper limit and have nice round stars, but I'm wondering if I could go a bit longer?

 If this is on your eq6r, you probably could go longer - I managed to squeeze 3 minutes at 250mm (unguided) out of mine, but had to throw away about 1/3 subs due trailing caused periodic error in the mount. I never tried running periodic error correction (PEC) training though, so that's something you could look into if you run into that problem. 

That said, whilst you could go longer, do you need to? Consider watching this video of a talk by Robin Glover (Sharpcap creator) concerning sub exposure lengths: 

 

 

[StuartT]

19 hours ago, The Lazy Astronomer said:

 

That said, whilst you could go longer, do you need to? Consider watching this video of a talk by Robin Glover (Sharpcap creator) concerning sub exposure lengths: 

Ok, fair point. If I hadn't been working the next day I would've just taken a load more subs (which I guess achieves the same effect). Pity I had to pack up really, as I had good polar alignment by the look of it.

[Clarkey]

Having said a while ago about tweaking the APP sliders I have done that and also removed the gradients and calibrated the star colours. Following on from this I did a bit more 'tweaking' in Affinity including some binning a got the second image - albeit cropped. To be honest, there is plenty of detail here that can be teased out. I only spent ten minutes on it.

As has been said above - the data was pretty good to work with. A bit of a gradient, but nothing that was difficult to remove. As with all AP, there are multiple versions of the same image, most of which is down to personal choice.

 

Edited June 21, 2021 by Clarkey

[rickwayne]

Second  the suggestion for  APP's Remove Light Pollution tool. It's one of the suite's best features.  My recent post shows what kind of nightmares that tool can solve.

Have you played around with the different stretch settings in the right-hand panel? That's often a good start.

I too was put off by the "data" jargon when I first started astro, but it actually serves as a good reminder of just  how much work and intervention has to go between photons generating electrons in a sensor and a picture on the wall. Processing is at least half of the art.