Total Noob needs help: Milky Way

[Effix]

Good morning/afternoon/evening to you all, ladies and gentlemen!

As the title says, I am quite new in StarTools, and in astrophotography in general. I have just discovered StarTools last week, and I think it will become my new image processing software, to replace that old version (5+ years) of Photoshop that I am currently using to try and get the best of my pictures, though with very poor results so far.

So, As for now, I have with me a picture of the Milky Way around the Summer Triangle (at least that's how we call it in French: Vega-Altaïr-Deneb). I think the pic is quite challenging, as I made several mistakes, like shooting in JPEG (won't do it again, I promise), and, after stacking in DeepSkyStaker, saving in TIFF (won't do again either, I've been reading the forum a bit already ) It is made of about 10 pictures of 30 seconds each, only with a tripod (the equatorial mount is planned but not yet aquired). The night was very wet, and I had to wipe my objective after each picture to remove the mist.

https://www.dropbox.com/s/1mv7e0xr4xf4h ... r.tif?dl=0

I tried to improve it with Photoshop and got a good enough result for me, but I am pretty sure it could be much much better, as it doesn't compare at all with pictures I can see here and there on the internet, which seems to have been taken in pretty much the same conditions... if at least I could remove this aweful noise...

https://www.dropbox.com/s/twr97e8z2igne ... 8.jpg?dl=0

I tried to develop it with the trial version of StarTools, which means that I unfortunately cannot show it to you (but I'm gonna ask Santa to buy me the full version). However I can tell you that, after the first stretch, it looked very "Modern Art" with a big patch of white in the middle, some red blobs here and there, and stuff made of green and blue... I tried to do some stuff, following the tutorial on the website's main page, but as I am not dealing with the same kind of picture (starfield vs nebula), I couldn't really reproduce the same thing. Also, the software crashed after too much thinking, the computer's getting old... So my humble request is, if you would accept the challenge, that you pros give it a try and tell me if you can get anything good out of it.

I'd be very happy and grateful if you could tell me what I can do to improve both my capture and my post-processing techniques. I want to have shiny pitctures of bright nebulaes through fields of stars too

Thanks a lot already

PS: pictures taken with an Olympus E600 at 3200 ISO f/4.3, 14mm focal length (28mm equivalent for 35mm) with the Hoya red enhancer filter to reduce light pollution.

[admin]

Hi and welcome!

I tried to develop it with the trial version of StarTools, which means that I unfortunately cannot show it to you (but I'm gonna ask Santa to buy me the full version).

There's always making a screenshot (Alt+PrtScr)

However I can tell you that, after the first stretch, it looked very "Modern Art" with a big patch of white in the middle, some red blobs here and there, and stuff made of green and blue...

It's AutoDev's job to show you everything that's 'wrong' with your image until you've fixed it, after which you can use it to perform stretching of the 'real' signal (IF you managed to fix everything, else use the Develop module).

As you've already found out, JPEGs leave very little signal to work and applies all sorts of other processing that pretty much destroys what's left
Add to that that 5mins is not a long time to gather much data with a filter in place on a 13.0x17.3mm sensor.
Try get more frames, shooting RAW, adding bias frames and (more importantly) flat frames.
When you get the chance, share that resulting data with us - it's a much better starting point for getting to grips with your gear and processing. When you start with compromised data, it's hard to give you pointers that will still work once you get 'real' data representative of what is going to be typical for your gear.

There is a hint of milkyway in there, but the signal is very, very low... Your camera certainly is up to the job though!

Cygnus CLear.jpg (215.25 KiB) Viewed 9626 times

You'll also want to reduce your data set's resolution (in exchange for noise reduction) by using the Bin module. It'll let your computer cope with the data much better and will let you 'hide' imperfections such as your tracking problems, focus problems and interpolation artifacts (all DSLR data has the latter at 'full' resolution).

Do you reckon you could get us a non-JPEG based stack?

[Effix]

Hi Ivo, thanks for your reply

First, I'm quite "glad' to see that your result is also quite noisy and kinda weird looking. At least I now what I have to improve the most, that is, the image capture.
I'm afraid I do not have any RAW data to provide you with: this picture was taken on October 19th, and since then the weather has never been good enough to allow me to go out at night. But as soon as the sky is clear I shall try again (Orion will be my next target!).
Meanwhile, would you know where can I get raw files with which I could practice?

[admin]

Glad I could (sort of) help!

Meanwhile, would you know where can I get raw files with which I could practice?

For some data sets, try here;
https://groups.yahoo.com/neo/groups/dsl ... ssing/info

And http://www.mistisoftware.com/astronomy/index_fits.htm

Most StarTools videos on YouTube come have their datasets linked in the description, so you could try your hand at those as well.

[Effix]

Thank you! I'll have a try at these and get back when I have something of my own to show you.

[admin]

Cool - we'll be here!

[Effix]

Hi there I'm back! And with a new set of pictures!

I took advantage of the clear sky last night to try and make more photos of the deep sky, as I currently have my motorised mount with me. So far I have only stacked M45. 25x40s exposures in RAW format (I've learnt), still with my Olympus E600 at ISO 3200, zoom 70-150mm set @ 150 (300mm equivalent), f/5.6. Filters Hoya Red Enhancer for light pollution and Hoya 4-star cross screen to give it a starry look.
I also made all the Darks, Flats, Dark Flats and Bias, with at least 15 pictures of each.
As I still don't own a key for the full version of StarTools, I'll show you only a screenshot of the final result. However I managed to trick some people into offer it to me for Christmas, so within a couple of days I shall be able to give you full-res pictures of my work.

For the time being, I present you only two screenshots of my work, the full size and a close-up of the Pleiades cluster. I am quite happy with my work, only there is always a kind of a violent noise across the whole picture which I cannot remove. if you look closely on the brightest stars, the nebulosity is a bit visible, but is kinda stained by dark pixels. Same for the general background, there is some blue pixels a bit everywhere which I assimilate to noise but that I cannot remove. As said, I shot in RAW, turning off all the noise reduction stuff of the camera, so I do'nt know where this comes from. Any comments and ideas are welcome.

Here's also the link to the original stacked image: https://www.dropbox.com/s/a5grx0o8a9hydja/M45.FTS?dl=0

M45
https://www.dropbox.com/s/4jbgx6qgi78vihg/M45.jpg?dl=0

M45 close-up
https://www.dropbox.com/s/7kayh8xwy5dvs ... p.jpg?dl=0

[admin]

Your data is pretty clean - congrats!
Looks like you're really getting the hang of it now and you're learning quick!

All you really need - by the looks of it - is more and longer exposures (try 120 seconds minimum). As you pointed out, the nebulosity is just about to show, unfortunately it is barely above the noise floor.
Try lowering the ISO to 400 or 800 so you don't overexpose anything when doing the longer exposures (don't worry - you're not losing any detail vs noise above ISO 400 or ISO 800 as digital cameras simply multiply the amount of photons they detect and store the result).

You're definitely making good (and fast) progress.

[Effix]

Thanks Ivo, that's very encouraging! I hope to be able to give it another try before the end of the year, but it's gonna be hard considering the weather here... The same night I also took M42 and M31, but M31 looks strange, it's very very blue, and the arms of hte galaxy are lost in the noise.
Also, I sadly still can't save my work. Con you have a look at the key requests from December 25th? there should be mine somewhere

[admin]

Thanks Ivo, that's very encouraging! I hope to be able to give it another try before the end of the year, but it's gonna be hard considering the weather here... The same night I also took M42 and M31, but M31 looks strange, it's very very blue, and the arms of hte galaxy are lost in the noise.

Great! The virgin stacked data being very blue is usually what happens if the data isn't color calibrated (which is a good thing!), though something else might be at play here. Happy to have a look at the M31 data if you like...

Also, I sadly still can't save my work. Con you have a look at the key requests from December 25th? there should be mine somewhere

Your license file was sent out to you (a GMail account) on the 26th (Australian time, so probably still 25th European time).
Maybe it's in your spam folder? Happy to resend or send it to an alternative e-mail address, etc.! Let me know.