Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

I realize this is a big ask, but...

  • Please log in to reply
19 replies to this topic

#1 JAS62

JAS62

    Sputnik

  • -----
  • topic starter
  • Posts: 44
  • Joined: 07 Jan 2019
  • Loc: Redmond, Oregon

Posted 19 November 2019 - 06:05 PM

Greetings!

 

I am wondering if anyone would be willing to download and try their hand at processing my raw data?  I am very interested in a more experienced opinion.  I am interested in finding out how my post processing techniques could improve by comparing my process with someone else.  Using the same data would make the comparison more meaningful.  If this request is too off the wall then just let me know.

 

The data is from my first ever fully successful DSO imaging session.  By "success" I mean all of the bits necessary to collect a series of stackable images actually worked.  No kidding, it was a monumental event for me.  The target was M33.  The seeing was good.  I live in a bortle class 4 area.  My equipment is pretty low-budget: Apertura 6" F/4 newt; Celestron CG-5 mount; Nikon D5300; SvBony 50mm guide scope; ASI120MC-S guide cam; APT for image acquisition; SharpCap for polar alignment.  I used DSS to stack (I stacked all 24 images) and StarTools for post processing.  I didn't take any dark, flat or bias images.  This is my end result:

 

M33 11/04/2019
 
My post processing skills are very weak.  I pretty much cut my teeth in StarTools on this image.  I am anticipating being able to improve my process, but post processing seems to be an art unto itself.
 
I'm pretty happy with the result being my first successful data collection.  I know I need a coma corrector.  I have one on the way.  I cropped the final image to get rid of the distortion.  I found out later that StarTools has a feature that may be able to correct the distorted stars.  The images are portrait in orientation.  Don't know if that matters.  Next time I think I'll orient the camera differently.
 
The source images are 24 .NEF files straight from the camera.  I noticed the first 18 or so have a bit of a glow to them.  It seems to surround the target.  The glow made it's way through the stacking.  I may have been able to correct the glow had I included darks or flats.  I'm not sure.  I do know I struggled with the glow through processing and couldn't figure out how to git rid of it.  It mainly shows on the left side of the finished image.  I'd be interested in what may have caused the glow.  Seems strange it faded out near the end of the session.
 
If anyone is interested in granting me the time and valued opinion, I'd be most grateful.  The file is about 550MB in size.

 

Here's the link to the images: M33_11042019.zip (Onedrive)

 

Clear skies!


  • zjc26138, starcam, Michael Harris and 6 others like this

#2 dmdouglass

dmdouglass

    Surveyor 1

  • *****
  • Posts: 1797
  • Joined: 23 Dec 2007
  • Loc: Tempe, AZ

Posted 19 November 2019 - 06:20 PM

From what i am seeing here, i would say ...  Pretty darn good, especially for first success !!

Well done.



#3 nimitz69

nimitz69

    Surveyor 1

  • *****
  • Posts: 1647
  • Joined: 21 Apr 2017
  • Loc: A barrier island 18 miles south of Cocoa Beach

Posted 19 November 2019 - 06:24 PM

I’ll let some of the real experts here process your stack but 2 things jump out at me immediately ...

 

1.  You need to get in the habit of taking cal frames EVERY time you image right now.  Darks and bias can be taken at any time and build a library. Only flats need to be taken after a session and that should only take about 5 mins. And. If you’re not dithering - start. For most capture.  S/w its just a check of a box and use the defaults until you know why you should change them.

 

2.  24 subs is not much data.  Total integration time trumps everything.  I don’t start any real processing until I have 3-5 hrs on a target ...

 

nice first image BTW!


Edited by nimitz69, 19 November 2019 - 06:26 PM.


#4 scadvice

scadvice

    Surveyor 1

  • *****
  • Vendors
  • Posts: 1552
  • Joined: 20 Feb 2018
  • Loc: Lodi, California

Posted 19 November 2019 - 07:08 PM

Yes very nice start!

 

If you post a picture here under 500k in jpeg. I can check a few things out for you in PixInsight real quick. Your zip file was giving me fits and wouldn't down under 20 mins. I think too much green? Maybe giving you the brownish tint and covering your red nebula in M33. Also I can pull the image hard quickly and see if more data/color is there. Of course I could be wrong and I could tell with the jpeg image.

 

For fun while testing a newly wired up usb and power setup I did a 60% moon, in cloudy hazy night and got this with my old T3i modified Canon. Your DSLR is nicer so it may be as stated by those before me. Flats, bias, and darks are a must.

 

Here is the image I did on a really bad night and details. I kept the images at 90 seconds each because of the clouds the seeing was horrible that night also.

 

https://www.astrobin...e4nf/B/?nc=user


  • JAS62 likes this

#5 SDTopensied

SDTopensied

    Viking 1

  • *****
  • Posts: 803
  • Joined: 25 Apr 2011
  • Loc: Atlanta

Posted 19 November 2019 - 07:36 PM

Nice image...I'll give it a shot.  PM me with a download location like Google Drive or something similar.  I'll be using PixInsight and I'll be happy to share my workflow with you.

 

-Steve



#6 kathyastro

kathyastro

    Viking 1

  • *****
  • Posts: 997
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 19 November 2019 - 07:48 PM

I took a shot at it.  ****, that's an enormous sensor you've got!  I have a fairly fast computer, and it took 42 minutes just to do the integration portion of the stacking!  Suggestion: post the stacked FITS file, rather than the individual frames.

 

You absolutely MUST take darks, flats and bias frames!!  The image is very noisy and has wicked vignetting.  Darks and flats will go a long way to fixing those issues.  Without the calibration frames, I was forced to make the dark pixels (i.e. the background) much darker than I like to hide the noise and bright corners.  And that is after I had cropped the image a lot.

 

You have coma, but you also have a sensor alignment problem.  The stars on the right side are much worse than those on the left side.

 

Anyway, here's what I got:

Image10.jpg


  • Scott Mitchell, lbim, entilza and 4 others like this

#7 Hondo

Hondo

    Ranger 4

  • -----
  • Posts: 376
  • Joined: 28 Oct 2017
  • Loc: NB, Canada

Posted 19 November 2019 - 08:07 PM

Hi Joe,

 

Here is what I did with Star Tools 1.3 and I have also attached the ST log. Thanks for the share.

 

 

Attached Thumbnails

  • Joes M33 CN_filtered.jpg

Attached Files


  • gundark, kathyastro, scadvice and 3 others like this

#8 Scott Mitchell

Scott Mitchell

    Viking 1

  • -----
  • Posts: 963
  • Joined: 28 Jul 2007
  • Loc: Atlanta-ish, GA

Posted 19 November 2019 - 08:31 PM

You're off to a great start! Get into the habit of using the full set of calibration frames and you'll love the results. 

 

Nice processing job @kathyastro!


  • kathyastro and JAS62 like this

#9 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 17373
  • Joined: 27 Oct 2014

Posted 19 November 2019 - 09:39 PM

It's a _whole_ lot less to ask if you upload the stack, rather than the individual frames.  That's the common approach here.



#10 JAS62

JAS62

    Sputnik

  • -----
  • topic starter
  • Posts: 44
  • Joined: 07 Jan 2019
  • Loc: Redmond, Oregon

Posted 19 November 2019 - 10:34 PM

Wow!  I'm kind of overwhelmed.  Did not expect this kind of response.  Thank you, everyone.

 

 

From what i am seeing here, i would say ...  Pretty darn good, especially for first success !!

Well done.

 

Really, I'm pretty proud of it.  Knowing (intimately) what it took to get the image, I actually feel fortunate.

 

 

I’ll let some of the real experts here process your stack but 2 things jump out at me immediately ...

 

1.  You need to get in the habit of taking cal frames EVERY time you image right now.  Darks and bias can be taken at any time and build a library. Only flats need to be taken after a session and that should only take about 5 mins. And. If you’re not dithering - start. For most capture.  S/w its just a check of a box and use the defaults until you know why you should change them.

 

2.  24 subs is not much data.  Total integration time trumps everything.  I don’t start any real processing until I have 3-5 hrs on a target ...

 

nice first image BTW!

 

Not taking cal frames is a bad habit, I know.  It's not like I didn't know that at the time.  Silly of me.  Thanks for the advice on acquiring darks and bias.   I was unaware that darks didn't have to mimic the imaging environment.  I will certainly get that done.

 

 

Yes very nice start!

 

If you post a picture here under 500k in jpeg. I can check a few things out for you in PixInsight real quick. Your zip file was giving me fits and wouldn't down under 20 mins. I think too much green? Maybe giving you the brownish tint and covering your red nebula in M33. Also I can pull the image hard quickly and see if more data/color is there. Of course I could be wrong and I could tell with the jpeg image.

 

For fun while testing a newly wired up usb and power setup I did a 60% moon, in cloudy hazy night and got this with my old T3i modified Canon. Your DSLR is nicer so it may be as stated by those before me. Flats, bias, and darks are a must.

 

Here is the image I did on a really bad night and details. I kept the images at 90 seconds each because of the clouds the seeing was horrible that night also.

 

https://www.astrobin...e4nf/B/?nc=user

 

Download size was a little unreasonable.  Your M33 shot is really nice.  Certainly for not so good seeing.  My shot had almost perfect conditions.

 

 

Nice image...I'll give it a shot.  PM me with a download location like Google Drive or something similar.  I'll be using PixInsight and I'll be happy to share my workflow with you.

 

-Steve

 

Thanks Steve.  I send a PM.  It's a big file.  Sorry.

 

 

I took a shot at it.  ****, that's an enormous sensor you've got!  I have a fairly fast computer, and it took 42 minutes just to do the integration portion of the stacking!  Suggestion: post the stacked FITS file, rather than the individual frames.

 

You absolutely MUST take darks, flats and bias frames!!  The image is very noisy and has wicked vignetting.  Darks and flats will go a long way to fixing those issues.  Without the calibration frames, I was forced to make the dark pixels (i.e. the background) much darker than I like to hide the noise and bright corners.  And that is after I had cropped the image a lot.

 

You have coma, but you also have a sensor alignment problem.  The stars on the right side are much worse than those on the left side.

 

Anyway, here's what I got:

attachicon.gif Image10.jpg

 

Ha!  So I picked up the Nikon D5300 used from B&H for about $350.  My research showed the D5XXX series camera use the same Sony sensor as used in several higher end Nikons and some dedicated imaging cameras.  It's supposed to have a good quality sensor.  Yes, it's big.  some where around 6000 x 4000 pix.

 

I've been thoroughly chastised for not taking calibration frames.  It will not happen again. frown.gif

 

Hmmm.  Sensor alignment.  I'll have to investigate that.  Could it be collimation?

 

Very nice image.  I like the coloration.  Soft.  Less red bias.  Much cleaner background.  Makes me want to reprocess and experiment with the coloring.

 

 

Hi Joe,

 

Here is what I did with Star Tools 1.3 and I have also attached the ST log. Thanks for the share.

 

Awesome.  Very nice image.  I think the theme for for M33 is to push the blue a bit.  I wonder the reasoning for that.  I really did not deviate a lot from the original color presentation.  Probably pushed the red a bit.

 

Thanks so much for the log.  I have watched a many videos on processing with StarTools, some are little hard to follow.  I will study your log thoroughly and see if I can replicate your process.  I think that will teach me a lot.

 

 

You're off to a great start! Get into the habit of using the full set of calibration frames and you'll love the results. 

 

Nice processing job @kathyastro!

 

Thank you!

 

 

It's a _whole_ lot less to ask if you upload the stack, rather than the individual frames.  That's the common approach here.

 

Heard.  I will be a little more kind to people's bandwidth.

 

 

Thank's everyone!  CN is the bomb!  Thursday night is looking good up here.  Temp going to be in the teens.  Going to put together a plan.  Maybe Pleiades...


  • Scott Mitchell likes this

#11 kathyastro

kathyastro

    Viking 1

  • *****
  • Posts: 997
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 20 November 2019 - 07:19 AM

Thanks for the advice on acquiring darks and bias.   I was unaware that darks didn't have to mimic the imaging environment.

...

Hmmm.  Sensor alignment.  I'll have to investigate that.  Could it be collimation?

...

I think the theme for for M33 is to push the blue a bit.  I wonder the reasoning for that.  I really did not deviate a lot from the original color presentation. 

Darks don't have to mimic the imaging physical setup.  They do have to match the lights in exposure time, ISO, and temperature.  Some people put their camera in the fridge to take darks!

 

The defect I was referring to could not be collimation.  The sensor was not sitting square to the light path.  So one of the physical connections between the scope and the camera was not straight.

 

Colour is one of the parameters that you routinely have to adjust as part of your processing.  What comes straight from the camera is seldom correct.  After seeing enough images of a popular object, you get a feel for how it looks.  In the case of M33, the spiral arms are mostly blue, with a smattering of red knots.  The red won't show up well if your DSLR is unmodified, so I didn't worry about trying to bring it out in my processing, but the blue should definitely predominate. 



#12 JAS62

JAS62

    Sputnik

  • -----
  • topic starter
  • Posts: 44
  • Joined: 07 Jan 2019
  • Loc: Redmond, Oregon

Posted 20 November 2019 - 11:47 AM

Darks don't have to mimic the imaging physical setup.  They do have to match the lights in exposure time, ISO, and temperature.  Some people put their camera in the fridge to take darks!

 

The defect I was referring to could not be collimation.  The sensor was not sitting square to the light path.  So one of the physical connections between the scope and the camera was not straight.

 

Colour is one of the parameters that you routinely have to adjust as part of your processing.  What comes straight from the camera is seldom correct.  After seeing enough images of a popular object, you get a feel for how it looks.  In the case of M33, the spiral arms are mostly blue, with a smattering of red knots.  The red won't show up well if your DSLR is unmodified, so I didn't worry about trying to bring it out in my processing, but the blue should definitely predominate. 

Thank you Kathy.  Great info.  I'll be shooting some darks and flats tonight once everything cools off.

 

Sensor alignment makes total sense to me know.  It's very likely I did not completely seat the nose piece, or more likely the reducer, in the focuser.  This could have left the camera off square.  I would have never have caught that.  I appreciate the explanation.

 

Since I am a total newb, there's much about image processing I still need to get caught up on.  Again, it make total sense to look at a series of existing images to develop an sense of "normal" when it comes to color patterns.  I might buy another DSLR and have it modified.  This one I'd like to keep in tact for normal use.  Probably should look at a dedicated image camera, but I think I'll leave that purchase decision for the next rig build. grin.gif

 

Joe



#13 scadvice

scadvice

    Surveyor 1

  • *****
  • Vendors
  • Posts: 1552
  • Joined: 20 Feb 2018
  • Loc: Lodi, California

Posted 20 November 2019 - 12:36 PM

You may know all this but I thought I'd past it on...

 

One other thing... It's recommended to shoot the darks AFTER your lights session because of temperature change.

 

Bias, my information says temp is not important but I just do them after darks.

 

Flats, I do them last and temp is not important. I just use this light panel on its lowest setting. No white T-shirt needed.

 

https://www.amazon.c...0?ie=UTF8&psc=1


  • JAS62 likes this

#14 Eric Seavey

Eric Seavey

    Viking 1

  • *****
  • Posts: 685
  • Joined: 17 Nov 2013
  • Loc: Irvine, California

Posted 20 November 2019 - 02:14 PM

Nice start! Congrats!

 

Here is my edit.  I used ImagesPlus, Photoshop with AstroFlat Pro, and reduced noise using Noise Control from The Plugin Site.  A good gradient reduction tool is important since you did not take flats.  As many here already have said, taking darks and flats would boost the image quality.

 

 

Attached Thumbnails

  • M33.jpg

  • kathyastro and JAS62 like this

#15 JAS62

JAS62

    Sputnik

  • -----
  • topic starter
  • Posts: 44
  • Joined: 07 Jan 2019
  • Loc: Redmond, Oregon

Posted 20 November 2019 - 05:58 PM

Nice start! Congrats!

 

Here is my edit.  I used ImagesPlus, Photoshop with AstroFlat Pro, and reduced noise using Noise Control from The Plugin Site.  A good gradient reduction tool is important since you did not take flats.  As many here already have said, taking darks and flats would boost the image quality.

Very nice.  The amount of noise reduction is pretty impressive.  Sounds like the right software tools are called for.  I might take another series tomorrow night along with darks and flats to generate a comparison  The results might make a good post.  Thanks Eric!


  • Eric Seavey likes this

#16 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 17373
  • Joined: 27 Oct 2014

Posted 20 November 2019 - 06:08 PM

Very nice.  The amount of noise reduction is pretty impressive.  Sounds like the right software tools are called for.  I might take another series tomorrow night along with darks and flats to generate a comparison  The results might make a good post.  Thanks Eric!

Re "darks and flats".  You do need bias also.  Otherwise the math for flats correction doesn't work properly.

 

Note that, if you omit the camera calibration frames, the minor problem is that your images won't be as good.

 

The major problem is that you're very likely to learn some bad habits in processing.  Processing is tough enough without having to unlearn bad habits.  <smile>


Edited by bobzeq25, 20 November 2019 - 06:11 PM.


#17 JAS62

JAS62

    Sputnik

  • -----
  • topic starter
  • Posts: 44
  • Joined: 07 Jan 2019
  • Loc: Redmond, Oregon

Posted 20 November 2019 - 06:41 PM

Re "darks and flats".  You do need bias also.  Otherwise the math for flats correction doesn't work properly.

 

Note that, if you omit the camera calibration frames, the minor problem is that your images won't be as good.

 

The major problem is that you're very likely to learn some bad habits in processing.  Processing is tough enough without having to unlearn bad habits.  <smile>

 

Yes.  I do realize I need bias frames.  It's just they they are so easy to do it's not work the extra few characters to mention.smirk.gif   I do appreciate the reminder.  This newb needs all the help he can get.

 

Thanks Bob!


  • bobzeq25 likes this

#18 Eric Seavey

Eric Seavey

    Viking 1

  • *****
  • Posts: 685
  • Joined: 17 Nov 2013
  • Loc: Irvine, California

Posted 21 November 2019 - 10:20 AM

Very nice.  The amount of noise reduction is pretty impressive.  Sounds like the right software tools are called for.  I might take another series tomorrow night along with darks and flats to generate a comparison  The results might make a good post.  Thanks Eric!

There are a number of photography software programs out there that are geared to the specific needs in astro photography.  One is called pixinsight and is able to tweak any setting you desire.  Another is astropixelprocessor, that I recommend to anyone starting out in astro photography at it automates many tasks and with little effort, creates great edits.  Deep sky stacker is what many people use, and it stacks pretty well, but stretching performance is OK, and further processing is limited or non-existent.  I personally like ImagesPlus because of the photoshop feel and the wealth of options, though it is a little outdated, which is why I use some plugins in photo shop to deal with gradients and chroma noise.  I know people who swear by CCDStack.  I tried it briefly and don't know much about it anymore.  Maxim DL and prism I never tried.

So there is a lot to choose from and it is worth trying out the software packages once you have a good set of images to process.

 

Good luck.


  • JAS62 likes this

#19 Dynan

Dynan

    Gemini

  • *****
  • Posts: 3215
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 21 November 2019 - 11:29 AM

Hi Joe,

 

Here is what I did with Star Tools 1.3 and I have also attached the ST log. Thanks for the share.

+++ Star Tools  - FTW!



#20 JAS62

JAS62

    Sputnik

  • -----
  • topic starter
  • Posts: 44
  • Joined: 07 Jan 2019
  • Loc: Redmond, Oregon

Posted 21 November 2019 - 01:04 PM

There are a number of photography software programs out there that are geared to the specific needs in astro photography.  One is called pixinsight and is able to tweak any setting you desire.  Another is astropixelprocessor, that I recommend to anyone starting out in astro photography at it automates many tasks and with little effort, creates great edits.  Deep sky stacker is what many people use, and it stacks pretty well, but stretching performance is OK, and further processing is limited or non-existent.  I personally like ImagesPlus because of the photoshop feel and the wealth of options, though it is a little outdated, which is why I use some plugins in photo shop to deal with gradients and chroma noise.  I know people who swear by CCDStack.  I tried it briefly and don't know much about it anymore.  Maxim DL and prism I never tried.

So there is a lot to choose from and it is worth trying out the software packages once you have a good set of images to process.

 

Good luck.

 

I have this book, "The Deep-Sky Imaging Primer" - Charles Bracken.  It's organized in three sections:  Understanding Images,  Acquiring Images, Processing Images.  It is very detailed.  Just reviewing the table of contents is a bit much.  However, the author approaches the most technical topic in a very practical way.  I've only gotten part way through the first section.  (If I can just get this day job out of the way.)  The first section, thus far, has really helped me understand the theory (for lack of a better word) of what I'm trying to achieve.

 

The processing section uses PixInsight to demonstrate the principles of image processing.  From an initial scan and what I've read thus far, it looks like it will be a reference I will return to frequently.  I give it the newb thumbs up. waytogo.gif


  • bobzeq25 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics